Research ArticleParameter Optimisation of Power Regeneration on theHydraulic Electric Regenerative Shock Absorber System
Peng Zheng 1 Ruichen Wang 2 Jingwei Gao1 and Xiang Zhang3
1College of Aerospace Science and Engineering National University of Defense Technology Changsha 410000 China2Institute of Railway Research University of Huddersfield Huddersfield UK3College of Intelligence Science and Technology National University of Defense Technology Changsha 410000 China
Correspondence should be addressed to Ruichen Wang rwanghudacuk
Received 15 December 2018 Revised 28 March 2019 Accepted 15 April 2019 Published 11 June 2019
Academic Editor Angelo Marcelo Tusset
Copyright copy 2019 Peng Zheng et al +is is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited
With the increasingly prominent energy issues regenerative shock absorber has attracted intensive attention in last two decadesfor the development of structure design However the researchers sometimes concentrate on conceptual designs withoutconsidering optimal parameter refinements +is paper proposes a regenerative shock absorber called the ldquohydraulic electricregenerative shock absorber (HERSA)rdquo which includes an analytical regeneration performance parameters optimisation approachto promote the regeneration efficiency and regenerated power +e developed HERSA model is able to convert oscillatory motioninto unidirectional rotary motion through the alteration of hydraulic flow while recovering power by a generator +e proposedmodel is also capable of obtaining the optimal parameters at certain condition as well as providing the flexibility of differentcomponent combinations to match specific system need +e results demonstrate that the proposed model can effectively decidethe optimal parameters in the system and also the recoverable power can achieve average power of 331W at 1Hz-25mmsinusoidal excitation in the system which is approximately 65 efficiency +is study can be further used to guide prototypedesign in future study
1 Introduction
With the rapid development of the global economy vehicleshave been popularised by millions of households Howeverthe explosion of vehicles has also led to huge energy con-sumption and serious environmental pollution For com-mercial vehicles only 10 to 16 of the fuel energy has beenutilised to provide proportion on driving Most of the energyis wasted on road roughness and thermal exhaust Amongthem the kinetic energy loss by shock absorberdamper isthe main energy dissipation [1] In the past two decadesmany researchers have done plenty studies on regenerativeshock absorber to achieve the goal of energy recovery andthe reduction of dissipation
According to the discovery of the enormous potentialfor vibration energy recovery in automotive vehicles [2ndash5]researchers have been more interested in the study ofvehicular regenerative shock absorbers In 1996 Okada
et al [6 7] proposed a regenerative suspension system It ismainly used for active dampers to reduce energy dissipa-tion +e shock absorber uses an electric actuator that cangenerate power at high driving speeds In 2003 Nakanoet al [8 9] proposed an electromagnetic damper which iscomposed of a DC motor a planetary gear and a ball screwmechanism +e DC motor can rotate in both directions tosupply power and recover energy +e results show that thedamper system can recover a certain amount of energyunder the low-frequency and large-amplitude excitationconditions However under the conditions of high-frequency excitation the suspension system has poordamping effect and less energy can be recovered In 2005Bose Corporation [10] developed an electromagnetic activesuspension system +e system is equipped with a linearelectromagnetic motor for each wheel It can not onlyachieve the reduction of vibration but also recover a smallamount of energy for reuse It has a simple structure with a
HindawiShock and VibrationVolume 2019 Article ID 5727849 13 pageshttpsdoiorg10115520195727849
low cost but may be damaged for the small magnetic polegap In 2007 Zhang et al [11] combined the ball screwsystem and DC motor to model a kind of power re-generation system Under the excitation of sine wave inputthis system can output the maximum voltage of 175 V In2012 Li et al [12] designed a built-in motion rectifierdevice for automotive suspension system It converts theuncertain movement into the steady rotation of generator+e simulation shows that the regeneration efficiency is upto 60 and the power output is nearly 15W when thevehicle travelled at a speed of 24 kmh In 2013 Fang et al[13] proposed a kind of hydraulic suspension system whichcan achieve the active control by using a pump to adjust thepressure in hydraulic cylinder In this system the checkvalves are used to control the direction of oil flow to ensurethe steady power output In 2014 Fang et al [14ndash16] de-veloped a hydraulic electromagnetic shock absorber whichcan regulate the fluid flow by applying a built-in hydraulicrectifier and internal accumulator in which the re-generation efficiency of the developed model is 166 at10Hz3mm harmonic excitation Li et al [17] designed andfabricated a hydraulic shock absorber prototype with ahydraulic rectifier to characterise and identify the severalkey mechanical and electrical parameters of an electro-mechanical model According to lack of model accuracythe parameter assumptions in the electromechanical modelcannot be always identified
In general the hydraulic electric regeneration shockabsorber (HERSA) has attracted wide attention due to itsstable work and compact structure [18] which is mainlyintegrated of hydraulic cylinder check valves (rectifiermodules) accumulators (stabilization modules) hydraulicmotors (power conversion modules) and generators (powerregeneration modules) [13ndash17 19ndash22] Some correspondingresearch studies have also been conducted based on theHERSA system In 2015 Xu and Guo [19] applied the geneticalgorithm in the power regenerating system to optimise theparameters of hydraulic pump +e results reveal that it canrecover power of 334W the excitation of sinusoidal wave is167Hz-50mm and the regeneration efficiency is around70 In 2016 Wang et al [20] designed a novel regenerativehydraulic shock absorber system A model which takes intoaccount the impacts of the dynamics of hydraulic flowrotational motion and power regeneration is developed Itcan be found that this system achieves recoverable power of260W with an efficiency of around 40 under sinusoidalexcitation of 1Hz and 25mm amplitude when the accu-mulator capacity is set 032 L with the load resistance 20ΩIn order to obtain a better damper performance in thehydraulic regenerative system Ahmad and Alam [21]analysed the influences of applied components used in thesystem such as hydraulic cylinder hydraulic motor andhydraulic accumulator In 2017 Zhang et al [22] proposed asemiactive regenerative suspension system+is model takesinto account the hydraulic losses in hydraulic system such asthrottle resistance of valves frictional resistance of pipelinesand resistance of motor
In previous studies it can be found that most researchpoints on the HERSA mainly focus on the improvement of
the modelling and experimental techniques but lack detailedresearch on parameter optimisation or optimal componentcombinations A HERSA system is therefore proposed andstudied in the following sections +is paper mainly focuseson the optimisation of power regeneration performanceparameters and studies key parameters such as the hydrauliccylinder size the check valve size the accumulator capacitythe motor displacement and the electrical load Based on theprinciple of orthogonal test the aim of the proposed HERSAis to find the optimal system parameters through modellinganalysis to provide a reliable basis for the prototype designon variable applications which can be universally used forgeneral HERSA system
As shown in Figure 1 this paper firstly introduces thevibration recovery potential and the modelling investigationof regeneration system Secondly the conceptual design ofHERSA system is introduced and then the model is createdfor optimising the parameters of key elements Finally theresults of optimal component combinations are pointed outand then the dynamic behaviours of the HERSA are dis-cussed and validated Based on the orthogonal test method apower regeneration optimisation method for generalHERSA system is also proposed in this paper
2 System Modelling of HERSA
21 Design Concept To understand the mechanism of aregenerative shock absorber system a mathematical model isdesigned on the basis of a typical configuration of the shockabsorber
As shown in Figure 2 a schematic design of the HERSAis proposed which consists of a hydraulic cylinder fourcheck valves a hydraulic accumulator a hydraulic motor apermanent magnetic generator pipelines battery and an oiltank +e key component of the system is the hydrauliccylinder that represents a conventional shock absorber ordamper
When the piston of hydraulic cylinder moves up anddown due to the excitation the shock oil in cylinder can beforced to flow through one-way valve and hydraulic accu-mulator and drive the hydraulic motor By the regulatingeffect of check valves the fluid flow will pass through themotor in one direction Meanwhile the accumulatorrsquossmoothing effect delivers a stable fluid flow to providehydraulic motor a more reliable and stable working con-dition Driven by the pressurised flow the hydraulic motorconverts linear motion into rotary motion as a motionconverter which will also drive the DC generator for thepurpose of power regeneration Finally the power regen-erated will be stored in battery or charged vehicle-mountedequipment
22 Hydraulic Flows +e road excitation is normally con-sisted by numerous sinusoidal waves [23] In order tosimplify the model the excitation is considered as the si-nusoidal wave which is defined as the fundamental elementof road irregularity +erefore the movement of piston isexpressed by
2 Shock and Vibration
Xa(t) X sin(2πft) (1)
where f is the excitation frequency and X is the maximumamplitude When the shaft moves up and down the volumechange of cap-end chamberVA and rod-end chamberVB canbe calculated by
VA AA X0 ∓Xa( 1113857 + Vcyd
VB AB X0 plusmn Xa( 1113857 + Vcyd
⎧⎨
⎩ up and down (2)
Vcyd is dead volume which refers to the fluid volume in thecylinder chambers at zero position It is considered as the
necessary part of volume change In addition AA and AB arethe areas of cap-end and rod-end on both sides of piston X0refers to the starting level of the piston
+e processes of fluid flows in designed hydraulic systemare shown in Figure 3 According to Bernoullirsquos principle
P +ρv2f
2+ ρgh constant (3)
whereP is the pressure ρ is the fluid density vf is the flow speedand h is the elevation of the point above a reference plane +eflow in cylinder and check valves can be expressed by
(a) Outflow of cylinder
QAout CCAC
2 PA minusPM1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PA gtPM
QBout CCAC
2 PB minusPM1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PB gtPM
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎩
(4)
(b) Inflow of cylinder
QAin CCAC
2 PM minusPA1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PM gtPA
QBin CCAC
2 PM minusPB1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PM gtPB
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎩
(5)
CC and AC are the flow coefficient and the area of inlinecheck valve port PA PB and PM represent the pressures
Introduction Researches of power regeneration system Aims of this paper
Accumulator flowNumerical modelling
Orthogonal simulation
Optimisation results andexperimental validation 50mm28mm 635mm
Hydraulic flow
Parameter optimisation of power regeneration on the hydraulic electric regenerative shock absorber system
Cylinder size
Check valve diameter
Accumulator capacity
Motor displacement
Electrical load
Test
desig
n
Key parameters
Sha speed
Hydraulic motor pressure
Regenerated power
Regeneration efficiency
Key indicators
Mea
n re
spon
se an
alys
is
Power regeneration
577cc063L 20Ω
Figure 1 +e study procedures of the HERSA
1
2
A3
4
B
56
7
8
9
10
(1) Accumulator(reserve cavity)
(2) Check valve(cap-end out)
(3) Cap-end chamber(4) Piston head(5) Check valve
(rod-end out)(6) Piston rod(7) Electric motor
(generator)(8) Hydraulic motor(9) Check valve
(cap-end in)(10) Check valve
(rod-end in)
(A) High-pressure side(B) Low-pressure side
Figure 2 Diagram view of the design concept for HERSA
Shock and Vibration 3
at the cap-end chamber the rod-end chamber and themotor inlet respectively
According to the influence of fluid compressibility inhydraulic elements the pressure out of the cylinder duringthe piston motion can be simultaneously expressed
(a) Up stroke
_PA βA AAv(t)minusQAout + QAin( 1113857
VA (6)
(b) Down stroke
_PB βB AB(minusv(t))minusQBout + QBin( 1113857
VB (7)
βA and βB are the effective bulk modulus in cap-endchamber and rod-end chamber
In general there are many empirical formulas for theeffective bulk modulus Given the relatively low pressure inthe HERSA system (under 100 bar) Boesrsquos model [24] isapplied to determine the bulk modulus of the fluid in thecylinder and motor
β βref middotlog 99PPref + 1( 1113857
2 (8)
According to the variations of the cylinder chambers(cap-end and rod-end chambers) the fluid flow rate andpressure in those two chambers can also oscillated signifi-cantly An accumulator is therefore used at the inlet ofmotor and the flow volume Vacf in accumulator can bewritten as
Vacf Vac 1minusPac
Pf1113888 1113889
1k
Pf gtPac (9)
where Vac is the volume of accumulator Pac is prechargedpressure to accumulator Pf is the fluid pressure of the ac-cumulator and k is the gas specific heat ratio of gas-chargedaccumulator and the fluid flow of accumulator can bewritten as
Qac CacAacsgn PM minusPf( 1113857
2 PM minusPf
11138681113868111386811138681113868
11138681113868111386811138681113868
ρ
11139741113972
(10)
where Cac is the accumulator flow coefficient and Aac is thearea of the accumulator inlet port +e volume variation ofaccumulator fluid Vp is
Vp minuskPfQac
_Pf (11)
Considering the smoothing effect of hydraulic accu-mulator a more accurate model of hydraulic motor can bedetermined including the variations of accumulator fluidflow and volume +erefore the pressure and fluid flow ofhydraulic motor can be represented
_PM βM QAout + QBout minusQac minusQM( 1113857
VT
QM DMωM
2π
(12)
where βM is the effective bulk modulus of the motorchamber DM is the displacement of the hydraulic motorand ωM is the shaft speed of the hydraulic motor andgenerator respectively
23 Power Regeneration Driven by the fluid flow the hy-draulic motor can produce rotary motion which is able todrive the DC generator to produce electricity +e workingprocess of power regeneration is shown in Figure 4
Where the mechanical efficiency of motor is ηM thedriving torque of motor TM is
TM DMPMηM
2π (13)
According to Newtonrsquos second law of motion the rotarymotion ωm is
_ωM TM minusTG( 1113857
Jt (14)
where Jt is the shaft moment of inertia In addition theelectromagnetic torque of generator TG would change withthe variation in the induced current and it is considered as theresistance torque because it provides the rotation to themotorin an opposite direction +erefore it can be written as
TG KTI (15)
+e electromotive force (EMF) can be expressed as
E KEωM (16)
UpDown
QBout
QBin
QAout
QAin
VA
VP
AP
PA PMωM
QMAA
PB
ACPf Qac
Aac
PacAB
VB
v
l
M
ρ
Figure 3 Schematic view of fluid flows in hydraulic system
4 Shock and Vibration
where KT is torque constant coefficient and KE is the elec-tromotive voltage constant coefficient
According to Kirchhoffrsquos voltage law [25] assuming thatthe susceptibility at any temperature and the flux that isestablished by the PM poles are constant the variation incurrent can be calculated by
_I Eminus Rin + RB( 1113857I( 1113857
LG (17)
where LG is the internal inductance RB is the externalelectrical load of battery and Rin is the internal resistance ofgenerator +en the instantaneous voltage can be expressedas
U IRB (18)
+erefore the output of regenerated power is
Pout I2RB
U2
RB (19)
In addition the effective input power of this system isconsidered as the sum of the piston damping force multi-plied by the effective piston area Given the areas of cap-endand rod-end are known the piston damping force can bewritten as
FA PAAA
FB PBAB1113896 (20)
+erefore the input power of the HERSA system is
Pin PAAA|v(t)| + PBAB|v(t)| up + down (21)
Hence the power regeneration efficiency can be calcu-lated from the following equation
ηreg Pout
Pin (22)
3 Simulation of HERSA
+is study focuses on the effect of the key parameters whichinclude the sizes of shock absorber body the size of checkvalve port hydraulic motor displacement and hydraulicaccumulator capacity Hence the investigation is performedon the proposed parameter optimisation method to studythe HERSArsquos behaviour and power level as well as thedesirable parameter solutions
31 Parameters Setting and Study In the modelling a fewassumptions made during this process were as follows
(a) +ere are no additional electrical losses in generatorconfiguration to be considered It means the outputpower of hydraulic motor is equal to the input powerof generator
(b) +e cylinder internal leakage between the chambersis not accounted for by the model
(c) +e external electrical load (batteryresistor) has noinfluence by the varying temperature
(d) Fluid compressibility is defined as using Boesrsquosmodel in hydraulic system
In practical applications there are various types of lossesin HERSA system such as hydraulic motor internal flowleakage pressure loss in the pipeline check valve pressureloss and hydraulic cylinder piston friction
To simplify the model the impact of these losses andinfluences in the system are not taken into account at thisstage +erefore the model is ideally configured and thefollowing main parameters are selected
311 Size of the Hydraulic Cylinder In this system thehydraulic cylinder is applied to replace the traditionalshock absorber and absorb vibration energy +is papertakes 4x4 SUV ldquoBeijing Jeep 2021rdquo as the example anddetermines that the maximum stroke of cylinder is 200mm+e sizes of the shock absorber body are determinedaccording to the standard ISO 3320-2013 [26] it is wellknown that a conventional viscous shock absorber hasasymmetrical damping characteristic due to its inherentdesign structure which can provide different dampingforces during the compression and extension strokes +episton diameter and rod diameter are therefore determinedas Table 1
312 Size of the Check Valve Port +e fluid pressurised bythe oscillation flows through check valve arrangement toensure fluid always flows through hydraulic motor in onedirection and enable the chambers in the cylinder can bereplenished as fast as possible for each run In terms of thestandard port size on double-acting cylinder the commonsizes of check valves are shown in Table 2
313 Hydraulic Accumulator Capacity A common hy-draulic accumulator is used to minimise the fluctuation ofthe pressurised flow Initially the gas chamber is prechargedto pressure Pac and set to 20 bar
In equations (9)ndash(11) several assumptions have beenmade to simplify the hydraulic accumulator model
(a) +e accumulator is set as a diaphragm type accu-mulator without heat exchange during the process
(b) +ere is a transient pressure balance inside the ac-cumulator between fluid chamber and gas chamber
(c) Frictions and thermal losses are neglected here
M G
DMTG
RB
KT
KE
LG
Rin
TM
JtPM
QM
ωM
Figure 4 Schematic view of energy conversion (mechanical toelectrical) and power circuit
Shock and Vibration 5
(d) Only fully charged and fully discharged states areconsidered in the hydraulic model
(e) +e precharge pressure in the accumulator is set at60 of the working pressure (20 bar) to providepressure pulsation damping
According to the damping forces in a traditional shockabsorber in a SUV the accumulator capacity has been es-timated with a peak of 100 bar which is shown in Table 3
314 Hydraulic Motor Displacement +e hydraulic motoris defined as a transfer device which is able to convert theunidirectional hydraulic flowpressure into rotationalmotiontorque Its main parameters are shown in Table 4
315 External Electrical Load For the regenerative shockabsorber the electrical load has significant effects on thecapability of power regeneration and the dynamics of thesystem [1] and it can be considered as Table 5
316 Other Key Parameters As shown in Table 6 the valuesof several other parameters are displayed
32 Orthogonal Simulation Test Design +e orthogonal testdesign is an important approach of statistical evaluationwhich can reduce the number of attempts and then effec-tively obtain desired results [27] Large-scale engineeringtesting is a complex system engineering +ere are manyfactors influencing the test results some factors play anindependent role and some factors can interact with othersto produce the comprehensive effect +e reason that affectsthe results during the test is called the test factor and thestate adopted under each factor is called the level Startingfrom Section 31 five-factor and four-level conditions havebeen applied in this test as shown in Table 7 +e simplestmethod is to apply an exhaustive method to test all factorsand levels at the same time
As shown in Table 7 the designed test has four 4-levelfactor and one 2-level factor If all factors and their levelsachieve their best performance for integration testing thenthe total of 512 tests need to be performed Obviously a largenumber of tests definitely demand a lot of manpower andmaterial resources as well as take a long time to reach thetarget+erefore this paper focuses on the problem that howto reasonably arrange the test and obtain the necessaryindication through less number of tests +e orthogonalmethod is an effective mathematical method to solve theproblem of multifactor test It also applies the orthogonaltable design scheme and the mathematical statistics methodto analyse the test data However the L16 (44 times 2) orthogonaltable (Table 8) is designed for the simulation test of HERSA
[28] and only 16 tests need to be carried out It is obviousthat the proposed parameter optimisation method can ef-fectively improve the efficiency of simulation and fastlyobtain the required simulation results
According to the above test design table numericalsimulation is carried out in mathematical model A sinu-soidal excitation 1Hz-25mm is set as predefined input Allother parameters are kept the same as in Table 6
Table 5 BatteryResistor resistance
Factor E Level 1 Level 2 Level 3 Level 4Electrical load (Ω) 10 20 30 40
Table 1 Hydraulic cylinder specification
Factor A Level 1 Level 2 Level 3 Level 4Piston diameter (mm) 50 40 32 25Rod diameter (mm) 28 25 20 16
Table 2 Check valves specification
Factor B Level 1 Level 2 Level 3 Level 4Check valve diameter (mm) 635 9525 127 1905
Table 3 Hydraulic accumulator specification
Factor C Level 1 Level 2Accumulator capacity (L) 063 1
Table 4 Hydraulic motor displacement
Factor D Level 1 Level 2 Level 3 Level 4Motor displacement (cc) 577 707 801 894
Table 6 Other key specifications of the HERSA
Symbol Value Unitf 1 HzXa 25 mmρ 872 kgm3
CC 07 mdashX0 200 mml 1 mβref 12times109 mdashPref 20 bark 14 mdashηm 100 Jt 0003 kgmiddotm2
KT 065 mdashKE 065 mdashRin 75 ΩLG 003 HVagd 01 V LDac Dcv mmCac 07 mdash
Table 7 Simulation test combinations
LevelFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Level 1 5028 635 063 577 10Level 2 4025 9525 1 707 20Level 3 3220 127 mdash 801 30Level 4 2516 1905 mdash 894 40
6 Shock and Vibration
33 Simulation Results and Discussion To use the modelwith orthogonal test approach developed in Sections 31and 32 for subsequent studies on the improvement ofparameter optimisation this section firstly presents studieson the investigation of key model parameters based onmodelling analysis +en it shows the quantitative be-haviours (pressure shaft speed regenerated power andregeneration efficiency) of the system under designed testtable gaining a preliminary understanding of the systembehaviours However the key results are summarised inTable 9 According to the principle of orthogonal test theaverages of outputs are calculated in the mean responseanalysis
As shown in Figure 5 the variation of the motor shaftspeed pressure mechanical power and regeneratedpower is dramatically levelled up at larger cylinder sizewith lower motor displacement accumulator capacity thesize of check valve and external load resistance +eaverages of those results are shown in Table 9 It is alsoclear that the change of component combination cansignificantly affect the system dynamics and regeneratedpower level to meet the demands of various vehiclesuspension systems By changing the component com-binations the maximum regenerated power of 4175Wwith the regeneration efficiency of approximately 534can be achieved on test 1 It is also obvious that the highestregeneration efficiency occurs at test 16 (approximately63) which is designed with small size of cylinder motordisplacement accumulator capacity and large check valveport Additionally the results of those 16 tests reveal thatthe change of motor displacement external loads andaccumulator capacity is capable of smoothing the flowoscillations and allow effective minimisation to the in-stability of hydraulic circuit thus altering the perfor-mance of the system and power capability
It can be summarised that the change of componentcombinations can not only affect the stability of systemdynamics but also dramatically impact on the level ofregenerated power
4 Parameter Optimisation Analysis
16 sets of simulation tests have been designed and performedto provide desirable solutions of the system parameters toenhance the system behaviours and power level After thatmean response analysis is used to take the average value of alltest results for each factor level and determine the extent thatthe factor affects the indicator based on the range meanvalues which is called the comprehensive equilibriummethod It can be applied to perform the multi-indicatoranalysis of shaft speed hydraulic motor pressure re-generative power and regeneration efficiency
41 Shaft Speed As shown in Figure 6 the size of cylinderhas the greatest influence on shaft speed with a visibly largerange+e larger cylinder size can deliver faster flowwhich iscapable of providing larger rotational torque on motor shaft+e smaller motor displacement is also beneficial on themotor rotation which is more effective than other threefactors Compared to the size of check valves the electricalload has an equivalent degree of influence to the shaft speedIn addition it is obvious that the capacity of accumulator isless significant than other factors for shaft speed It can beconcluded that the optimal combination of the shaft speed asthe evaluation criterion is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
42 Hydraulic Motor Pressure As shown in Figure 7 thedisplacement of motor and the load of generator have almostequivalent influences on working pressure and they takeslightly lower impacts in comparison to the size of hydrauliccylinder Additionally it reveals that smaller check valvediameter can significantly lead to higher pressure+e higheraverage pressure means much more power can be producedby the generator to raise the regenerated power level+erefore optimal combination of hydraulic motor pressureis A1B1C1D1E1 where A1 5028mm B1 635mmC1 063 L D1 577 cc and E1 10Ω
43 Regenerated Power As shown in Figure 8 the size of thecylinder has significant influence on the power output of thepower circuit which is more important than other factorsObviously the small values of other factors can also con-tribute positive effects for more regenerated power From themean response analysis of Figure 8 the optimal combinationof the regenerated power is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
44 Regeneration Efficiency Regeneration efficiency is asignificant indicator to assess the performance of HERSA Asshown in Figure 9 the electrical load has the greatest in-fluence on the regeneration efficiency When the electricalload is 20Ω the proposed system can reach a high re-generation efficiency which increases slowly with the largervalue of electrical load Compared to other three indicators
Table 8 Orthogonal simulation test design list L16 (44 times 2)
NoFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Test 1 5028 635 063 577 10Test 2 5028 9525 063 707 20Test 3 5028 127 1 801 30Test 4 5028 1905 1 894 40Test 5 4025 635 1 707 30Test 6 4025 9525 1 577 40Test 7 4025 127 063 894 10Test 8 4025 1905 063 801 20Test 9 3220 635 063 801 40Test 10 3220 9525 063 894 30Test 11 3220 127 1 577 20Test 12 3220 1905 1 707 10Test 13 2516 635 1 894 20Test 14 2516 9525 1 801 10Test 15 2516 127 063 707 40Test 16 2516 1905 063 577 30
Shock and Vibration 7
higher regeneration efficiency can be found with larger thesize of check valve port and the capacity of accumulator It isclear that the larger size of hydraulic cylinder also con-tributes to regeneration efficiency However the optimalcombination of regeneration efficiency is A1B4C2D1E4where A1 5028mm B4 1905mm C2 1 L D1 577 ccand E4 40Ω
45 Optimisation Results From the mean response analysisof Figures 6ndash9 it can be concluded that the optimal com-bination of each parameter for more regenerated power onthe HERSA is A1B1C1D1E1 However a high regenerationefficiency is also one of the important criteria It is necessaryto adjust the optimal combination for the higher re-generation efficiency According to the range mean values ofdifferent factors the extent that the factor affects the in-dicator can be determined+e rank of factors based on theirdegree of importance and their corresponding best com-bination are shown in Table 10
It is obvious that the largest size of cylinder (A1) and thesmallest displacement of motor (D1) can provide betterdynamics and power level for the proposed HERSAAccording to the principle of the comprehensive equilib-rium method the selection of other factors is as follows
(a) Given the rank of factor B on different indicators isfourth the superior level selection is considered asthe one that occurs most frequently +e selection offactor B therefore is level 1 where B1 = 635mm
(b) Identically the rank of factor C is fifth on differentindicators Given the best level selection of factor Cfor three indicators is level 1 which is the one thatoccurs most frequently +erefore the selection offactor C is level 1 where C1 = 063 L
(c) When the factor has a different degree of influenceon all indicators the more important indicatorsshould be satisfied firstly +e regeneration efficiencyand regenerated power are the main criteria of thissystem When the factor E is selected the level 2 maybe better As shown in Figures 8 and 9 the efficiency
at level 2 is over 60 which is much higher than thatat level 1 and nearly to that at level 3 and 4Meanwhile the regenerated power at level 2 is alsohigh +e selection of factor E is level 2 whereE2 = 20Ω
+erefore the optimal combination of parameters inHERSA is A1B1C1D1E2 which is capable of delivering a largeelectrical power as well as a high regeneration efficiency +eoptimisation results are shown as follows +e size of cyl-inder is 50mm (piston) and 28mm (rod) the diameter ofcheck valve is 635mm the accumulator capacity is 063 Lthe displacement of motor is 577 cc and the electrical loadof generator is 20Ω respectively Furthermore for thegeneral HERSA system (including hydraulic cylinders checkvalves accumulators motors and generators) a set ofoptimisation methods about power regeneration perfor-mance can be summarised as follows
(a) Determine the range of key parameters(b) Design the corresponding orthogonal test(c) Conduct the mean response analysis(d) Discuss the optimisation results
It is considered that the proposed optimisation pro-cedure is suitable for the general HERSA system Figure 10shows the results of the HERSA applied with the optimalcomponent combinations
As shown in Figure 10 the shaft speed is 1634 rpm theaccumulator pressure is about 30 bar the power output isabout 331W and the regeneration efficiency is approxi-mately 65 Both larger power and higher regenerated ef-ficiency are achieved
46 Experimental Validation +e setup of the test rig isshown in Figure 11 According to the schematic in Figure 3the parameters of the key components are listed in Tables 6and 11 A corresponding test rig was designed and fabri-cated according to the design concept and model devel-opment to validate the prediction of the optimised HERSAmodel Based on the result of parameter optimisation the
Table 9 Summary of simulation tests
Test no Shaft speed (rpm) Hydraulic motor pressure (bar) Input power (W) Regenerated power (W) Regeneration efficiencyTest 1 16513928 464087 781801 4175 053402Test 2 13400569 198006 3427223 2239 065334Test 3 11756696 11384 2041776 1398 068482Test 4 1055314 7317 1371768 942 068667Test 5 8106328 88879 1015277 6643 065427Test 6 9885237 10621 1171773 8261 070496Test 7 6652814 11886 1295599 6665 050674Test 8 7275167 94879 659557 6596 062528Test 9 4612982 35676 300727 1798 059788Test 10 4107713 35638 296083 1706 057626Test 11 6481021 117337 818903 5231 063877Test 12 5315388 121918 847933 4318 050923Test 13 2516925 29404 154995 7888 05089Test 14 2842986 57556 263095 1235 046945Test 15 3120371 27366 145937 8228 056378Test 16 3888416 52267 242388 1528 063042
8 Shock and Vibration
key components of the test rig including hydraulic cyl-inder hydraulic accumulator hydraulic motor DC gen-erator inline check valve and hydraulic circuits of theHERSA system are carefully selected and equipped on thedesigned test bench +e HERSA test rig selected the pa-rameters of the main components that are closest to theoptimal results of the parameter optimisation Table 11shows the main parameter differences between model andtest rig
+e input controller and road actuator are designedwhich are able to provide predicted input excitations in-cluding excitation displacement excitation velocity and
frequency Data acquisition and transducers are also appliedto synchronise and measure the regenerated voltage andcurrent across the electrical load
+e voltage and current across the electrical load weremeasured on the design test rig +e measured regeneratedpower is compared with predicted power as shown inFigure 12 +e both results of simulation and measurementare under the harmonic excitation of 1Hz-25mm with anelectrical load of 20Ω It is worth to mention that themeasured average regenerated power is approximately326W compared to the predicted power of up to 331W andthe results show a good agreement especially for the visible
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65
0
200
400
600
800
1000
1200
1400
1600
1800Sh
a sp
eed
(rpm
)
t (s)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(a)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
10
Hyd
raul
ic m
otor
pre
ssur
e (ba
r)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(b)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
200
400
600
800
1000
1200
1400
1600
Mec
hani
cal p
ower
(W)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(c)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
100
200
300
400500
Rege
nera
ted
pow
er (W
)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(d)
Figure 5 Hydraulic motor shaft speed pressure mechanical power and regenerated power at different 16 tests
Shock and Vibration 9
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
low cost but may be damaged for the small magnetic polegap In 2007 Zhang et al [11] combined the ball screwsystem and DC motor to model a kind of power re-generation system Under the excitation of sine wave inputthis system can output the maximum voltage of 175 V In2012 Li et al [12] designed a built-in motion rectifierdevice for automotive suspension system It converts theuncertain movement into the steady rotation of generator+e simulation shows that the regeneration efficiency is upto 60 and the power output is nearly 15W when thevehicle travelled at a speed of 24 kmh In 2013 Fang et al[13] proposed a kind of hydraulic suspension system whichcan achieve the active control by using a pump to adjust thepressure in hydraulic cylinder In this system the checkvalves are used to control the direction of oil flow to ensurethe steady power output In 2014 Fang et al [14ndash16] de-veloped a hydraulic electromagnetic shock absorber whichcan regulate the fluid flow by applying a built-in hydraulicrectifier and internal accumulator in which the re-generation efficiency of the developed model is 166 at10Hz3mm harmonic excitation Li et al [17] designed andfabricated a hydraulic shock absorber prototype with ahydraulic rectifier to characterise and identify the severalkey mechanical and electrical parameters of an electro-mechanical model According to lack of model accuracythe parameter assumptions in the electromechanical modelcannot be always identified
In general the hydraulic electric regeneration shockabsorber (HERSA) has attracted wide attention due to itsstable work and compact structure [18] which is mainlyintegrated of hydraulic cylinder check valves (rectifiermodules) accumulators (stabilization modules) hydraulicmotors (power conversion modules) and generators (powerregeneration modules) [13ndash17 19ndash22] Some correspondingresearch studies have also been conducted based on theHERSA system In 2015 Xu and Guo [19] applied the geneticalgorithm in the power regenerating system to optimise theparameters of hydraulic pump +e results reveal that it canrecover power of 334W the excitation of sinusoidal wave is167Hz-50mm and the regeneration efficiency is around70 In 2016 Wang et al [20] designed a novel regenerativehydraulic shock absorber system A model which takes intoaccount the impacts of the dynamics of hydraulic flowrotational motion and power regeneration is developed Itcan be found that this system achieves recoverable power of260W with an efficiency of around 40 under sinusoidalexcitation of 1Hz and 25mm amplitude when the accu-mulator capacity is set 032 L with the load resistance 20ΩIn order to obtain a better damper performance in thehydraulic regenerative system Ahmad and Alam [21]analysed the influences of applied components used in thesystem such as hydraulic cylinder hydraulic motor andhydraulic accumulator In 2017 Zhang et al [22] proposed asemiactive regenerative suspension system+is model takesinto account the hydraulic losses in hydraulic system such asthrottle resistance of valves frictional resistance of pipelinesand resistance of motor
In previous studies it can be found that most researchpoints on the HERSA mainly focus on the improvement of
the modelling and experimental techniques but lack detailedresearch on parameter optimisation or optimal componentcombinations A HERSA system is therefore proposed andstudied in the following sections +is paper mainly focuseson the optimisation of power regeneration performanceparameters and studies key parameters such as the hydrauliccylinder size the check valve size the accumulator capacitythe motor displacement and the electrical load Based on theprinciple of orthogonal test the aim of the proposed HERSAis to find the optimal system parameters through modellinganalysis to provide a reliable basis for the prototype designon variable applications which can be universally used forgeneral HERSA system
As shown in Figure 1 this paper firstly introduces thevibration recovery potential and the modelling investigationof regeneration system Secondly the conceptual design ofHERSA system is introduced and then the model is createdfor optimising the parameters of key elements Finally theresults of optimal component combinations are pointed outand then the dynamic behaviours of the HERSA are dis-cussed and validated Based on the orthogonal test method apower regeneration optimisation method for generalHERSA system is also proposed in this paper
2 System Modelling of HERSA
21 Design Concept To understand the mechanism of aregenerative shock absorber system a mathematical model isdesigned on the basis of a typical configuration of the shockabsorber
As shown in Figure 2 a schematic design of the HERSAis proposed which consists of a hydraulic cylinder fourcheck valves a hydraulic accumulator a hydraulic motor apermanent magnetic generator pipelines battery and an oiltank +e key component of the system is the hydrauliccylinder that represents a conventional shock absorber ordamper
When the piston of hydraulic cylinder moves up anddown due to the excitation the shock oil in cylinder can beforced to flow through one-way valve and hydraulic accu-mulator and drive the hydraulic motor By the regulatingeffect of check valves the fluid flow will pass through themotor in one direction Meanwhile the accumulatorrsquossmoothing effect delivers a stable fluid flow to providehydraulic motor a more reliable and stable working con-dition Driven by the pressurised flow the hydraulic motorconverts linear motion into rotary motion as a motionconverter which will also drive the DC generator for thepurpose of power regeneration Finally the power regen-erated will be stored in battery or charged vehicle-mountedequipment
22 Hydraulic Flows +e road excitation is normally con-sisted by numerous sinusoidal waves [23] In order tosimplify the model the excitation is considered as the si-nusoidal wave which is defined as the fundamental elementof road irregularity +erefore the movement of piston isexpressed by
2 Shock and Vibration
Xa(t) X sin(2πft) (1)
where f is the excitation frequency and X is the maximumamplitude When the shaft moves up and down the volumechange of cap-end chamberVA and rod-end chamberVB canbe calculated by
VA AA X0 ∓Xa( 1113857 + Vcyd
VB AB X0 plusmn Xa( 1113857 + Vcyd
⎧⎨
⎩ up and down (2)
Vcyd is dead volume which refers to the fluid volume in thecylinder chambers at zero position It is considered as the
necessary part of volume change In addition AA and AB arethe areas of cap-end and rod-end on both sides of piston X0refers to the starting level of the piston
+e processes of fluid flows in designed hydraulic systemare shown in Figure 3 According to Bernoullirsquos principle
P +ρv2f
2+ ρgh constant (3)
whereP is the pressure ρ is the fluid density vf is the flow speedand h is the elevation of the point above a reference plane +eflow in cylinder and check valves can be expressed by
(a) Outflow of cylinder
QAout CCAC
2 PA minusPM1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PA gtPM
QBout CCAC
2 PB minusPM1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PB gtPM
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎩
(4)
(b) Inflow of cylinder
QAin CCAC
2 PM minusPA1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PM gtPA
QBin CCAC
2 PM minusPB1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PM gtPB
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎩
(5)
CC and AC are the flow coefficient and the area of inlinecheck valve port PA PB and PM represent the pressures
Introduction Researches of power regeneration system Aims of this paper
Accumulator flowNumerical modelling
Orthogonal simulation
Optimisation results andexperimental validation 50mm28mm 635mm
Hydraulic flow
Parameter optimisation of power regeneration on the hydraulic electric regenerative shock absorber system
Cylinder size
Check valve diameter
Accumulator capacity
Motor displacement
Electrical load
Test
desig
n
Key parameters
Sha speed
Hydraulic motor pressure
Regenerated power
Regeneration efficiency
Key indicators
Mea
n re
spon
se an
alys
is
Power regeneration
577cc063L 20Ω
Figure 1 +e study procedures of the HERSA
1
2
A3
4
B
56
7
8
9
10
(1) Accumulator(reserve cavity)
(2) Check valve(cap-end out)
(3) Cap-end chamber(4) Piston head(5) Check valve
(rod-end out)(6) Piston rod(7) Electric motor
(generator)(8) Hydraulic motor(9) Check valve
(cap-end in)(10) Check valve
(rod-end in)
(A) High-pressure side(B) Low-pressure side
Figure 2 Diagram view of the design concept for HERSA
Shock and Vibration 3
at the cap-end chamber the rod-end chamber and themotor inlet respectively
According to the influence of fluid compressibility inhydraulic elements the pressure out of the cylinder duringthe piston motion can be simultaneously expressed
(a) Up stroke
_PA βA AAv(t)minusQAout + QAin( 1113857
VA (6)
(b) Down stroke
_PB βB AB(minusv(t))minusQBout + QBin( 1113857
VB (7)
βA and βB are the effective bulk modulus in cap-endchamber and rod-end chamber
In general there are many empirical formulas for theeffective bulk modulus Given the relatively low pressure inthe HERSA system (under 100 bar) Boesrsquos model [24] isapplied to determine the bulk modulus of the fluid in thecylinder and motor
β βref middotlog 99PPref + 1( 1113857
2 (8)
According to the variations of the cylinder chambers(cap-end and rod-end chambers) the fluid flow rate andpressure in those two chambers can also oscillated signifi-cantly An accumulator is therefore used at the inlet ofmotor and the flow volume Vacf in accumulator can bewritten as
Vacf Vac 1minusPac
Pf1113888 1113889
1k
Pf gtPac (9)
where Vac is the volume of accumulator Pac is prechargedpressure to accumulator Pf is the fluid pressure of the ac-cumulator and k is the gas specific heat ratio of gas-chargedaccumulator and the fluid flow of accumulator can bewritten as
Qac CacAacsgn PM minusPf( 1113857
2 PM minusPf
11138681113868111386811138681113868
11138681113868111386811138681113868
ρ
11139741113972
(10)
where Cac is the accumulator flow coefficient and Aac is thearea of the accumulator inlet port +e volume variation ofaccumulator fluid Vp is
Vp minuskPfQac
_Pf (11)
Considering the smoothing effect of hydraulic accu-mulator a more accurate model of hydraulic motor can bedetermined including the variations of accumulator fluidflow and volume +erefore the pressure and fluid flow ofhydraulic motor can be represented
_PM βM QAout + QBout minusQac minusQM( 1113857
VT
QM DMωM
2π
(12)
where βM is the effective bulk modulus of the motorchamber DM is the displacement of the hydraulic motorand ωM is the shaft speed of the hydraulic motor andgenerator respectively
23 Power Regeneration Driven by the fluid flow the hy-draulic motor can produce rotary motion which is able todrive the DC generator to produce electricity +e workingprocess of power regeneration is shown in Figure 4
Where the mechanical efficiency of motor is ηM thedriving torque of motor TM is
TM DMPMηM
2π (13)
According to Newtonrsquos second law of motion the rotarymotion ωm is
_ωM TM minusTG( 1113857
Jt (14)
where Jt is the shaft moment of inertia In addition theelectromagnetic torque of generator TG would change withthe variation in the induced current and it is considered as theresistance torque because it provides the rotation to themotorin an opposite direction +erefore it can be written as
TG KTI (15)
+e electromotive force (EMF) can be expressed as
E KEωM (16)
UpDown
QBout
QBin
QAout
QAin
VA
VP
AP
PA PMωM
QMAA
PB
ACPf Qac
Aac
PacAB
VB
v
l
M
ρ
Figure 3 Schematic view of fluid flows in hydraulic system
4 Shock and Vibration
where KT is torque constant coefficient and KE is the elec-tromotive voltage constant coefficient
According to Kirchhoffrsquos voltage law [25] assuming thatthe susceptibility at any temperature and the flux that isestablished by the PM poles are constant the variation incurrent can be calculated by
_I Eminus Rin + RB( 1113857I( 1113857
LG (17)
where LG is the internal inductance RB is the externalelectrical load of battery and Rin is the internal resistance ofgenerator +en the instantaneous voltage can be expressedas
U IRB (18)
+erefore the output of regenerated power is
Pout I2RB
U2
RB (19)
In addition the effective input power of this system isconsidered as the sum of the piston damping force multi-plied by the effective piston area Given the areas of cap-endand rod-end are known the piston damping force can bewritten as
FA PAAA
FB PBAB1113896 (20)
+erefore the input power of the HERSA system is
Pin PAAA|v(t)| + PBAB|v(t)| up + down (21)
Hence the power regeneration efficiency can be calcu-lated from the following equation
ηreg Pout
Pin (22)
3 Simulation of HERSA
+is study focuses on the effect of the key parameters whichinclude the sizes of shock absorber body the size of checkvalve port hydraulic motor displacement and hydraulicaccumulator capacity Hence the investigation is performedon the proposed parameter optimisation method to studythe HERSArsquos behaviour and power level as well as thedesirable parameter solutions
31 Parameters Setting and Study In the modelling a fewassumptions made during this process were as follows
(a) +ere are no additional electrical losses in generatorconfiguration to be considered It means the outputpower of hydraulic motor is equal to the input powerof generator
(b) +e cylinder internal leakage between the chambersis not accounted for by the model
(c) +e external electrical load (batteryresistor) has noinfluence by the varying temperature
(d) Fluid compressibility is defined as using Boesrsquosmodel in hydraulic system
In practical applications there are various types of lossesin HERSA system such as hydraulic motor internal flowleakage pressure loss in the pipeline check valve pressureloss and hydraulic cylinder piston friction
To simplify the model the impact of these losses andinfluences in the system are not taken into account at thisstage +erefore the model is ideally configured and thefollowing main parameters are selected
311 Size of the Hydraulic Cylinder In this system thehydraulic cylinder is applied to replace the traditionalshock absorber and absorb vibration energy +is papertakes 4x4 SUV ldquoBeijing Jeep 2021rdquo as the example anddetermines that the maximum stroke of cylinder is 200mm+e sizes of the shock absorber body are determinedaccording to the standard ISO 3320-2013 [26] it is wellknown that a conventional viscous shock absorber hasasymmetrical damping characteristic due to its inherentdesign structure which can provide different dampingforces during the compression and extension strokes +episton diameter and rod diameter are therefore determinedas Table 1
312 Size of the Check Valve Port +e fluid pressurised bythe oscillation flows through check valve arrangement toensure fluid always flows through hydraulic motor in onedirection and enable the chambers in the cylinder can bereplenished as fast as possible for each run In terms of thestandard port size on double-acting cylinder the commonsizes of check valves are shown in Table 2
313 Hydraulic Accumulator Capacity A common hy-draulic accumulator is used to minimise the fluctuation ofthe pressurised flow Initially the gas chamber is prechargedto pressure Pac and set to 20 bar
In equations (9)ndash(11) several assumptions have beenmade to simplify the hydraulic accumulator model
(a) +e accumulator is set as a diaphragm type accu-mulator without heat exchange during the process
(b) +ere is a transient pressure balance inside the ac-cumulator between fluid chamber and gas chamber
(c) Frictions and thermal losses are neglected here
M G
DMTG
RB
KT
KE
LG
Rin
TM
JtPM
QM
ωM
Figure 4 Schematic view of energy conversion (mechanical toelectrical) and power circuit
Shock and Vibration 5
(d) Only fully charged and fully discharged states areconsidered in the hydraulic model
(e) +e precharge pressure in the accumulator is set at60 of the working pressure (20 bar) to providepressure pulsation damping
According to the damping forces in a traditional shockabsorber in a SUV the accumulator capacity has been es-timated with a peak of 100 bar which is shown in Table 3
314 Hydraulic Motor Displacement +e hydraulic motoris defined as a transfer device which is able to convert theunidirectional hydraulic flowpressure into rotationalmotiontorque Its main parameters are shown in Table 4
315 External Electrical Load For the regenerative shockabsorber the electrical load has significant effects on thecapability of power regeneration and the dynamics of thesystem [1] and it can be considered as Table 5
316 Other Key Parameters As shown in Table 6 the valuesof several other parameters are displayed
32 Orthogonal Simulation Test Design +e orthogonal testdesign is an important approach of statistical evaluationwhich can reduce the number of attempts and then effec-tively obtain desired results [27] Large-scale engineeringtesting is a complex system engineering +ere are manyfactors influencing the test results some factors play anindependent role and some factors can interact with othersto produce the comprehensive effect +e reason that affectsthe results during the test is called the test factor and thestate adopted under each factor is called the level Startingfrom Section 31 five-factor and four-level conditions havebeen applied in this test as shown in Table 7 +e simplestmethod is to apply an exhaustive method to test all factorsand levels at the same time
As shown in Table 7 the designed test has four 4-levelfactor and one 2-level factor If all factors and their levelsachieve their best performance for integration testing thenthe total of 512 tests need to be performed Obviously a largenumber of tests definitely demand a lot of manpower andmaterial resources as well as take a long time to reach thetarget+erefore this paper focuses on the problem that howto reasonably arrange the test and obtain the necessaryindication through less number of tests +e orthogonalmethod is an effective mathematical method to solve theproblem of multifactor test It also applies the orthogonaltable design scheme and the mathematical statistics methodto analyse the test data However the L16 (44 times 2) orthogonaltable (Table 8) is designed for the simulation test of HERSA
[28] and only 16 tests need to be carried out It is obviousthat the proposed parameter optimisation method can ef-fectively improve the efficiency of simulation and fastlyobtain the required simulation results
According to the above test design table numericalsimulation is carried out in mathematical model A sinu-soidal excitation 1Hz-25mm is set as predefined input Allother parameters are kept the same as in Table 6
Table 5 BatteryResistor resistance
Factor E Level 1 Level 2 Level 3 Level 4Electrical load (Ω) 10 20 30 40
Table 1 Hydraulic cylinder specification
Factor A Level 1 Level 2 Level 3 Level 4Piston diameter (mm) 50 40 32 25Rod diameter (mm) 28 25 20 16
Table 2 Check valves specification
Factor B Level 1 Level 2 Level 3 Level 4Check valve diameter (mm) 635 9525 127 1905
Table 3 Hydraulic accumulator specification
Factor C Level 1 Level 2Accumulator capacity (L) 063 1
Table 4 Hydraulic motor displacement
Factor D Level 1 Level 2 Level 3 Level 4Motor displacement (cc) 577 707 801 894
Table 6 Other key specifications of the HERSA
Symbol Value Unitf 1 HzXa 25 mmρ 872 kgm3
CC 07 mdashX0 200 mml 1 mβref 12times109 mdashPref 20 bark 14 mdashηm 100 Jt 0003 kgmiddotm2
KT 065 mdashKE 065 mdashRin 75 ΩLG 003 HVagd 01 V LDac Dcv mmCac 07 mdash
Table 7 Simulation test combinations
LevelFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Level 1 5028 635 063 577 10Level 2 4025 9525 1 707 20Level 3 3220 127 mdash 801 30Level 4 2516 1905 mdash 894 40
6 Shock and Vibration
33 Simulation Results and Discussion To use the modelwith orthogonal test approach developed in Sections 31and 32 for subsequent studies on the improvement ofparameter optimisation this section firstly presents studieson the investigation of key model parameters based onmodelling analysis +en it shows the quantitative be-haviours (pressure shaft speed regenerated power andregeneration efficiency) of the system under designed testtable gaining a preliminary understanding of the systembehaviours However the key results are summarised inTable 9 According to the principle of orthogonal test theaverages of outputs are calculated in the mean responseanalysis
As shown in Figure 5 the variation of the motor shaftspeed pressure mechanical power and regeneratedpower is dramatically levelled up at larger cylinder sizewith lower motor displacement accumulator capacity thesize of check valve and external load resistance +eaverages of those results are shown in Table 9 It is alsoclear that the change of component combination cansignificantly affect the system dynamics and regeneratedpower level to meet the demands of various vehiclesuspension systems By changing the component com-binations the maximum regenerated power of 4175Wwith the regeneration efficiency of approximately 534can be achieved on test 1 It is also obvious that the highestregeneration efficiency occurs at test 16 (approximately63) which is designed with small size of cylinder motordisplacement accumulator capacity and large check valveport Additionally the results of those 16 tests reveal thatthe change of motor displacement external loads andaccumulator capacity is capable of smoothing the flowoscillations and allow effective minimisation to the in-stability of hydraulic circuit thus altering the perfor-mance of the system and power capability
It can be summarised that the change of componentcombinations can not only affect the stability of systemdynamics but also dramatically impact on the level ofregenerated power
4 Parameter Optimisation Analysis
16 sets of simulation tests have been designed and performedto provide desirable solutions of the system parameters toenhance the system behaviours and power level After thatmean response analysis is used to take the average value of alltest results for each factor level and determine the extent thatthe factor affects the indicator based on the range meanvalues which is called the comprehensive equilibriummethod It can be applied to perform the multi-indicatoranalysis of shaft speed hydraulic motor pressure re-generative power and regeneration efficiency
41 Shaft Speed As shown in Figure 6 the size of cylinderhas the greatest influence on shaft speed with a visibly largerange+e larger cylinder size can deliver faster flowwhich iscapable of providing larger rotational torque on motor shaft+e smaller motor displacement is also beneficial on themotor rotation which is more effective than other threefactors Compared to the size of check valves the electricalload has an equivalent degree of influence to the shaft speedIn addition it is obvious that the capacity of accumulator isless significant than other factors for shaft speed It can beconcluded that the optimal combination of the shaft speed asthe evaluation criterion is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
42 Hydraulic Motor Pressure As shown in Figure 7 thedisplacement of motor and the load of generator have almostequivalent influences on working pressure and they takeslightly lower impacts in comparison to the size of hydrauliccylinder Additionally it reveals that smaller check valvediameter can significantly lead to higher pressure+e higheraverage pressure means much more power can be producedby the generator to raise the regenerated power level+erefore optimal combination of hydraulic motor pressureis A1B1C1D1E1 where A1 5028mm B1 635mmC1 063 L D1 577 cc and E1 10Ω
43 Regenerated Power As shown in Figure 8 the size of thecylinder has significant influence on the power output of thepower circuit which is more important than other factorsObviously the small values of other factors can also con-tribute positive effects for more regenerated power From themean response analysis of Figure 8 the optimal combinationof the regenerated power is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
44 Regeneration Efficiency Regeneration efficiency is asignificant indicator to assess the performance of HERSA Asshown in Figure 9 the electrical load has the greatest in-fluence on the regeneration efficiency When the electricalload is 20Ω the proposed system can reach a high re-generation efficiency which increases slowly with the largervalue of electrical load Compared to other three indicators
Table 8 Orthogonal simulation test design list L16 (44 times 2)
NoFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Test 1 5028 635 063 577 10Test 2 5028 9525 063 707 20Test 3 5028 127 1 801 30Test 4 5028 1905 1 894 40Test 5 4025 635 1 707 30Test 6 4025 9525 1 577 40Test 7 4025 127 063 894 10Test 8 4025 1905 063 801 20Test 9 3220 635 063 801 40Test 10 3220 9525 063 894 30Test 11 3220 127 1 577 20Test 12 3220 1905 1 707 10Test 13 2516 635 1 894 20Test 14 2516 9525 1 801 10Test 15 2516 127 063 707 40Test 16 2516 1905 063 577 30
Shock and Vibration 7
higher regeneration efficiency can be found with larger thesize of check valve port and the capacity of accumulator It isclear that the larger size of hydraulic cylinder also con-tributes to regeneration efficiency However the optimalcombination of regeneration efficiency is A1B4C2D1E4where A1 5028mm B4 1905mm C2 1 L D1 577 ccand E4 40Ω
45 Optimisation Results From the mean response analysisof Figures 6ndash9 it can be concluded that the optimal com-bination of each parameter for more regenerated power onthe HERSA is A1B1C1D1E1 However a high regenerationefficiency is also one of the important criteria It is necessaryto adjust the optimal combination for the higher re-generation efficiency According to the range mean values ofdifferent factors the extent that the factor affects the in-dicator can be determined+e rank of factors based on theirdegree of importance and their corresponding best com-bination are shown in Table 10
It is obvious that the largest size of cylinder (A1) and thesmallest displacement of motor (D1) can provide betterdynamics and power level for the proposed HERSAAccording to the principle of the comprehensive equilib-rium method the selection of other factors is as follows
(a) Given the rank of factor B on different indicators isfourth the superior level selection is considered asthe one that occurs most frequently +e selection offactor B therefore is level 1 where B1 = 635mm
(b) Identically the rank of factor C is fifth on differentindicators Given the best level selection of factor Cfor three indicators is level 1 which is the one thatoccurs most frequently +erefore the selection offactor C is level 1 where C1 = 063 L
(c) When the factor has a different degree of influenceon all indicators the more important indicatorsshould be satisfied firstly +e regeneration efficiencyand regenerated power are the main criteria of thissystem When the factor E is selected the level 2 maybe better As shown in Figures 8 and 9 the efficiency
at level 2 is over 60 which is much higher than thatat level 1 and nearly to that at level 3 and 4Meanwhile the regenerated power at level 2 is alsohigh +e selection of factor E is level 2 whereE2 = 20Ω
+erefore the optimal combination of parameters inHERSA is A1B1C1D1E2 which is capable of delivering a largeelectrical power as well as a high regeneration efficiency +eoptimisation results are shown as follows +e size of cyl-inder is 50mm (piston) and 28mm (rod) the diameter ofcheck valve is 635mm the accumulator capacity is 063 Lthe displacement of motor is 577 cc and the electrical loadof generator is 20Ω respectively Furthermore for thegeneral HERSA system (including hydraulic cylinders checkvalves accumulators motors and generators) a set ofoptimisation methods about power regeneration perfor-mance can be summarised as follows
(a) Determine the range of key parameters(b) Design the corresponding orthogonal test(c) Conduct the mean response analysis(d) Discuss the optimisation results
It is considered that the proposed optimisation pro-cedure is suitable for the general HERSA system Figure 10shows the results of the HERSA applied with the optimalcomponent combinations
As shown in Figure 10 the shaft speed is 1634 rpm theaccumulator pressure is about 30 bar the power output isabout 331W and the regeneration efficiency is approxi-mately 65 Both larger power and higher regenerated ef-ficiency are achieved
46 Experimental Validation +e setup of the test rig isshown in Figure 11 According to the schematic in Figure 3the parameters of the key components are listed in Tables 6and 11 A corresponding test rig was designed and fabri-cated according to the design concept and model devel-opment to validate the prediction of the optimised HERSAmodel Based on the result of parameter optimisation the
Table 9 Summary of simulation tests
Test no Shaft speed (rpm) Hydraulic motor pressure (bar) Input power (W) Regenerated power (W) Regeneration efficiencyTest 1 16513928 464087 781801 4175 053402Test 2 13400569 198006 3427223 2239 065334Test 3 11756696 11384 2041776 1398 068482Test 4 1055314 7317 1371768 942 068667Test 5 8106328 88879 1015277 6643 065427Test 6 9885237 10621 1171773 8261 070496Test 7 6652814 11886 1295599 6665 050674Test 8 7275167 94879 659557 6596 062528Test 9 4612982 35676 300727 1798 059788Test 10 4107713 35638 296083 1706 057626Test 11 6481021 117337 818903 5231 063877Test 12 5315388 121918 847933 4318 050923Test 13 2516925 29404 154995 7888 05089Test 14 2842986 57556 263095 1235 046945Test 15 3120371 27366 145937 8228 056378Test 16 3888416 52267 242388 1528 063042
8 Shock and Vibration
key components of the test rig including hydraulic cyl-inder hydraulic accumulator hydraulic motor DC gen-erator inline check valve and hydraulic circuits of theHERSA system are carefully selected and equipped on thedesigned test bench +e HERSA test rig selected the pa-rameters of the main components that are closest to theoptimal results of the parameter optimisation Table 11shows the main parameter differences between model andtest rig
+e input controller and road actuator are designedwhich are able to provide predicted input excitations in-cluding excitation displacement excitation velocity and
frequency Data acquisition and transducers are also appliedto synchronise and measure the regenerated voltage andcurrent across the electrical load
+e voltage and current across the electrical load weremeasured on the design test rig +e measured regeneratedpower is compared with predicted power as shown inFigure 12 +e both results of simulation and measurementare under the harmonic excitation of 1Hz-25mm with anelectrical load of 20Ω It is worth to mention that themeasured average regenerated power is approximately326W compared to the predicted power of up to 331W andthe results show a good agreement especially for the visible
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65
0
200
400
600
800
1000
1200
1400
1600
1800Sh
a sp
eed
(rpm
)
t (s)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(a)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
10
Hyd
raul
ic m
otor
pre
ssur
e (ba
r)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(b)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
200
400
600
800
1000
1200
1400
1600
Mec
hani
cal p
ower
(W)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(c)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
100
200
300
400500
Rege
nera
ted
pow
er (W
)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(d)
Figure 5 Hydraulic motor shaft speed pressure mechanical power and regenerated power at different 16 tests
Shock and Vibration 9
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
Xa(t) X sin(2πft) (1)
where f is the excitation frequency and X is the maximumamplitude When the shaft moves up and down the volumechange of cap-end chamberVA and rod-end chamberVB canbe calculated by
VA AA X0 ∓Xa( 1113857 + Vcyd
VB AB X0 plusmn Xa( 1113857 + Vcyd
⎧⎨
⎩ up and down (2)
Vcyd is dead volume which refers to the fluid volume in thecylinder chambers at zero position It is considered as the
necessary part of volume change In addition AA and AB arethe areas of cap-end and rod-end on both sides of piston X0refers to the starting level of the piston
+e processes of fluid flows in designed hydraulic systemare shown in Figure 3 According to Bernoullirsquos principle
P +ρv2f
2+ ρgh constant (3)
whereP is the pressure ρ is the fluid density vf is the flow speedand h is the elevation of the point above a reference plane +eflow in cylinder and check valves can be expressed by
(a) Outflow of cylinder
QAout CCAC
2 PA minusPM1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PA gtPM
QBout CCAC
2 PB minusPM1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PB gtPM
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎩
(4)
(b) Inflow of cylinder
QAin CCAC
2 PM minusPA1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PM gtPA
QBin CCAC
2 PM minusPB1113868111386811138681113868
1113868111386811138681113868
ρ
1113971
PM gtPB
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎩
(5)
CC and AC are the flow coefficient and the area of inlinecheck valve port PA PB and PM represent the pressures
Introduction Researches of power regeneration system Aims of this paper
Accumulator flowNumerical modelling
Orthogonal simulation
Optimisation results andexperimental validation 50mm28mm 635mm
Hydraulic flow
Parameter optimisation of power regeneration on the hydraulic electric regenerative shock absorber system
Cylinder size
Check valve diameter
Accumulator capacity
Motor displacement
Electrical load
Test
desig
n
Key parameters
Sha speed
Hydraulic motor pressure
Regenerated power
Regeneration efficiency
Key indicators
Mea
n re
spon
se an
alys
is
Power regeneration
577cc063L 20Ω
Figure 1 +e study procedures of the HERSA
1
2
A3
4
B
56
7
8
9
10
(1) Accumulator(reserve cavity)
(2) Check valve(cap-end out)
(3) Cap-end chamber(4) Piston head(5) Check valve
(rod-end out)(6) Piston rod(7) Electric motor
(generator)(8) Hydraulic motor(9) Check valve
(cap-end in)(10) Check valve
(rod-end in)
(A) High-pressure side(B) Low-pressure side
Figure 2 Diagram view of the design concept for HERSA
Shock and Vibration 3
at the cap-end chamber the rod-end chamber and themotor inlet respectively
According to the influence of fluid compressibility inhydraulic elements the pressure out of the cylinder duringthe piston motion can be simultaneously expressed
(a) Up stroke
_PA βA AAv(t)minusQAout + QAin( 1113857
VA (6)
(b) Down stroke
_PB βB AB(minusv(t))minusQBout + QBin( 1113857
VB (7)
βA and βB are the effective bulk modulus in cap-endchamber and rod-end chamber
In general there are many empirical formulas for theeffective bulk modulus Given the relatively low pressure inthe HERSA system (under 100 bar) Boesrsquos model [24] isapplied to determine the bulk modulus of the fluid in thecylinder and motor
β βref middotlog 99PPref + 1( 1113857
2 (8)
According to the variations of the cylinder chambers(cap-end and rod-end chambers) the fluid flow rate andpressure in those two chambers can also oscillated signifi-cantly An accumulator is therefore used at the inlet ofmotor and the flow volume Vacf in accumulator can bewritten as
Vacf Vac 1minusPac
Pf1113888 1113889
1k
Pf gtPac (9)
where Vac is the volume of accumulator Pac is prechargedpressure to accumulator Pf is the fluid pressure of the ac-cumulator and k is the gas specific heat ratio of gas-chargedaccumulator and the fluid flow of accumulator can bewritten as
Qac CacAacsgn PM minusPf( 1113857
2 PM minusPf
11138681113868111386811138681113868
11138681113868111386811138681113868
ρ
11139741113972
(10)
where Cac is the accumulator flow coefficient and Aac is thearea of the accumulator inlet port +e volume variation ofaccumulator fluid Vp is
Vp minuskPfQac
_Pf (11)
Considering the smoothing effect of hydraulic accu-mulator a more accurate model of hydraulic motor can bedetermined including the variations of accumulator fluidflow and volume +erefore the pressure and fluid flow ofhydraulic motor can be represented
_PM βM QAout + QBout minusQac minusQM( 1113857
VT
QM DMωM
2π
(12)
where βM is the effective bulk modulus of the motorchamber DM is the displacement of the hydraulic motorand ωM is the shaft speed of the hydraulic motor andgenerator respectively
23 Power Regeneration Driven by the fluid flow the hy-draulic motor can produce rotary motion which is able todrive the DC generator to produce electricity +e workingprocess of power regeneration is shown in Figure 4
Where the mechanical efficiency of motor is ηM thedriving torque of motor TM is
TM DMPMηM
2π (13)
According to Newtonrsquos second law of motion the rotarymotion ωm is
_ωM TM minusTG( 1113857
Jt (14)
where Jt is the shaft moment of inertia In addition theelectromagnetic torque of generator TG would change withthe variation in the induced current and it is considered as theresistance torque because it provides the rotation to themotorin an opposite direction +erefore it can be written as
TG KTI (15)
+e electromotive force (EMF) can be expressed as
E KEωM (16)
UpDown
QBout
QBin
QAout
QAin
VA
VP
AP
PA PMωM
QMAA
PB
ACPf Qac
Aac
PacAB
VB
v
l
M
ρ
Figure 3 Schematic view of fluid flows in hydraulic system
4 Shock and Vibration
where KT is torque constant coefficient and KE is the elec-tromotive voltage constant coefficient
According to Kirchhoffrsquos voltage law [25] assuming thatthe susceptibility at any temperature and the flux that isestablished by the PM poles are constant the variation incurrent can be calculated by
_I Eminus Rin + RB( 1113857I( 1113857
LG (17)
where LG is the internal inductance RB is the externalelectrical load of battery and Rin is the internal resistance ofgenerator +en the instantaneous voltage can be expressedas
U IRB (18)
+erefore the output of regenerated power is
Pout I2RB
U2
RB (19)
In addition the effective input power of this system isconsidered as the sum of the piston damping force multi-plied by the effective piston area Given the areas of cap-endand rod-end are known the piston damping force can bewritten as
FA PAAA
FB PBAB1113896 (20)
+erefore the input power of the HERSA system is
Pin PAAA|v(t)| + PBAB|v(t)| up + down (21)
Hence the power regeneration efficiency can be calcu-lated from the following equation
ηreg Pout
Pin (22)
3 Simulation of HERSA
+is study focuses on the effect of the key parameters whichinclude the sizes of shock absorber body the size of checkvalve port hydraulic motor displacement and hydraulicaccumulator capacity Hence the investigation is performedon the proposed parameter optimisation method to studythe HERSArsquos behaviour and power level as well as thedesirable parameter solutions
31 Parameters Setting and Study In the modelling a fewassumptions made during this process were as follows
(a) +ere are no additional electrical losses in generatorconfiguration to be considered It means the outputpower of hydraulic motor is equal to the input powerof generator
(b) +e cylinder internal leakage between the chambersis not accounted for by the model
(c) +e external electrical load (batteryresistor) has noinfluence by the varying temperature
(d) Fluid compressibility is defined as using Boesrsquosmodel in hydraulic system
In practical applications there are various types of lossesin HERSA system such as hydraulic motor internal flowleakage pressure loss in the pipeline check valve pressureloss and hydraulic cylinder piston friction
To simplify the model the impact of these losses andinfluences in the system are not taken into account at thisstage +erefore the model is ideally configured and thefollowing main parameters are selected
311 Size of the Hydraulic Cylinder In this system thehydraulic cylinder is applied to replace the traditionalshock absorber and absorb vibration energy +is papertakes 4x4 SUV ldquoBeijing Jeep 2021rdquo as the example anddetermines that the maximum stroke of cylinder is 200mm+e sizes of the shock absorber body are determinedaccording to the standard ISO 3320-2013 [26] it is wellknown that a conventional viscous shock absorber hasasymmetrical damping characteristic due to its inherentdesign structure which can provide different dampingforces during the compression and extension strokes +episton diameter and rod diameter are therefore determinedas Table 1
312 Size of the Check Valve Port +e fluid pressurised bythe oscillation flows through check valve arrangement toensure fluid always flows through hydraulic motor in onedirection and enable the chambers in the cylinder can bereplenished as fast as possible for each run In terms of thestandard port size on double-acting cylinder the commonsizes of check valves are shown in Table 2
313 Hydraulic Accumulator Capacity A common hy-draulic accumulator is used to minimise the fluctuation ofthe pressurised flow Initially the gas chamber is prechargedto pressure Pac and set to 20 bar
In equations (9)ndash(11) several assumptions have beenmade to simplify the hydraulic accumulator model
(a) +e accumulator is set as a diaphragm type accu-mulator without heat exchange during the process
(b) +ere is a transient pressure balance inside the ac-cumulator between fluid chamber and gas chamber
(c) Frictions and thermal losses are neglected here
M G
DMTG
RB
KT
KE
LG
Rin
TM
JtPM
QM
ωM
Figure 4 Schematic view of energy conversion (mechanical toelectrical) and power circuit
Shock and Vibration 5
(d) Only fully charged and fully discharged states areconsidered in the hydraulic model
(e) +e precharge pressure in the accumulator is set at60 of the working pressure (20 bar) to providepressure pulsation damping
According to the damping forces in a traditional shockabsorber in a SUV the accumulator capacity has been es-timated with a peak of 100 bar which is shown in Table 3
314 Hydraulic Motor Displacement +e hydraulic motoris defined as a transfer device which is able to convert theunidirectional hydraulic flowpressure into rotationalmotiontorque Its main parameters are shown in Table 4
315 External Electrical Load For the regenerative shockabsorber the electrical load has significant effects on thecapability of power regeneration and the dynamics of thesystem [1] and it can be considered as Table 5
316 Other Key Parameters As shown in Table 6 the valuesof several other parameters are displayed
32 Orthogonal Simulation Test Design +e orthogonal testdesign is an important approach of statistical evaluationwhich can reduce the number of attempts and then effec-tively obtain desired results [27] Large-scale engineeringtesting is a complex system engineering +ere are manyfactors influencing the test results some factors play anindependent role and some factors can interact with othersto produce the comprehensive effect +e reason that affectsthe results during the test is called the test factor and thestate adopted under each factor is called the level Startingfrom Section 31 five-factor and four-level conditions havebeen applied in this test as shown in Table 7 +e simplestmethod is to apply an exhaustive method to test all factorsand levels at the same time
As shown in Table 7 the designed test has four 4-levelfactor and one 2-level factor If all factors and their levelsachieve their best performance for integration testing thenthe total of 512 tests need to be performed Obviously a largenumber of tests definitely demand a lot of manpower andmaterial resources as well as take a long time to reach thetarget+erefore this paper focuses on the problem that howto reasonably arrange the test and obtain the necessaryindication through less number of tests +e orthogonalmethod is an effective mathematical method to solve theproblem of multifactor test It also applies the orthogonaltable design scheme and the mathematical statistics methodto analyse the test data However the L16 (44 times 2) orthogonaltable (Table 8) is designed for the simulation test of HERSA
[28] and only 16 tests need to be carried out It is obviousthat the proposed parameter optimisation method can ef-fectively improve the efficiency of simulation and fastlyobtain the required simulation results
According to the above test design table numericalsimulation is carried out in mathematical model A sinu-soidal excitation 1Hz-25mm is set as predefined input Allother parameters are kept the same as in Table 6
Table 5 BatteryResistor resistance
Factor E Level 1 Level 2 Level 3 Level 4Electrical load (Ω) 10 20 30 40
Table 1 Hydraulic cylinder specification
Factor A Level 1 Level 2 Level 3 Level 4Piston diameter (mm) 50 40 32 25Rod diameter (mm) 28 25 20 16
Table 2 Check valves specification
Factor B Level 1 Level 2 Level 3 Level 4Check valve diameter (mm) 635 9525 127 1905
Table 3 Hydraulic accumulator specification
Factor C Level 1 Level 2Accumulator capacity (L) 063 1
Table 4 Hydraulic motor displacement
Factor D Level 1 Level 2 Level 3 Level 4Motor displacement (cc) 577 707 801 894
Table 6 Other key specifications of the HERSA
Symbol Value Unitf 1 HzXa 25 mmρ 872 kgm3
CC 07 mdashX0 200 mml 1 mβref 12times109 mdashPref 20 bark 14 mdashηm 100 Jt 0003 kgmiddotm2
KT 065 mdashKE 065 mdashRin 75 ΩLG 003 HVagd 01 V LDac Dcv mmCac 07 mdash
Table 7 Simulation test combinations
LevelFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Level 1 5028 635 063 577 10Level 2 4025 9525 1 707 20Level 3 3220 127 mdash 801 30Level 4 2516 1905 mdash 894 40
6 Shock and Vibration
33 Simulation Results and Discussion To use the modelwith orthogonal test approach developed in Sections 31and 32 for subsequent studies on the improvement ofparameter optimisation this section firstly presents studieson the investigation of key model parameters based onmodelling analysis +en it shows the quantitative be-haviours (pressure shaft speed regenerated power andregeneration efficiency) of the system under designed testtable gaining a preliminary understanding of the systembehaviours However the key results are summarised inTable 9 According to the principle of orthogonal test theaverages of outputs are calculated in the mean responseanalysis
As shown in Figure 5 the variation of the motor shaftspeed pressure mechanical power and regeneratedpower is dramatically levelled up at larger cylinder sizewith lower motor displacement accumulator capacity thesize of check valve and external load resistance +eaverages of those results are shown in Table 9 It is alsoclear that the change of component combination cansignificantly affect the system dynamics and regeneratedpower level to meet the demands of various vehiclesuspension systems By changing the component com-binations the maximum regenerated power of 4175Wwith the regeneration efficiency of approximately 534can be achieved on test 1 It is also obvious that the highestregeneration efficiency occurs at test 16 (approximately63) which is designed with small size of cylinder motordisplacement accumulator capacity and large check valveport Additionally the results of those 16 tests reveal thatthe change of motor displacement external loads andaccumulator capacity is capable of smoothing the flowoscillations and allow effective minimisation to the in-stability of hydraulic circuit thus altering the perfor-mance of the system and power capability
It can be summarised that the change of componentcombinations can not only affect the stability of systemdynamics but also dramatically impact on the level ofregenerated power
4 Parameter Optimisation Analysis
16 sets of simulation tests have been designed and performedto provide desirable solutions of the system parameters toenhance the system behaviours and power level After thatmean response analysis is used to take the average value of alltest results for each factor level and determine the extent thatthe factor affects the indicator based on the range meanvalues which is called the comprehensive equilibriummethod It can be applied to perform the multi-indicatoranalysis of shaft speed hydraulic motor pressure re-generative power and regeneration efficiency
41 Shaft Speed As shown in Figure 6 the size of cylinderhas the greatest influence on shaft speed with a visibly largerange+e larger cylinder size can deliver faster flowwhich iscapable of providing larger rotational torque on motor shaft+e smaller motor displacement is also beneficial on themotor rotation which is more effective than other threefactors Compared to the size of check valves the electricalload has an equivalent degree of influence to the shaft speedIn addition it is obvious that the capacity of accumulator isless significant than other factors for shaft speed It can beconcluded that the optimal combination of the shaft speed asthe evaluation criterion is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
42 Hydraulic Motor Pressure As shown in Figure 7 thedisplacement of motor and the load of generator have almostequivalent influences on working pressure and they takeslightly lower impacts in comparison to the size of hydrauliccylinder Additionally it reveals that smaller check valvediameter can significantly lead to higher pressure+e higheraverage pressure means much more power can be producedby the generator to raise the regenerated power level+erefore optimal combination of hydraulic motor pressureis A1B1C1D1E1 where A1 5028mm B1 635mmC1 063 L D1 577 cc and E1 10Ω
43 Regenerated Power As shown in Figure 8 the size of thecylinder has significant influence on the power output of thepower circuit which is more important than other factorsObviously the small values of other factors can also con-tribute positive effects for more regenerated power From themean response analysis of Figure 8 the optimal combinationof the regenerated power is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
44 Regeneration Efficiency Regeneration efficiency is asignificant indicator to assess the performance of HERSA Asshown in Figure 9 the electrical load has the greatest in-fluence on the regeneration efficiency When the electricalload is 20Ω the proposed system can reach a high re-generation efficiency which increases slowly with the largervalue of electrical load Compared to other three indicators
Table 8 Orthogonal simulation test design list L16 (44 times 2)
NoFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Test 1 5028 635 063 577 10Test 2 5028 9525 063 707 20Test 3 5028 127 1 801 30Test 4 5028 1905 1 894 40Test 5 4025 635 1 707 30Test 6 4025 9525 1 577 40Test 7 4025 127 063 894 10Test 8 4025 1905 063 801 20Test 9 3220 635 063 801 40Test 10 3220 9525 063 894 30Test 11 3220 127 1 577 20Test 12 3220 1905 1 707 10Test 13 2516 635 1 894 20Test 14 2516 9525 1 801 10Test 15 2516 127 063 707 40Test 16 2516 1905 063 577 30
Shock and Vibration 7
higher regeneration efficiency can be found with larger thesize of check valve port and the capacity of accumulator It isclear that the larger size of hydraulic cylinder also con-tributes to regeneration efficiency However the optimalcombination of regeneration efficiency is A1B4C2D1E4where A1 5028mm B4 1905mm C2 1 L D1 577 ccand E4 40Ω
45 Optimisation Results From the mean response analysisof Figures 6ndash9 it can be concluded that the optimal com-bination of each parameter for more regenerated power onthe HERSA is A1B1C1D1E1 However a high regenerationefficiency is also one of the important criteria It is necessaryto adjust the optimal combination for the higher re-generation efficiency According to the range mean values ofdifferent factors the extent that the factor affects the in-dicator can be determined+e rank of factors based on theirdegree of importance and their corresponding best com-bination are shown in Table 10
It is obvious that the largest size of cylinder (A1) and thesmallest displacement of motor (D1) can provide betterdynamics and power level for the proposed HERSAAccording to the principle of the comprehensive equilib-rium method the selection of other factors is as follows
(a) Given the rank of factor B on different indicators isfourth the superior level selection is considered asthe one that occurs most frequently +e selection offactor B therefore is level 1 where B1 = 635mm
(b) Identically the rank of factor C is fifth on differentindicators Given the best level selection of factor Cfor three indicators is level 1 which is the one thatoccurs most frequently +erefore the selection offactor C is level 1 where C1 = 063 L
(c) When the factor has a different degree of influenceon all indicators the more important indicatorsshould be satisfied firstly +e regeneration efficiencyand regenerated power are the main criteria of thissystem When the factor E is selected the level 2 maybe better As shown in Figures 8 and 9 the efficiency
at level 2 is over 60 which is much higher than thatat level 1 and nearly to that at level 3 and 4Meanwhile the regenerated power at level 2 is alsohigh +e selection of factor E is level 2 whereE2 = 20Ω
+erefore the optimal combination of parameters inHERSA is A1B1C1D1E2 which is capable of delivering a largeelectrical power as well as a high regeneration efficiency +eoptimisation results are shown as follows +e size of cyl-inder is 50mm (piston) and 28mm (rod) the diameter ofcheck valve is 635mm the accumulator capacity is 063 Lthe displacement of motor is 577 cc and the electrical loadof generator is 20Ω respectively Furthermore for thegeneral HERSA system (including hydraulic cylinders checkvalves accumulators motors and generators) a set ofoptimisation methods about power regeneration perfor-mance can be summarised as follows
(a) Determine the range of key parameters(b) Design the corresponding orthogonal test(c) Conduct the mean response analysis(d) Discuss the optimisation results
It is considered that the proposed optimisation pro-cedure is suitable for the general HERSA system Figure 10shows the results of the HERSA applied with the optimalcomponent combinations
As shown in Figure 10 the shaft speed is 1634 rpm theaccumulator pressure is about 30 bar the power output isabout 331W and the regeneration efficiency is approxi-mately 65 Both larger power and higher regenerated ef-ficiency are achieved
46 Experimental Validation +e setup of the test rig isshown in Figure 11 According to the schematic in Figure 3the parameters of the key components are listed in Tables 6and 11 A corresponding test rig was designed and fabri-cated according to the design concept and model devel-opment to validate the prediction of the optimised HERSAmodel Based on the result of parameter optimisation the
Table 9 Summary of simulation tests
Test no Shaft speed (rpm) Hydraulic motor pressure (bar) Input power (W) Regenerated power (W) Regeneration efficiencyTest 1 16513928 464087 781801 4175 053402Test 2 13400569 198006 3427223 2239 065334Test 3 11756696 11384 2041776 1398 068482Test 4 1055314 7317 1371768 942 068667Test 5 8106328 88879 1015277 6643 065427Test 6 9885237 10621 1171773 8261 070496Test 7 6652814 11886 1295599 6665 050674Test 8 7275167 94879 659557 6596 062528Test 9 4612982 35676 300727 1798 059788Test 10 4107713 35638 296083 1706 057626Test 11 6481021 117337 818903 5231 063877Test 12 5315388 121918 847933 4318 050923Test 13 2516925 29404 154995 7888 05089Test 14 2842986 57556 263095 1235 046945Test 15 3120371 27366 145937 8228 056378Test 16 3888416 52267 242388 1528 063042
8 Shock and Vibration
key components of the test rig including hydraulic cyl-inder hydraulic accumulator hydraulic motor DC gen-erator inline check valve and hydraulic circuits of theHERSA system are carefully selected and equipped on thedesigned test bench +e HERSA test rig selected the pa-rameters of the main components that are closest to theoptimal results of the parameter optimisation Table 11shows the main parameter differences between model andtest rig
+e input controller and road actuator are designedwhich are able to provide predicted input excitations in-cluding excitation displacement excitation velocity and
frequency Data acquisition and transducers are also appliedto synchronise and measure the regenerated voltage andcurrent across the electrical load
+e voltage and current across the electrical load weremeasured on the design test rig +e measured regeneratedpower is compared with predicted power as shown inFigure 12 +e both results of simulation and measurementare under the harmonic excitation of 1Hz-25mm with anelectrical load of 20Ω It is worth to mention that themeasured average regenerated power is approximately326W compared to the predicted power of up to 331W andthe results show a good agreement especially for the visible
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65
0
200
400
600
800
1000
1200
1400
1600
1800Sh
a sp
eed
(rpm
)
t (s)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(a)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
10
Hyd
raul
ic m
otor
pre
ssur
e (ba
r)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(b)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
200
400
600
800
1000
1200
1400
1600
Mec
hani
cal p
ower
(W)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(c)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
100
200
300
400500
Rege
nera
ted
pow
er (W
)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(d)
Figure 5 Hydraulic motor shaft speed pressure mechanical power and regenerated power at different 16 tests
Shock and Vibration 9
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
at the cap-end chamber the rod-end chamber and themotor inlet respectively
According to the influence of fluid compressibility inhydraulic elements the pressure out of the cylinder duringthe piston motion can be simultaneously expressed
(a) Up stroke
_PA βA AAv(t)minusQAout + QAin( 1113857
VA (6)
(b) Down stroke
_PB βB AB(minusv(t))minusQBout + QBin( 1113857
VB (7)
βA and βB are the effective bulk modulus in cap-endchamber and rod-end chamber
In general there are many empirical formulas for theeffective bulk modulus Given the relatively low pressure inthe HERSA system (under 100 bar) Boesrsquos model [24] isapplied to determine the bulk modulus of the fluid in thecylinder and motor
β βref middotlog 99PPref + 1( 1113857
2 (8)
According to the variations of the cylinder chambers(cap-end and rod-end chambers) the fluid flow rate andpressure in those two chambers can also oscillated signifi-cantly An accumulator is therefore used at the inlet ofmotor and the flow volume Vacf in accumulator can bewritten as
Vacf Vac 1minusPac
Pf1113888 1113889
1k
Pf gtPac (9)
where Vac is the volume of accumulator Pac is prechargedpressure to accumulator Pf is the fluid pressure of the ac-cumulator and k is the gas specific heat ratio of gas-chargedaccumulator and the fluid flow of accumulator can bewritten as
Qac CacAacsgn PM minusPf( 1113857
2 PM minusPf
11138681113868111386811138681113868
11138681113868111386811138681113868
ρ
11139741113972
(10)
where Cac is the accumulator flow coefficient and Aac is thearea of the accumulator inlet port +e volume variation ofaccumulator fluid Vp is
Vp minuskPfQac
_Pf (11)
Considering the smoothing effect of hydraulic accu-mulator a more accurate model of hydraulic motor can bedetermined including the variations of accumulator fluidflow and volume +erefore the pressure and fluid flow ofhydraulic motor can be represented
_PM βM QAout + QBout minusQac minusQM( 1113857
VT
QM DMωM
2π
(12)
where βM is the effective bulk modulus of the motorchamber DM is the displacement of the hydraulic motorand ωM is the shaft speed of the hydraulic motor andgenerator respectively
23 Power Regeneration Driven by the fluid flow the hy-draulic motor can produce rotary motion which is able todrive the DC generator to produce electricity +e workingprocess of power regeneration is shown in Figure 4
Where the mechanical efficiency of motor is ηM thedriving torque of motor TM is
TM DMPMηM
2π (13)
According to Newtonrsquos second law of motion the rotarymotion ωm is
_ωM TM minusTG( 1113857
Jt (14)
where Jt is the shaft moment of inertia In addition theelectromagnetic torque of generator TG would change withthe variation in the induced current and it is considered as theresistance torque because it provides the rotation to themotorin an opposite direction +erefore it can be written as
TG KTI (15)
+e electromotive force (EMF) can be expressed as
E KEωM (16)
UpDown
QBout
QBin
QAout
QAin
VA
VP
AP
PA PMωM
QMAA
PB
ACPf Qac
Aac
PacAB
VB
v
l
M
ρ
Figure 3 Schematic view of fluid flows in hydraulic system
4 Shock and Vibration
where KT is torque constant coefficient and KE is the elec-tromotive voltage constant coefficient
According to Kirchhoffrsquos voltage law [25] assuming thatthe susceptibility at any temperature and the flux that isestablished by the PM poles are constant the variation incurrent can be calculated by
_I Eminus Rin + RB( 1113857I( 1113857
LG (17)
where LG is the internal inductance RB is the externalelectrical load of battery and Rin is the internal resistance ofgenerator +en the instantaneous voltage can be expressedas
U IRB (18)
+erefore the output of regenerated power is
Pout I2RB
U2
RB (19)
In addition the effective input power of this system isconsidered as the sum of the piston damping force multi-plied by the effective piston area Given the areas of cap-endand rod-end are known the piston damping force can bewritten as
FA PAAA
FB PBAB1113896 (20)
+erefore the input power of the HERSA system is
Pin PAAA|v(t)| + PBAB|v(t)| up + down (21)
Hence the power regeneration efficiency can be calcu-lated from the following equation
ηreg Pout
Pin (22)
3 Simulation of HERSA
+is study focuses on the effect of the key parameters whichinclude the sizes of shock absorber body the size of checkvalve port hydraulic motor displacement and hydraulicaccumulator capacity Hence the investigation is performedon the proposed parameter optimisation method to studythe HERSArsquos behaviour and power level as well as thedesirable parameter solutions
31 Parameters Setting and Study In the modelling a fewassumptions made during this process were as follows
(a) +ere are no additional electrical losses in generatorconfiguration to be considered It means the outputpower of hydraulic motor is equal to the input powerof generator
(b) +e cylinder internal leakage between the chambersis not accounted for by the model
(c) +e external electrical load (batteryresistor) has noinfluence by the varying temperature
(d) Fluid compressibility is defined as using Boesrsquosmodel in hydraulic system
In practical applications there are various types of lossesin HERSA system such as hydraulic motor internal flowleakage pressure loss in the pipeline check valve pressureloss and hydraulic cylinder piston friction
To simplify the model the impact of these losses andinfluences in the system are not taken into account at thisstage +erefore the model is ideally configured and thefollowing main parameters are selected
311 Size of the Hydraulic Cylinder In this system thehydraulic cylinder is applied to replace the traditionalshock absorber and absorb vibration energy +is papertakes 4x4 SUV ldquoBeijing Jeep 2021rdquo as the example anddetermines that the maximum stroke of cylinder is 200mm+e sizes of the shock absorber body are determinedaccording to the standard ISO 3320-2013 [26] it is wellknown that a conventional viscous shock absorber hasasymmetrical damping characteristic due to its inherentdesign structure which can provide different dampingforces during the compression and extension strokes +episton diameter and rod diameter are therefore determinedas Table 1
312 Size of the Check Valve Port +e fluid pressurised bythe oscillation flows through check valve arrangement toensure fluid always flows through hydraulic motor in onedirection and enable the chambers in the cylinder can bereplenished as fast as possible for each run In terms of thestandard port size on double-acting cylinder the commonsizes of check valves are shown in Table 2
313 Hydraulic Accumulator Capacity A common hy-draulic accumulator is used to minimise the fluctuation ofthe pressurised flow Initially the gas chamber is prechargedto pressure Pac and set to 20 bar
In equations (9)ndash(11) several assumptions have beenmade to simplify the hydraulic accumulator model
(a) +e accumulator is set as a diaphragm type accu-mulator without heat exchange during the process
(b) +ere is a transient pressure balance inside the ac-cumulator between fluid chamber and gas chamber
(c) Frictions and thermal losses are neglected here
M G
DMTG
RB
KT
KE
LG
Rin
TM
JtPM
QM
ωM
Figure 4 Schematic view of energy conversion (mechanical toelectrical) and power circuit
Shock and Vibration 5
(d) Only fully charged and fully discharged states areconsidered in the hydraulic model
(e) +e precharge pressure in the accumulator is set at60 of the working pressure (20 bar) to providepressure pulsation damping
According to the damping forces in a traditional shockabsorber in a SUV the accumulator capacity has been es-timated with a peak of 100 bar which is shown in Table 3
314 Hydraulic Motor Displacement +e hydraulic motoris defined as a transfer device which is able to convert theunidirectional hydraulic flowpressure into rotationalmotiontorque Its main parameters are shown in Table 4
315 External Electrical Load For the regenerative shockabsorber the electrical load has significant effects on thecapability of power regeneration and the dynamics of thesystem [1] and it can be considered as Table 5
316 Other Key Parameters As shown in Table 6 the valuesof several other parameters are displayed
32 Orthogonal Simulation Test Design +e orthogonal testdesign is an important approach of statistical evaluationwhich can reduce the number of attempts and then effec-tively obtain desired results [27] Large-scale engineeringtesting is a complex system engineering +ere are manyfactors influencing the test results some factors play anindependent role and some factors can interact with othersto produce the comprehensive effect +e reason that affectsthe results during the test is called the test factor and thestate adopted under each factor is called the level Startingfrom Section 31 five-factor and four-level conditions havebeen applied in this test as shown in Table 7 +e simplestmethod is to apply an exhaustive method to test all factorsand levels at the same time
As shown in Table 7 the designed test has four 4-levelfactor and one 2-level factor If all factors and their levelsachieve their best performance for integration testing thenthe total of 512 tests need to be performed Obviously a largenumber of tests definitely demand a lot of manpower andmaterial resources as well as take a long time to reach thetarget+erefore this paper focuses on the problem that howto reasonably arrange the test and obtain the necessaryindication through less number of tests +e orthogonalmethod is an effective mathematical method to solve theproblem of multifactor test It also applies the orthogonaltable design scheme and the mathematical statistics methodto analyse the test data However the L16 (44 times 2) orthogonaltable (Table 8) is designed for the simulation test of HERSA
[28] and only 16 tests need to be carried out It is obviousthat the proposed parameter optimisation method can ef-fectively improve the efficiency of simulation and fastlyobtain the required simulation results
According to the above test design table numericalsimulation is carried out in mathematical model A sinu-soidal excitation 1Hz-25mm is set as predefined input Allother parameters are kept the same as in Table 6
Table 5 BatteryResistor resistance
Factor E Level 1 Level 2 Level 3 Level 4Electrical load (Ω) 10 20 30 40
Table 1 Hydraulic cylinder specification
Factor A Level 1 Level 2 Level 3 Level 4Piston diameter (mm) 50 40 32 25Rod diameter (mm) 28 25 20 16
Table 2 Check valves specification
Factor B Level 1 Level 2 Level 3 Level 4Check valve diameter (mm) 635 9525 127 1905
Table 3 Hydraulic accumulator specification
Factor C Level 1 Level 2Accumulator capacity (L) 063 1
Table 4 Hydraulic motor displacement
Factor D Level 1 Level 2 Level 3 Level 4Motor displacement (cc) 577 707 801 894
Table 6 Other key specifications of the HERSA
Symbol Value Unitf 1 HzXa 25 mmρ 872 kgm3
CC 07 mdashX0 200 mml 1 mβref 12times109 mdashPref 20 bark 14 mdashηm 100 Jt 0003 kgmiddotm2
KT 065 mdashKE 065 mdashRin 75 ΩLG 003 HVagd 01 V LDac Dcv mmCac 07 mdash
Table 7 Simulation test combinations
LevelFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Level 1 5028 635 063 577 10Level 2 4025 9525 1 707 20Level 3 3220 127 mdash 801 30Level 4 2516 1905 mdash 894 40
6 Shock and Vibration
33 Simulation Results and Discussion To use the modelwith orthogonal test approach developed in Sections 31and 32 for subsequent studies on the improvement ofparameter optimisation this section firstly presents studieson the investigation of key model parameters based onmodelling analysis +en it shows the quantitative be-haviours (pressure shaft speed regenerated power andregeneration efficiency) of the system under designed testtable gaining a preliminary understanding of the systembehaviours However the key results are summarised inTable 9 According to the principle of orthogonal test theaverages of outputs are calculated in the mean responseanalysis
As shown in Figure 5 the variation of the motor shaftspeed pressure mechanical power and regeneratedpower is dramatically levelled up at larger cylinder sizewith lower motor displacement accumulator capacity thesize of check valve and external load resistance +eaverages of those results are shown in Table 9 It is alsoclear that the change of component combination cansignificantly affect the system dynamics and regeneratedpower level to meet the demands of various vehiclesuspension systems By changing the component com-binations the maximum regenerated power of 4175Wwith the regeneration efficiency of approximately 534can be achieved on test 1 It is also obvious that the highestregeneration efficiency occurs at test 16 (approximately63) which is designed with small size of cylinder motordisplacement accumulator capacity and large check valveport Additionally the results of those 16 tests reveal thatthe change of motor displacement external loads andaccumulator capacity is capable of smoothing the flowoscillations and allow effective minimisation to the in-stability of hydraulic circuit thus altering the perfor-mance of the system and power capability
It can be summarised that the change of componentcombinations can not only affect the stability of systemdynamics but also dramatically impact on the level ofregenerated power
4 Parameter Optimisation Analysis
16 sets of simulation tests have been designed and performedto provide desirable solutions of the system parameters toenhance the system behaviours and power level After thatmean response analysis is used to take the average value of alltest results for each factor level and determine the extent thatthe factor affects the indicator based on the range meanvalues which is called the comprehensive equilibriummethod It can be applied to perform the multi-indicatoranalysis of shaft speed hydraulic motor pressure re-generative power and regeneration efficiency
41 Shaft Speed As shown in Figure 6 the size of cylinderhas the greatest influence on shaft speed with a visibly largerange+e larger cylinder size can deliver faster flowwhich iscapable of providing larger rotational torque on motor shaft+e smaller motor displacement is also beneficial on themotor rotation which is more effective than other threefactors Compared to the size of check valves the electricalload has an equivalent degree of influence to the shaft speedIn addition it is obvious that the capacity of accumulator isless significant than other factors for shaft speed It can beconcluded that the optimal combination of the shaft speed asthe evaluation criterion is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
42 Hydraulic Motor Pressure As shown in Figure 7 thedisplacement of motor and the load of generator have almostequivalent influences on working pressure and they takeslightly lower impacts in comparison to the size of hydrauliccylinder Additionally it reveals that smaller check valvediameter can significantly lead to higher pressure+e higheraverage pressure means much more power can be producedby the generator to raise the regenerated power level+erefore optimal combination of hydraulic motor pressureis A1B1C1D1E1 where A1 5028mm B1 635mmC1 063 L D1 577 cc and E1 10Ω
43 Regenerated Power As shown in Figure 8 the size of thecylinder has significant influence on the power output of thepower circuit which is more important than other factorsObviously the small values of other factors can also con-tribute positive effects for more regenerated power From themean response analysis of Figure 8 the optimal combinationof the regenerated power is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
44 Regeneration Efficiency Regeneration efficiency is asignificant indicator to assess the performance of HERSA Asshown in Figure 9 the electrical load has the greatest in-fluence on the regeneration efficiency When the electricalload is 20Ω the proposed system can reach a high re-generation efficiency which increases slowly with the largervalue of electrical load Compared to other three indicators
Table 8 Orthogonal simulation test design list L16 (44 times 2)
NoFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Test 1 5028 635 063 577 10Test 2 5028 9525 063 707 20Test 3 5028 127 1 801 30Test 4 5028 1905 1 894 40Test 5 4025 635 1 707 30Test 6 4025 9525 1 577 40Test 7 4025 127 063 894 10Test 8 4025 1905 063 801 20Test 9 3220 635 063 801 40Test 10 3220 9525 063 894 30Test 11 3220 127 1 577 20Test 12 3220 1905 1 707 10Test 13 2516 635 1 894 20Test 14 2516 9525 1 801 10Test 15 2516 127 063 707 40Test 16 2516 1905 063 577 30
Shock and Vibration 7
higher regeneration efficiency can be found with larger thesize of check valve port and the capacity of accumulator It isclear that the larger size of hydraulic cylinder also con-tributes to regeneration efficiency However the optimalcombination of regeneration efficiency is A1B4C2D1E4where A1 5028mm B4 1905mm C2 1 L D1 577 ccand E4 40Ω
45 Optimisation Results From the mean response analysisof Figures 6ndash9 it can be concluded that the optimal com-bination of each parameter for more regenerated power onthe HERSA is A1B1C1D1E1 However a high regenerationefficiency is also one of the important criteria It is necessaryto adjust the optimal combination for the higher re-generation efficiency According to the range mean values ofdifferent factors the extent that the factor affects the in-dicator can be determined+e rank of factors based on theirdegree of importance and their corresponding best com-bination are shown in Table 10
It is obvious that the largest size of cylinder (A1) and thesmallest displacement of motor (D1) can provide betterdynamics and power level for the proposed HERSAAccording to the principle of the comprehensive equilib-rium method the selection of other factors is as follows
(a) Given the rank of factor B on different indicators isfourth the superior level selection is considered asthe one that occurs most frequently +e selection offactor B therefore is level 1 where B1 = 635mm
(b) Identically the rank of factor C is fifth on differentindicators Given the best level selection of factor Cfor three indicators is level 1 which is the one thatoccurs most frequently +erefore the selection offactor C is level 1 where C1 = 063 L
(c) When the factor has a different degree of influenceon all indicators the more important indicatorsshould be satisfied firstly +e regeneration efficiencyand regenerated power are the main criteria of thissystem When the factor E is selected the level 2 maybe better As shown in Figures 8 and 9 the efficiency
at level 2 is over 60 which is much higher than thatat level 1 and nearly to that at level 3 and 4Meanwhile the regenerated power at level 2 is alsohigh +e selection of factor E is level 2 whereE2 = 20Ω
+erefore the optimal combination of parameters inHERSA is A1B1C1D1E2 which is capable of delivering a largeelectrical power as well as a high regeneration efficiency +eoptimisation results are shown as follows +e size of cyl-inder is 50mm (piston) and 28mm (rod) the diameter ofcheck valve is 635mm the accumulator capacity is 063 Lthe displacement of motor is 577 cc and the electrical loadof generator is 20Ω respectively Furthermore for thegeneral HERSA system (including hydraulic cylinders checkvalves accumulators motors and generators) a set ofoptimisation methods about power regeneration perfor-mance can be summarised as follows
(a) Determine the range of key parameters(b) Design the corresponding orthogonal test(c) Conduct the mean response analysis(d) Discuss the optimisation results
It is considered that the proposed optimisation pro-cedure is suitable for the general HERSA system Figure 10shows the results of the HERSA applied with the optimalcomponent combinations
As shown in Figure 10 the shaft speed is 1634 rpm theaccumulator pressure is about 30 bar the power output isabout 331W and the regeneration efficiency is approxi-mately 65 Both larger power and higher regenerated ef-ficiency are achieved
46 Experimental Validation +e setup of the test rig isshown in Figure 11 According to the schematic in Figure 3the parameters of the key components are listed in Tables 6and 11 A corresponding test rig was designed and fabri-cated according to the design concept and model devel-opment to validate the prediction of the optimised HERSAmodel Based on the result of parameter optimisation the
Table 9 Summary of simulation tests
Test no Shaft speed (rpm) Hydraulic motor pressure (bar) Input power (W) Regenerated power (W) Regeneration efficiencyTest 1 16513928 464087 781801 4175 053402Test 2 13400569 198006 3427223 2239 065334Test 3 11756696 11384 2041776 1398 068482Test 4 1055314 7317 1371768 942 068667Test 5 8106328 88879 1015277 6643 065427Test 6 9885237 10621 1171773 8261 070496Test 7 6652814 11886 1295599 6665 050674Test 8 7275167 94879 659557 6596 062528Test 9 4612982 35676 300727 1798 059788Test 10 4107713 35638 296083 1706 057626Test 11 6481021 117337 818903 5231 063877Test 12 5315388 121918 847933 4318 050923Test 13 2516925 29404 154995 7888 05089Test 14 2842986 57556 263095 1235 046945Test 15 3120371 27366 145937 8228 056378Test 16 3888416 52267 242388 1528 063042
8 Shock and Vibration
key components of the test rig including hydraulic cyl-inder hydraulic accumulator hydraulic motor DC gen-erator inline check valve and hydraulic circuits of theHERSA system are carefully selected and equipped on thedesigned test bench +e HERSA test rig selected the pa-rameters of the main components that are closest to theoptimal results of the parameter optimisation Table 11shows the main parameter differences between model andtest rig
+e input controller and road actuator are designedwhich are able to provide predicted input excitations in-cluding excitation displacement excitation velocity and
frequency Data acquisition and transducers are also appliedto synchronise and measure the regenerated voltage andcurrent across the electrical load
+e voltage and current across the electrical load weremeasured on the design test rig +e measured regeneratedpower is compared with predicted power as shown inFigure 12 +e both results of simulation and measurementare under the harmonic excitation of 1Hz-25mm with anelectrical load of 20Ω It is worth to mention that themeasured average regenerated power is approximately326W compared to the predicted power of up to 331W andthe results show a good agreement especially for the visible
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65
0
200
400
600
800
1000
1200
1400
1600
1800Sh
a sp
eed
(rpm
)
t (s)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(a)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
10
Hyd
raul
ic m
otor
pre
ssur
e (ba
r)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(b)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
200
400
600
800
1000
1200
1400
1600
Mec
hani
cal p
ower
(W)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(c)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
100
200
300
400500
Rege
nera
ted
pow
er (W
)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(d)
Figure 5 Hydraulic motor shaft speed pressure mechanical power and regenerated power at different 16 tests
Shock and Vibration 9
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
where KT is torque constant coefficient and KE is the elec-tromotive voltage constant coefficient
According to Kirchhoffrsquos voltage law [25] assuming thatthe susceptibility at any temperature and the flux that isestablished by the PM poles are constant the variation incurrent can be calculated by
_I Eminus Rin + RB( 1113857I( 1113857
LG (17)
where LG is the internal inductance RB is the externalelectrical load of battery and Rin is the internal resistance ofgenerator +en the instantaneous voltage can be expressedas
U IRB (18)
+erefore the output of regenerated power is
Pout I2RB
U2
RB (19)
In addition the effective input power of this system isconsidered as the sum of the piston damping force multi-plied by the effective piston area Given the areas of cap-endand rod-end are known the piston damping force can bewritten as
FA PAAA
FB PBAB1113896 (20)
+erefore the input power of the HERSA system is
Pin PAAA|v(t)| + PBAB|v(t)| up + down (21)
Hence the power regeneration efficiency can be calcu-lated from the following equation
ηreg Pout
Pin (22)
3 Simulation of HERSA
+is study focuses on the effect of the key parameters whichinclude the sizes of shock absorber body the size of checkvalve port hydraulic motor displacement and hydraulicaccumulator capacity Hence the investigation is performedon the proposed parameter optimisation method to studythe HERSArsquos behaviour and power level as well as thedesirable parameter solutions
31 Parameters Setting and Study In the modelling a fewassumptions made during this process were as follows
(a) +ere are no additional electrical losses in generatorconfiguration to be considered It means the outputpower of hydraulic motor is equal to the input powerof generator
(b) +e cylinder internal leakage between the chambersis not accounted for by the model
(c) +e external electrical load (batteryresistor) has noinfluence by the varying temperature
(d) Fluid compressibility is defined as using Boesrsquosmodel in hydraulic system
In practical applications there are various types of lossesin HERSA system such as hydraulic motor internal flowleakage pressure loss in the pipeline check valve pressureloss and hydraulic cylinder piston friction
To simplify the model the impact of these losses andinfluences in the system are not taken into account at thisstage +erefore the model is ideally configured and thefollowing main parameters are selected
311 Size of the Hydraulic Cylinder In this system thehydraulic cylinder is applied to replace the traditionalshock absorber and absorb vibration energy +is papertakes 4x4 SUV ldquoBeijing Jeep 2021rdquo as the example anddetermines that the maximum stroke of cylinder is 200mm+e sizes of the shock absorber body are determinedaccording to the standard ISO 3320-2013 [26] it is wellknown that a conventional viscous shock absorber hasasymmetrical damping characteristic due to its inherentdesign structure which can provide different dampingforces during the compression and extension strokes +episton diameter and rod diameter are therefore determinedas Table 1
312 Size of the Check Valve Port +e fluid pressurised bythe oscillation flows through check valve arrangement toensure fluid always flows through hydraulic motor in onedirection and enable the chambers in the cylinder can bereplenished as fast as possible for each run In terms of thestandard port size on double-acting cylinder the commonsizes of check valves are shown in Table 2
313 Hydraulic Accumulator Capacity A common hy-draulic accumulator is used to minimise the fluctuation ofthe pressurised flow Initially the gas chamber is prechargedto pressure Pac and set to 20 bar
In equations (9)ndash(11) several assumptions have beenmade to simplify the hydraulic accumulator model
(a) +e accumulator is set as a diaphragm type accu-mulator without heat exchange during the process
(b) +ere is a transient pressure balance inside the ac-cumulator between fluid chamber and gas chamber
(c) Frictions and thermal losses are neglected here
M G
DMTG
RB
KT
KE
LG
Rin
TM
JtPM
QM
ωM
Figure 4 Schematic view of energy conversion (mechanical toelectrical) and power circuit
Shock and Vibration 5
(d) Only fully charged and fully discharged states areconsidered in the hydraulic model
(e) +e precharge pressure in the accumulator is set at60 of the working pressure (20 bar) to providepressure pulsation damping
According to the damping forces in a traditional shockabsorber in a SUV the accumulator capacity has been es-timated with a peak of 100 bar which is shown in Table 3
314 Hydraulic Motor Displacement +e hydraulic motoris defined as a transfer device which is able to convert theunidirectional hydraulic flowpressure into rotationalmotiontorque Its main parameters are shown in Table 4
315 External Electrical Load For the regenerative shockabsorber the electrical load has significant effects on thecapability of power regeneration and the dynamics of thesystem [1] and it can be considered as Table 5
316 Other Key Parameters As shown in Table 6 the valuesof several other parameters are displayed
32 Orthogonal Simulation Test Design +e orthogonal testdesign is an important approach of statistical evaluationwhich can reduce the number of attempts and then effec-tively obtain desired results [27] Large-scale engineeringtesting is a complex system engineering +ere are manyfactors influencing the test results some factors play anindependent role and some factors can interact with othersto produce the comprehensive effect +e reason that affectsthe results during the test is called the test factor and thestate adopted under each factor is called the level Startingfrom Section 31 five-factor and four-level conditions havebeen applied in this test as shown in Table 7 +e simplestmethod is to apply an exhaustive method to test all factorsand levels at the same time
As shown in Table 7 the designed test has four 4-levelfactor and one 2-level factor If all factors and their levelsachieve their best performance for integration testing thenthe total of 512 tests need to be performed Obviously a largenumber of tests definitely demand a lot of manpower andmaterial resources as well as take a long time to reach thetarget+erefore this paper focuses on the problem that howto reasonably arrange the test and obtain the necessaryindication through less number of tests +e orthogonalmethod is an effective mathematical method to solve theproblem of multifactor test It also applies the orthogonaltable design scheme and the mathematical statistics methodto analyse the test data However the L16 (44 times 2) orthogonaltable (Table 8) is designed for the simulation test of HERSA
[28] and only 16 tests need to be carried out It is obviousthat the proposed parameter optimisation method can ef-fectively improve the efficiency of simulation and fastlyobtain the required simulation results
According to the above test design table numericalsimulation is carried out in mathematical model A sinu-soidal excitation 1Hz-25mm is set as predefined input Allother parameters are kept the same as in Table 6
Table 5 BatteryResistor resistance
Factor E Level 1 Level 2 Level 3 Level 4Electrical load (Ω) 10 20 30 40
Table 1 Hydraulic cylinder specification
Factor A Level 1 Level 2 Level 3 Level 4Piston diameter (mm) 50 40 32 25Rod diameter (mm) 28 25 20 16
Table 2 Check valves specification
Factor B Level 1 Level 2 Level 3 Level 4Check valve diameter (mm) 635 9525 127 1905
Table 3 Hydraulic accumulator specification
Factor C Level 1 Level 2Accumulator capacity (L) 063 1
Table 4 Hydraulic motor displacement
Factor D Level 1 Level 2 Level 3 Level 4Motor displacement (cc) 577 707 801 894
Table 6 Other key specifications of the HERSA
Symbol Value Unitf 1 HzXa 25 mmρ 872 kgm3
CC 07 mdashX0 200 mml 1 mβref 12times109 mdashPref 20 bark 14 mdashηm 100 Jt 0003 kgmiddotm2
KT 065 mdashKE 065 mdashRin 75 ΩLG 003 HVagd 01 V LDac Dcv mmCac 07 mdash
Table 7 Simulation test combinations
LevelFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Level 1 5028 635 063 577 10Level 2 4025 9525 1 707 20Level 3 3220 127 mdash 801 30Level 4 2516 1905 mdash 894 40
6 Shock and Vibration
33 Simulation Results and Discussion To use the modelwith orthogonal test approach developed in Sections 31and 32 for subsequent studies on the improvement ofparameter optimisation this section firstly presents studieson the investigation of key model parameters based onmodelling analysis +en it shows the quantitative be-haviours (pressure shaft speed regenerated power andregeneration efficiency) of the system under designed testtable gaining a preliminary understanding of the systembehaviours However the key results are summarised inTable 9 According to the principle of orthogonal test theaverages of outputs are calculated in the mean responseanalysis
As shown in Figure 5 the variation of the motor shaftspeed pressure mechanical power and regeneratedpower is dramatically levelled up at larger cylinder sizewith lower motor displacement accumulator capacity thesize of check valve and external load resistance +eaverages of those results are shown in Table 9 It is alsoclear that the change of component combination cansignificantly affect the system dynamics and regeneratedpower level to meet the demands of various vehiclesuspension systems By changing the component com-binations the maximum regenerated power of 4175Wwith the regeneration efficiency of approximately 534can be achieved on test 1 It is also obvious that the highestregeneration efficiency occurs at test 16 (approximately63) which is designed with small size of cylinder motordisplacement accumulator capacity and large check valveport Additionally the results of those 16 tests reveal thatthe change of motor displacement external loads andaccumulator capacity is capable of smoothing the flowoscillations and allow effective minimisation to the in-stability of hydraulic circuit thus altering the perfor-mance of the system and power capability
It can be summarised that the change of componentcombinations can not only affect the stability of systemdynamics but also dramatically impact on the level ofregenerated power
4 Parameter Optimisation Analysis
16 sets of simulation tests have been designed and performedto provide desirable solutions of the system parameters toenhance the system behaviours and power level After thatmean response analysis is used to take the average value of alltest results for each factor level and determine the extent thatthe factor affects the indicator based on the range meanvalues which is called the comprehensive equilibriummethod It can be applied to perform the multi-indicatoranalysis of shaft speed hydraulic motor pressure re-generative power and regeneration efficiency
41 Shaft Speed As shown in Figure 6 the size of cylinderhas the greatest influence on shaft speed with a visibly largerange+e larger cylinder size can deliver faster flowwhich iscapable of providing larger rotational torque on motor shaft+e smaller motor displacement is also beneficial on themotor rotation which is more effective than other threefactors Compared to the size of check valves the electricalload has an equivalent degree of influence to the shaft speedIn addition it is obvious that the capacity of accumulator isless significant than other factors for shaft speed It can beconcluded that the optimal combination of the shaft speed asthe evaluation criterion is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
42 Hydraulic Motor Pressure As shown in Figure 7 thedisplacement of motor and the load of generator have almostequivalent influences on working pressure and they takeslightly lower impacts in comparison to the size of hydrauliccylinder Additionally it reveals that smaller check valvediameter can significantly lead to higher pressure+e higheraverage pressure means much more power can be producedby the generator to raise the regenerated power level+erefore optimal combination of hydraulic motor pressureis A1B1C1D1E1 where A1 5028mm B1 635mmC1 063 L D1 577 cc and E1 10Ω
43 Regenerated Power As shown in Figure 8 the size of thecylinder has significant influence on the power output of thepower circuit which is more important than other factorsObviously the small values of other factors can also con-tribute positive effects for more regenerated power From themean response analysis of Figure 8 the optimal combinationof the regenerated power is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
44 Regeneration Efficiency Regeneration efficiency is asignificant indicator to assess the performance of HERSA Asshown in Figure 9 the electrical load has the greatest in-fluence on the regeneration efficiency When the electricalload is 20Ω the proposed system can reach a high re-generation efficiency which increases slowly with the largervalue of electrical load Compared to other three indicators
Table 8 Orthogonal simulation test design list L16 (44 times 2)
NoFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Test 1 5028 635 063 577 10Test 2 5028 9525 063 707 20Test 3 5028 127 1 801 30Test 4 5028 1905 1 894 40Test 5 4025 635 1 707 30Test 6 4025 9525 1 577 40Test 7 4025 127 063 894 10Test 8 4025 1905 063 801 20Test 9 3220 635 063 801 40Test 10 3220 9525 063 894 30Test 11 3220 127 1 577 20Test 12 3220 1905 1 707 10Test 13 2516 635 1 894 20Test 14 2516 9525 1 801 10Test 15 2516 127 063 707 40Test 16 2516 1905 063 577 30
Shock and Vibration 7
higher regeneration efficiency can be found with larger thesize of check valve port and the capacity of accumulator It isclear that the larger size of hydraulic cylinder also con-tributes to regeneration efficiency However the optimalcombination of regeneration efficiency is A1B4C2D1E4where A1 5028mm B4 1905mm C2 1 L D1 577 ccand E4 40Ω
45 Optimisation Results From the mean response analysisof Figures 6ndash9 it can be concluded that the optimal com-bination of each parameter for more regenerated power onthe HERSA is A1B1C1D1E1 However a high regenerationefficiency is also one of the important criteria It is necessaryto adjust the optimal combination for the higher re-generation efficiency According to the range mean values ofdifferent factors the extent that the factor affects the in-dicator can be determined+e rank of factors based on theirdegree of importance and their corresponding best com-bination are shown in Table 10
It is obvious that the largest size of cylinder (A1) and thesmallest displacement of motor (D1) can provide betterdynamics and power level for the proposed HERSAAccording to the principle of the comprehensive equilib-rium method the selection of other factors is as follows
(a) Given the rank of factor B on different indicators isfourth the superior level selection is considered asthe one that occurs most frequently +e selection offactor B therefore is level 1 where B1 = 635mm
(b) Identically the rank of factor C is fifth on differentindicators Given the best level selection of factor Cfor three indicators is level 1 which is the one thatoccurs most frequently +erefore the selection offactor C is level 1 where C1 = 063 L
(c) When the factor has a different degree of influenceon all indicators the more important indicatorsshould be satisfied firstly +e regeneration efficiencyand regenerated power are the main criteria of thissystem When the factor E is selected the level 2 maybe better As shown in Figures 8 and 9 the efficiency
at level 2 is over 60 which is much higher than thatat level 1 and nearly to that at level 3 and 4Meanwhile the regenerated power at level 2 is alsohigh +e selection of factor E is level 2 whereE2 = 20Ω
+erefore the optimal combination of parameters inHERSA is A1B1C1D1E2 which is capable of delivering a largeelectrical power as well as a high regeneration efficiency +eoptimisation results are shown as follows +e size of cyl-inder is 50mm (piston) and 28mm (rod) the diameter ofcheck valve is 635mm the accumulator capacity is 063 Lthe displacement of motor is 577 cc and the electrical loadof generator is 20Ω respectively Furthermore for thegeneral HERSA system (including hydraulic cylinders checkvalves accumulators motors and generators) a set ofoptimisation methods about power regeneration perfor-mance can be summarised as follows
(a) Determine the range of key parameters(b) Design the corresponding orthogonal test(c) Conduct the mean response analysis(d) Discuss the optimisation results
It is considered that the proposed optimisation pro-cedure is suitable for the general HERSA system Figure 10shows the results of the HERSA applied with the optimalcomponent combinations
As shown in Figure 10 the shaft speed is 1634 rpm theaccumulator pressure is about 30 bar the power output isabout 331W and the regeneration efficiency is approxi-mately 65 Both larger power and higher regenerated ef-ficiency are achieved
46 Experimental Validation +e setup of the test rig isshown in Figure 11 According to the schematic in Figure 3the parameters of the key components are listed in Tables 6and 11 A corresponding test rig was designed and fabri-cated according to the design concept and model devel-opment to validate the prediction of the optimised HERSAmodel Based on the result of parameter optimisation the
Table 9 Summary of simulation tests
Test no Shaft speed (rpm) Hydraulic motor pressure (bar) Input power (W) Regenerated power (W) Regeneration efficiencyTest 1 16513928 464087 781801 4175 053402Test 2 13400569 198006 3427223 2239 065334Test 3 11756696 11384 2041776 1398 068482Test 4 1055314 7317 1371768 942 068667Test 5 8106328 88879 1015277 6643 065427Test 6 9885237 10621 1171773 8261 070496Test 7 6652814 11886 1295599 6665 050674Test 8 7275167 94879 659557 6596 062528Test 9 4612982 35676 300727 1798 059788Test 10 4107713 35638 296083 1706 057626Test 11 6481021 117337 818903 5231 063877Test 12 5315388 121918 847933 4318 050923Test 13 2516925 29404 154995 7888 05089Test 14 2842986 57556 263095 1235 046945Test 15 3120371 27366 145937 8228 056378Test 16 3888416 52267 242388 1528 063042
8 Shock and Vibration
key components of the test rig including hydraulic cyl-inder hydraulic accumulator hydraulic motor DC gen-erator inline check valve and hydraulic circuits of theHERSA system are carefully selected and equipped on thedesigned test bench +e HERSA test rig selected the pa-rameters of the main components that are closest to theoptimal results of the parameter optimisation Table 11shows the main parameter differences between model andtest rig
+e input controller and road actuator are designedwhich are able to provide predicted input excitations in-cluding excitation displacement excitation velocity and
frequency Data acquisition and transducers are also appliedto synchronise and measure the regenerated voltage andcurrent across the electrical load
+e voltage and current across the electrical load weremeasured on the design test rig +e measured regeneratedpower is compared with predicted power as shown inFigure 12 +e both results of simulation and measurementare under the harmonic excitation of 1Hz-25mm with anelectrical load of 20Ω It is worth to mention that themeasured average regenerated power is approximately326W compared to the predicted power of up to 331W andthe results show a good agreement especially for the visible
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65
0
200
400
600
800
1000
1200
1400
1600
1800Sh
a sp
eed
(rpm
)
t (s)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(a)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
10
Hyd
raul
ic m
otor
pre
ssur
e (ba
r)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(b)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
200
400
600
800
1000
1200
1400
1600
Mec
hani
cal p
ower
(W)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(c)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
100
200
300
400500
Rege
nera
ted
pow
er (W
)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(d)
Figure 5 Hydraulic motor shaft speed pressure mechanical power and regenerated power at different 16 tests
Shock and Vibration 9
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
(d) Only fully charged and fully discharged states areconsidered in the hydraulic model
(e) +e precharge pressure in the accumulator is set at60 of the working pressure (20 bar) to providepressure pulsation damping
According to the damping forces in a traditional shockabsorber in a SUV the accumulator capacity has been es-timated with a peak of 100 bar which is shown in Table 3
314 Hydraulic Motor Displacement +e hydraulic motoris defined as a transfer device which is able to convert theunidirectional hydraulic flowpressure into rotationalmotiontorque Its main parameters are shown in Table 4
315 External Electrical Load For the regenerative shockabsorber the electrical load has significant effects on thecapability of power regeneration and the dynamics of thesystem [1] and it can be considered as Table 5
316 Other Key Parameters As shown in Table 6 the valuesof several other parameters are displayed
32 Orthogonal Simulation Test Design +e orthogonal testdesign is an important approach of statistical evaluationwhich can reduce the number of attempts and then effec-tively obtain desired results [27] Large-scale engineeringtesting is a complex system engineering +ere are manyfactors influencing the test results some factors play anindependent role and some factors can interact with othersto produce the comprehensive effect +e reason that affectsthe results during the test is called the test factor and thestate adopted under each factor is called the level Startingfrom Section 31 five-factor and four-level conditions havebeen applied in this test as shown in Table 7 +e simplestmethod is to apply an exhaustive method to test all factorsand levels at the same time
As shown in Table 7 the designed test has four 4-levelfactor and one 2-level factor If all factors and their levelsachieve their best performance for integration testing thenthe total of 512 tests need to be performed Obviously a largenumber of tests definitely demand a lot of manpower andmaterial resources as well as take a long time to reach thetarget+erefore this paper focuses on the problem that howto reasonably arrange the test and obtain the necessaryindication through less number of tests +e orthogonalmethod is an effective mathematical method to solve theproblem of multifactor test It also applies the orthogonaltable design scheme and the mathematical statistics methodto analyse the test data However the L16 (44 times 2) orthogonaltable (Table 8) is designed for the simulation test of HERSA
[28] and only 16 tests need to be carried out It is obviousthat the proposed parameter optimisation method can ef-fectively improve the efficiency of simulation and fastlyobtain the required simulation results
According to the above test design table numericalsimulation is carried out in mathematical model A sinu-soidal excitation 1Hz-25mm is set as predefined input Allother parameters are kept the same as in Table 6
Table 5 BatteryResistor resistance
Factor E Level 1 Level 2 Level 3 Level 4Electrical load (Ω) 10 20 30 40
Table 1 Hydraulic cylinder specification
Factor A Level 1 Level 2 Level 3 Level 4Piston diameter (mm) 50 40 32 25Rod diameter (mm) 28 25 20 16
Table 2 Check valves specification
Factor B Level 1 Level 2 Level 3 Level 4Check valve diameter (mm) 635 9525 127 1905
Table 3 Hydraulic accumulator specification
Factor C Level 1 Level 2Accumulator capacity (L) 063 1
Table 4 Hydraulic motor displacement
Factor D Level 1 Level 2 Level 3 Level 4Motor displacement (cc) 577 707 801 894
Table 6 Other key specifications of the HERSA
Symbol Value Unitf 1 HzXa 25 mmρ 872 kgm3
CC 07 mdashX0 200 mml 1 mβref 12times109 mdashPref 20 bark 14 mdashηm 100 Jt 0003 kgmiddotm2
KT 065 mdashKE 065 mdashRin 75 ΩLG 003 HVagd 01 V LDac Dcv mmCac 07 mdash
Table 7 Simulation test combinations
LevelFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Level 1 5028 635 063 577 10Level 2 4025 9525 1 707 20Level 3 3220 127 mdash 801 30Level 4 2516 1905 mdash 894 40
6 Shock and Vibration
33 Simulation Results and Discussion To use the modelwith orthogonal test approach developed in Sections 31and 32 for subsequent studies on the improvement ofparameter optimisation this section firstly presents studieson the investigation of key model parameters based onmodelling analysis +en it shows the quantitative be-haviours (pressure shaft speed regenerated power andregeneration efficiency) of the system under designed testtable gaining a preliminary understanding of the systembehaviours However the key results are summarised inTable 9 According to the principle of orthogonal test theaverages of outputs are calculated in the mean responseanalysis
As shown in Figure 5 the variation of the motor shaftspeed pressure mechanical power and regeneratedpower is dramatically levelled up at larger cylinder sizewith lower motor displacement accumulator capacity thesize of check valve and external load resistance +eaverages of those results are shown in Table 9 It is alsoclear that the change of component combination cansignificantly affect the system dynamics and regeneratedpower level to meet the demands of various vehiclesuspension systems By changing the component com-binations the maximum regenerated power of 4175Wwith the regeneration efficiency of approximately 534can be achieved on test 1 It is also obvious that the highestregeneration efficiency occurs at test 16 (approximately63) which is designed with small size of cylinder motordisplacement accumulator capacity and large check valveport Additionally the results of those 16 tests reveal thatthe change of motor displacement external loads andaccumulator capacity is capable of smoothing the flowoscillations and allow effective minimisation to the in-stability of hydraulic circuit thus altering the perfor-mance of the system and power capability
It can be summarised that the change of componentcombinations can not only affect the stability of systemdynamics but also dramatically impact on the level ofregenerated power
4 Parameter Optimisation Analysis
16 sets of simulation tests have been designed and performedto provide desirable solutions of the system parameters toenhance the system behaviours and power level After thatmean response analysis is used to take the average value of alltest results for each factor level and determine the extent thatthe factor affects the indicator based on the range meanvalues which is called the comprehensive equilibriummethod It can be applied to perform the multi-indicatoranalysis of shaft speed hydraulic motor pressure re-generative power and regeneration efficiency
41 Shaft Speed As shown in Figure 6 the size of cylinderhas the greatest influence on shaft speed with a visibly largerange+e larger cylinder size can deliver faster flowwhich iscapable of providing larger rotational torque on motor shaft+e smaller motor displacement is also beneficial on themotor rotation which is more effective than other threefactors Compared to the size of check valves the electricalload has an equivalent degree of influence to the shaft speedIn addition it is obvious that the capacity of accumulator isless significant than other factors for shaft speed It can beconcluded that the optimal combination of the shaft speed asthe evaluation criterion is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
42 Hydraulic Motor Pressure As shown in Figure 7 thedisplacement of motor and the load of generator have almostequivalent influences on working pressure and they takeslightly lower impacts in comparison to the size of hydrauliccylinder Additionally it reveals that smaller check valvediameter can significantly lead to higher pressure+e higheraverage pressure means much more power can be producedby the generator to raise the regenerated power level+erefore optimal combination of hydraulic motor pressureis A1B1C1D1E1 where A1 5028mm B1 635mmC1 063 L D1 577 cc and E1 10Ω
43 Regenerated Power As shown in Figure 8 the size of thecylinder has significant influence on the power output of thepower circuit which is more important than other factorsObviously the small values of other factors can also con-tribute positive effects for more regenerated power From themean response analysis of Figure 8 the optimal combinationof the regenerated power is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
44 Regeneration Efficiency Regeneration efficiency is asignificant indicator to assess the performance of HERSA Asshown in Figure 9 the electrical load has the greatest in-fluence on the regeneration efficiency When the electricalload is 20Ω the proposed system can reach a high re-generation efficiency which increases slowly with the largervalue of electrical load Compared to other three indicators
Table 8 Orthogonal simulation test design list L16 (44 times 2)
NoFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Test 1 5028 635 063 577 10Test 2 5028 9525 063 707 20Test 3 5028 127 1 801 30Test 4 5028 1905 1 894 40Test 5 4025 635 1 707 30Test 6 4025 9525 1 577 40Test 7 4025 127 063 894 10Test 8 4025 1905 063 801 20Test 9 3220 635 063 801 40Test 10 3220 9525 063 894 30Test 11 3220 127 1 577 20Test 12 3220 1905 1 707 10Test 13 2516 635 1 894 20Test 14 2516 9525 1 801 10Test 15 2516 127 063 707 40Test 16 2516 1905 063 577 30
Shock and Vibration 7
higher regeneration efficiency can be found with larger thesize of check valve port and the capacity of accumulator It isclear that the larger size of hydraulic cylinder also con-tributes to regeneration efficiency However the optimalcombination of regeneration efficiency is A1B4C2D1E4where A1 5028mm B4 1905mm C2 1 L D1 577 ccand E4 40Ω
45 Optimisation Results From the mean response analysisof Figures 6ndash9 it can be concluded that the optimal com-bination of each parameter for more regenerated power onthe HERSA is A1B1C1D1E1 However a high regenerationefficiency is also one of the important criteria It is necessaryto adjust the optimal combination for the higher re-generation efficiency According to the range mean values ofdifferent factors the extent that the factor affects the in-dicator can be determined+e rank of factors based on theirdegree of importance and their corresponding best com-bination are shown in Table 10
It is obvious that the largest size of cylinder (A1) and thesmallest displacement of motor (D1) can provide betterdynamics and power level for the proposed HERSAAccording to the principle of the comprehensive equilib-rium method the selection of other factors is as follows
(a) Given the rank of factor B on different indicators isfourth the superior level selection is considered asthe one that occurs most frequently +e selection offactor B therefore is level 1 where B1 = 635mm
(b) Identically the rank of factor C is fifth on differentindicators Given the best level selection of factor Cfor three indicators is level 1 which is the one thatoccurs most frequently +erefore the selection offactor C is level 1 where C1 = 063 L
(c) When the factor has a different degree of influenceon all indicators the more important indicatorsshould be satisfied firstly +e regeneration efficiencyand regenerated power are the main criteria of thissystem When the factor E is selected the level 2 maybe better As shown in Figures 8 and 9 the efficiency
at level 2 is over 60 which is much higher than thatat level 1 and nearly to that at level 3 and 4Meanwhile the regenerated power at level 2 is alsohigh +e selection of factor E is level 2 whereE2 = 20Ω
+erefore the optimal combination of parameters inHERSA is A1B1C1D1E2 which is capable of delivering a largeelectrical power as well as a high regeneration efficiency +eoptimisation results are shown as follows +e size of cyl-inder is 50mm (piston) and 28mm (rod) the diameter ofcheck valve is 635mm the accumulator capacity is 063 Lthe displacement of motor is 577 cc and the electrical loadof generator is 20Ω respectively Furthermore for thegeneral HERSA system (including hydraulic cylinders checkvalves accumulators motors and generators) a set ofoptimisation methods about power regeneration perfor-mance can be summarised as follows
(a) Determine the range of key parameters(b) Design the corresponding orthogonal test(c) Conduct the mean response analysis(d) Discuss the optimisation results
It is considered that the proposed optimisation pro-cedure is suitable for the general HERSA system Figure 10shows the results of the HERSA applied with the optimalcomponent combinations
As shown in Figure 10 the shaft speed is 1634 rpm theaccumulator pressure is about 30 bar the power output isabout 331W and the regeneration efficiency is approxi-mately 65 Both larger power and higher regenerated ef-ficiency are achieved
46 Experimental Validation +e setup of the test rig isshown in Figure 11 According to the schematic in Figure 3the parameters of the key components are listed in Tables 6and 11 A corresponding test rig was designed and fabri-cated according to the design concept and model devel-opment to validate the prediction of the optimised HERSAmodel Based on the result of parameter optimisation the
Table 9 Summary of simulation tests
Test no Shaft speed (rpm) Hydraulic motor pressure (bar) Input power (W) Regenerated power (W) Regeneration efficiencyTest 1 16513928 464087 781801 4175 053402Test 2 13400569 198006 3427223 2239 065334Test 3 11756696 11384 2041776 1398 068482Test 4 1055314 7317 1371768 942 068667Test 5 8106328 88879 1015277 6643 065427Test 6 9885237 10621 1171773 8261 070496Test 7 6652814 11886 1295599 6665 050674Test 8 7275167 94879 659557 6596 062528Test 9 4612982 35676 300727 1798 059788Test 10 4107713 35638 296083 1706 057626Test 11 6481021 117337 818903 5231 063877Test 12 5315388 121918 847933 4318 050923Test 13 2516925 29404 154995 7888 05089Test 14 2842986 57556 263095 1235 046945Test 15 3120371 27366 145937 8228 056378Test 16 3888416 52267 242388 1528 063042
8 Shock and Vibration
key components of the test rig including hydraulic cyl-inder hydraulic accumulator hydraulic motor DC gen-erator inline check valve and hydraulic circuits of theHERSA system are carefully selected and equipped on thedesigned test bench +e HERSA test rig selected the pa-rameters of the main components that are closest to theoptimal results of the parameter optimisation Table 11shows the main parameter differences between model andtest rig
+e input controller and road actuator are designedwhich are able to provide predicted input excitations in-cluding excitation displacement excitation velocity and
frequency Data acquisition and transducers are also appliedto synchronise and measure the regenerated voltage andcurrent across the electrical load
+e voltage and current across the electrical load weremeasured on the design test rig +e measured regeneratedpower is compared with predicted power as shown inFigure 12 +e both results of simulation and measurementare under the harmonic excitation of 1Hz-25mm with anelectrical load of 20Ω It is worth to mention that themeasured average regenerated power is approximately326W compared to the predicted power of up to 331W andthe results show a good agreement especially for the visible
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65
0
200
400
600
800
1000
1200
1400
1600
1800Sh
a sp
eed
(rpm
)
t (s)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(a)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
10
Hyd
raul
ic m
otor
pre
ssur
e (ba
r)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(b)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
200
400
600
800
1000
1200
1400
1600
Mec
hani
cal p
ower
(W)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(c)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
100
200
300
400500
Rege
nera
ted
pow
er (W
)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(d)
Figure 5 Hydraulic motor shaft speed pressure mechanical power and regenerated power at different 16 tests
Shock and Vibration 9
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
33 Simulation Results and Discussion To use the modelwith orthogonal test approach developed in Sections 31and 32 for subsequent studies on the improvement ofparameter optimisation this section firstly presents studieson the investigation of key model parameters based onmodelling analysis +en it shows the quantitative be-haviours (pressure shaft speed regenerated power andregeneration efficiency) of the system under designed testtable gaining a preliminary understanding of the systembehaviours However the key results are summarised inTable 9 According to the principle of orthogonal test theaverages of outputs are calculated in the mean responseanalysis
As shown in Figure 5 the variation of the motor shaftspeed pressure mechanical power and regeneratedpower is dramatically levelled up at larger cylinder sizewith lower motor displacement accumulator capacity thesize of check valve and external load resistance +eaverages of those results are shown in Table 9 It is alsoclear that the change of component combination cansignificantly affect the system dynamics and regeneratedpower level to meet the demands of various vehiclesuspension systems By changing the component com-binations the maximum regenerated power of 4175Wwith the regeneration efficiency of approximately 534can be achieved on test 1 It is also obvious that the highestregeneration efficiency occurs at test 16 (approximately63) which is designed with small size of cylinder motordisplacement accumulator capacity and large check valveport Additionally the results of those 16 tests reveal thatthe change of motor displacement external loads andaccumulator capacity is capable of smoothing the flowoscillations and allow effective minimisation to the in-stability of hydraulic circuit thus altering the perfor-mance of the system and power capability
It can be summarised that the change of componentcombinations can not only affect the stability of systemdynamics but also dramatically impact on the level ofregenerated power
4 Parameter Optimisation Analysis
16 sets of simulation tests have been designed and performedto provide desirable solutions of the system parameters toenhance the system behaviours and power level After thatmean response analysis is used to take the average value of alltest results for each factor level and determine the extent thatthe factor affects the indicator based on the range meanvalues which is called the comprehensive equilibriummethod It can be applied to perform the multi-indicatoranalysis of shaft speed hydraulic motor pressure re-generative power and regeneration efficiency
41 Shaft Speed As shown in Figure 6 the size of cylinderhas the greatest influence on shaft speed with a visibly largerange+e larger cylinder size can deliver faster flowwhich iscapable of providing larger rotational torque on motor shaft+e smaller motor displacement is also beneficial on themotor rotation which is more effective than other threefactors Compared to the size of check valves the electricalload has an equivalent degree of influence to the shaft speedIn addition it is obvious that the capacity of accumulator isless significant than other factors for shaft speed It can beconcluded that the optimal combination of the shaft speed asthe evaluation criterion is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
42 Hydraulic Motor Pressure As shown in Figure 7 thedisplacement of motor and the load of generator have almostequivalent influences on working pressure and they takeslightly lower impacts in comparison to the size of hydrauliccylinder Additionally it reveals that smaller check valvediameter can significantly lead to higher pressure+e higheraverage pressure means much more power can be producedby the generator to raise the regenerated power level+erefore optimal combination of hydraulic motor pressureis A1B1C1D1E1 where A1 5028mm B1 635mmC1 063 L D1 577 cc and E1 10Ω
43 Regenerated Power As shown in Figure 8 the size of thecylinder has significant influence on the power output of thepower circuit which is more important than other factorsObviously the small values of other factors can also con-tribute positive effects for more regenerated power From themean response analysis of Figure 8 the optimal combinationof the regenerated power is A1B1C1D1E1 where A1 5028mm B1 635mm C1 063 L D1 577 cc andE1 10Ω
44 Regeneration Efficiency Regeneration efficiency is asignificant indicator to assess the performance of HERSA Asshown in Figure 9 the electrical load has the greatest in-fluence on the regeneration efficiency When the electricalload is 20Ω the proposed system can reach a high re-generation efficiency which increases slowly with the largervalue of electrical load Compared to other three indicators
Table 8 Orthogonal simulation test design list L16 (44 times 2)
NoFactor
A (mm) B (mm) C (L) D (cc) E (Ω)Test 1 5028 635 063 577 10Test 2 5028 9525 063 707 20Test 3 5028 127 1 801 30Test 4 5028 1905 1 894 40Test 5 4025 635 1 707 30Test 6 4025 9525 1 577 40Test 7 4025 127 063 894 10Test 8 4025 1905 063 801 20Test 9 3220 635 063 801 40Test 10 3220 9525 063 894 30Test 11 3220 127 1 577 20Test 12 3220 1905 1 707 10Test 13 2516 635 1 894 20Test 14 2516 9525 1 801 10Test 15 2516 127 063 707 40Test 16 2516 1905 063 577 30
Shock and Vibration 7
higher regeneration efficiency can be found with larger thesize of check valve port and the capacity of accumulator It isclear that the larger size of hydraulic cylinder also con-tributes to regeneration efficiency However the optimalcombination of regeneration efficiency is A1B4C2D1E4where A1 5028mm B4 1905mm C2 1 L D1 577 ccand E4 40Ω
45 Optimisation Results From the mean response analysisof Figures 6ndash9 it can be concluded that the optimal com-bination of each parameter for more regenerated power onthe HERSA is A1B1C1D1E1 However a high regenerationefficiency is also one of the important criteria It is necessaryto adjust the optimal combination for the higher re-generation efficiency According to the range mean values ofdifferent factors the extent that the factor affects the in-dicator can be determined+e rank of factors based on theirdegree of importance and their corresponding best com-bination are shown in Table 10
It is obvious that the largest size of cylinder (A1) and thesmallest displacement of motor (D1) can provide betterdynamics and power level for the proposed HERSAAccording to the principle of the comprehensive equilib-rium method the selection of other factors is as follows
(a) Given the rank of factor B on different indicators isfourth the superior level selection is considered asthe one that occurs most frequently +e selection offactor B therefore is level 1 where B1 = 635mm
(b) Identically the rank of factor C is fifth on differentindicators Given the best level selection of factor Cfor three indicators is level 1 which is the one thatoccurs most frequently +erefore the selection offactor C is level 1 where C1 = 063 L
(c) When the factor has a different degree of influenceon all indicators the more important indicatorsshould be satisfied firstly +e regeneration efficiencyand regenerated power are the main criteria of thissystem When the factor E is selected the level 2 maybe better As shown in Figures 8 and 9 the efficiency
at level 2 is over 60 which is much higher than thatat level 1 and nearly to that at level 3 and 4Meanwhile the regenerated power at level 2 is alsohigh +e selection of factor E is level 2 whereE2 = 20Ω
+erefore the optimal combination of parameters inHERSA is A1B1C1D1E2 which is capable of delivering a largeelectrical power as well as a high regeneration efficiency +eoptimisation results are shown as follows +e size of cyl-inder is 50mm (piston) and 28mm (rod) the diameter ofcheck valve is 635mm the accumulator capacity is 063 Lthe displacement of motor is 577 cc and the electrical loadof generator is 20Ω respectively Furthermore for thegeneral HERSA system (including hydraulic cylinders checkvalves accumulators motors and generators) a set ofoptimisation methods about power regeneration perfor-mance can be summarised as follows
(a) Determine the range of key parameters(b) Design the corresponding orthogonal test(c) Conduct the mean response analysis(d) Discuss the optimisation results
It is considered that the proposed optimisation pro-cedure is suitable for the general HERSA system Figure 10shows the results of the HERSA applied with the optimalcomponent combinations
As shown in Figure 10 the shaft speed is 1634 rpm theaccumulator pressure is about 30 bar the power output isabout 331W and the regeneration efficiency is approxi-mately 65 Both larger power and higher regenerated ef-ficiency are achieved
46 Experimental Validation +e setup of the test rig isshown in Figure 11 According to the schematic in Figure 3the parameters of the key components are listed in Tables 6and 11 A corresponding test rig was designed and fabri-cated according to the design concept and model devel-opment to validate the prediction of the optimised HERSAmodel Based on the result of parameter optimisation the
Table 9 Summary of simulation tests
Test no Shaft speed (rpm) Hydraulic motor pressure (bar) Input power (W) Regenerated power (W) Regeneration efficiencyTest 1 16513928 464087 781801 4175 053402Test 2 13400569 198006 3427223 2239 065334Test 3 11756696 11384 2041776 1398 068482Test 4 1055314 7317 1371768 942 068667Test 5 8106328 88879 1015277 6643 065427Test 6 9885237 10621 1171773 8261 070496Test 7 6652814 11886 1295599 6665 050674Test 8 7275167 94879 659557 6596 062528Test 9 4612982 35676 300727 1798 059788Test 10 4107713 35638 296083 1706 057626Test 11 6481021 117337 818903 5231 063877Test 12 5315388 121918 847933 4318 050923Test 13 2516925 29404 154995 7888 05089Test 14 2842986 57556 263095 1235 046945Test 15 3120371 27366 145937 8228 056378Test 16 3888416 52267 242388 1528 063042
8 Shock and Vibration
key components of the test rig including hydraulic cyl-inder hydraulic accumulator hydraulic motor DC gen-erator inline check valve and hydraulic circuits of theHERSA system are carefully selected and equipped on thedesigned test bench +e HERSA test rig selected the pa-rameters of the main components that are closest to theoptimal results of the parameter optimisation Table 11shows the main parameter differences between model andtest rig
+e input controller and road actuator are designedwhich are able to provide predicted input excitations in-cluding excitation displacement excitation velocity and
frequency Data acquisition and transducers are also appliedto synchronise and measure the regenerated voltage andcurrent across the electrical load
+e voltage and current across the electrical load weremeasured on the design test rig +e measured regeneratedpower is compared with predicted power as shown inFigure 12 +e both results of simulation and measurementare under the harmonic excitation of 1Hz-25mm with anelectrical load of 20Ω It is worth to mention that themeasured average regenerated power is approximately326W compared to the predicted power of up to 331W andthe results show a good agreement especially for the visible
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65
0
200
400
600
800
1000
1200
1400
1600
1800Sh
a sp
eed
(rpm
)
t (s)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(a)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
10
Hyd
raul
ic m
otor
pre
ssur
e (ba
r)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(b)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
200
400
600
800
1000
1200
1400
1600
Mec
hani
cal p
ower
(W)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(c)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
100
200
300
400500
Rege
nera
ted
pow
er (W
)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(d)
Figure 5 Hydraulic motor shaft speed pressure mechanical power and regenerated power at different 16 tests
Shock and Vibration 9
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
higher regeneration efficiency can be found with larger thesize of check valve port and the capacity of accumulator It isclear that the larger size of hydraulic cylinder also con-tributes to regeneration efficiency However the optimalcombination of regeneration efficiency is A1B4C2D1E4where A1 5028mm B4 1905mm C2 1 L D1 577 ccand E4 40Ω
45 Optimisation Results From the mean response analysisof Figures 6ndash9 it can be concluded that the optimal com-bination of each parameter for more regenerated power onthe HERSA is A1B1C1D1E1 However a high regenerationefficiency is also one of the important criteria It is necessaryto adjust the optimal combination for the higher re-generation efficiency According to the range mean values ofdifferent factors the extent that the factor affects the in-dicator can be determined+e rank of factors based on theirdegree of importance and their corresponding best com-bination are shown in Table 10
It is obvious that the largest size of cylinder (A1) and thesmallest displacement of motor (D1) can provide betterdynamics and power level for the proposed HERSAAccording to the principle of the comprehensive equilib-rium method the selection of other factors is as follows
(a) Given the rank of factor B on different indicators isfourth the superior level selection is considered asthe one that occurs most frequently +e selection offactor B therefore is level 1 where B1 = 635mm
(b) Identically the rank of factor C is fifth on differentindicators Given the best level selection of factor Cfor three indicators is level 1 which is the one thatoccurs most frequently +erefore the selection offactor C is level 1 where C1 = 063 L
(c) When the factor has a different degree of influenceon all indicators the more important indicatorsshould be satisfied firstly +e regeneration efficiencyand regenerated power are the main criteria of thissystem When the factor E is selected the level 2 maybe better As shown in Figures 8 and 9 the efficiency
at level 2 is over 60 which is much higher than thatat level 1 and nearly to that at level 3 and 4Meanwhile the regenerated power at level 2 is alsohigh +e selection of factor E is level 2 whereE2 = 20Ω
+erefore the optimal combination of parameters inHERSA is A1B1C1D1E2 which is capable of delivering a largeelectrical power as well as a high regeneration efficiency +eoptimisation results are shown as follows +e size of cyl-inder is 50mm (piston) and 28mm (rod) the diameter ofcheck valve is 635mm the accumulator capacity is 063 Lthe displacement of motor is 577 cc and the electrical loadof generator is 20Ω respectively Furthermore for thegeneral HERSA system (including hydraulic cylinders checkvalves accumulators motors and generators) a set ofoptimisation methods about power regeneration perfor-mance can be summarised as follows
(a) Determine the range of key parameters(b) Design the corresponding orthogonal test(c) Conduct the mean response analysis(d) Discuss the optimisation results
It is considered that the proposed optimisation pro-cedure is suitable for the general HERSA system Figure 10shows the results of the HERSA applied with the optimalcomponent combinations
As shown in Figure 10 the shaft speed is 1634 rpm theaccumulator pressure is about 30 bar the power output isabout 331W and the regeneration efficiency is approxi-mately 65 Both larger power and higher regenerated ef-ficiency are achieved
46 Experimental Validation +e setup of the test rig isshown in Figure 11 According to the schematic in Figure 3the parameters of the key components are listed in Tables 6and 11 A corresponding test rig was designed and fabri-cated according to the design concept and model devel-opment to validate the prediction of the optimised HERSAmodel Based on the result of parameter optimisation the
Table 9 Summary of simulation tests
Test no Shaft speed (rpm) Hydraulic motor pressure (bar) Input power (W) Regenerated power (W) Regeneration efficiencyTest 1 16513928 464087 781801 4175 053402Test 2 13400569 198006 3427223 2239 065334Test 3 11756696 11384 2041776 1398 068482Test 4 1055314 7317 1371768 942 068667Test 5 8106328 88879 1015277 6643 065427Test 6 9885237 10621 1171773 8261 070496Test 7 6652814 11886 1295599 6665 050674Test 8 7275167 94879 659557 6596 062528Test 9 4612982 35676 300727 1798 059788Test 10 4107713 35638 296083 1706 057626Test 11 6481021 117337 818903 5231 063877Test 12 5315388 121918 847933 4318 050923Test 13 2516925 29404 154995 7888 05089Test 14 2842986 57556 263095 1235 046945Test 15 3120371 27366 145937 8228 056378Test 16 3888416 52267 242388 1528 063042
8 Shock and Vibration
key components of the test rig including hydraulic cyl-inder hydraulic accumulator hydraulic motor DC gen-erator inline check valve and hydraulic circuits of theHERSA system are carefully selected and equipped on thedesigned test bench +e HERSA test rig selected the pa-rameters of the main components that are closest to theoptimal results of the parameter optimisation Table 11shows the main parameter differences between model andtest rig
+e input controller and road actuator are designedwhich are able to provide predicted input excitations in-cluding excitation displacement excitation velocity and
frequency Data acquisition and transducers are also appliedto synchronise and measure the regenerated voltage andcurrent across the electrical load
+e voltage and current across the electrical load weremeasured on the design test rig +e measured regeneratedpower is compared with predicted power as shown inFigure 12 +e both results of simulation and measurementare under the harmonic excitation of 1Hz-25mm with anelectrical load of 20Ω It is worth to mention that themeasured average regenerated power is approximately326W compared to the predicted power of up to 331W andthe results show a good agreement especially for the visible
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65
0
200
400
600
800
1000
1200
1400
1600
1800Sh
a sp
eed
(rpm
)
t (s)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(a)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
10
Hyd
raul
ic m
otor
pre
ssur
e (ba
r)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(b)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
200
400
600
800
1000
1200
1400
1600
Mec
hani
cal p
ower
(W)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(c)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
100
200
300
400500
Rege
nera
ted
pow
er (W
)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(d)
Figure 5 Hydraulic motor shaft speed pressure mechanical power and regenerated power at different 16 tests
Shock and Vibration 9
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
key components of the test rig including hydraulic cyl-inder hydraulic accumulator hydraulic motor DC gen-erator inline check valve and hydraulic circuits of theHERSA system are carefully selected and equipped on thedesigned test bench +e HERSA test rig selected the pa-rameters of the main components that are closest to theoptimal results of the parameter optimisation Table 11shows the main parameter differences between model andtest rig
+e input controller and road actuator are designedwhich are able to provide predicted input excitations in-cluding excitation displacement excitation velocity and
frequency Data acquisition and transducers are also appliedto synchronise and measure the regenerated voltage andcurrent across the electrical load
+e voltage and current across the electrical load weremeasured on the design test rig +e measured regeneratedpower is compared with predicted power as shown inFigure 12 +e both results of simulation and measurementare under the harmonic excitation of 1Hz-25mm with anelectrical load of 20Ω It is worth to mention that themeasured average regenerated power is approximately326W compared to the predicted power of up to 331W andthe results show a good agreement especially for the visible
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65
0
200
400
600
800
1000
1200
1400
1600
1800Sh
a sp
eed
(rpm
)
t (s)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(a)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
10
Hyd
raul
ic m
otor
pre
ssur
e (ba
r)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(b)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
200
400
600
800
1000
1200
1400
1600
Mec
hani
cal p
ower
(W)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(c)
0ndash05 05 10 15 20 25 30 35 40 45 50 55 60 65t (s)
0
100
200
300
400500
Rege
nera
ted
pow
er (W
)
Test 1Test 2Test 3Test 4
Test 5Test 6Test 7Test 8
Test 9Test 10Test 11Test 12
Test 13Test 14Test 15Test 16
(d)
Figure 5 Hydraulic motor shaft speed pressure mechanical power and regenerated power at different 16 tests
Shock and Vibration 9
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
variation trend in compression and extension strokes ofcylinder
In Figure 12 it also indicates that the peaks of powerare slightly different and predicted peak is higher than thatof the measured +is is because the smaller motor dis-placement can provide high pressure to obtain higher shaftspeed with more generated power Additionally in realexperiment the road actuator cannot practically provideabsolute stability of sinusoidal excitation due to un-accepted input error in the operating process especially onthe top and bottom of the sinusoidal waveform It is alsothe underlying cause of reducing the nadir of measuredvalue
5 Conclusions
In this paper a hydraulic electric regenerative shock ab-sorber (HERSA) is designed modelled and fabricated toregenerate the kinematic energy of the suspension systemTo maximise the level of regenerated power and powerefficiency a parameter optimisation approach has beenproposed and the result has been validated
A mathematical model has been proposed firstly whichconsists of hydraulic cylinder check valves accumulatorhydraulic motor and other components In the dynamicmodel it considers the flow variation in different chambersof cylinder (compression stroke and extension stroke) the
A1 A2 A3 A4
200
400
600
800
1000
1200
1400
Sha
spee
d (r
pm)
Check valveMotor
CylinderAccumulatorGenerator
B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4Factor level
Figure 6 +e index level of shaft speed at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
2
4
6
8
10
12
14
16
18
20
22
Pres
sure
(bar
)
Factor level
Figure 7 +e index level of motor pressure at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
0
50
100
150
200
250
Elec
tric
al o
utpu
t pow
er (W
)
Factor level
Figure 8 +e index level of regenerated power at different factors
Check valveMotor
CylinderAccumulatorGenerator
A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 E1 E2 E3 E4
050
052
054
056
058
060
062
064
Rege
nera
tion
effic
ienc
y
Factor level
Figure 9 +e index level of regeneration efficiency at differentfactors
10 Shock and Vibration
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
fluid bulk modulus and the accumulator smoothing whichare beneficial to comprehensively understand the systembehaviours in order to further contribute and develop thecorresponding prototype
+e parameters needed to be optimised in HERSAsystem have been pointed out which consist of the size ofhydraulic cylinder the size of check valves the capacity ofaccumulator the displacement of hydraulic motor and theelectrical load +e optimal values of the key componentscan be determined by using the orthogonal method
In parameter optimisation 16 tests were designed andthe corresponding simulated results were obtainedAccording to the principle of the comprehensive equilib-rium method the optimal component combinations of theHERSA were determined and contribute to the selection ofthe components of the test rig +e best combinations of thekey components have been determined the size of cylinder50mm (piston) and 28mm (rod) the diameter of checkvalve 635mm the accumulator capacity 063 L the dis-placement of motor 577 cc and the electrical load of
Table 10 +e rank of factors and their combination
Indicators Rank of factors Best level combinationsShaft speed A D E B C A1D1E1B1C1Hydraulic motor pressure A E D B C A1E1D1B1C1Regenerated power A D E B C A1D1E1B1C1Regeneration efficiency E A D B C E4A1D1B4C2
1 15 2 25 3 35 4 45 50
5001000
Pow
er (W
) Average mechanical power = 509W
1 15 2 25 3 35 4 45 5Time (s)
250300350400
Pow
er (W
) Regenerated power = 331W regeneration efficiency = 65
0 05 1 15 2 25 3 35 4 45 50
1000
2000
Spee
d (r
pm) Shaft speed = 1634rpm
1 15 2 25 3 35 4 45 5253035
Pres
sure
(bar
) Pressure of accumulator = 30bar
Figure 10 +e results with optimal parameter combinations
Actuator
Hydraulic cylinder
Check valve AinCheck valve Aout
Check valve Bin
Check valve Bout
Pipeline
AccumulatorElectrical load
Voltage and current transducer
Control system
Figure 11 +e HERSA test rig
Shock and Vibration 11
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
generator 20Ω respectively It is considered that the pro-posed optimisation procedure based on the orthogonal testis suitable for any general HERSA system
It is worth mentioning that the model prediction hasconfirmed that the recoverable power can be achieved withthe average of 331W at 1Hz-25mm sinusoidal excitationwith the efficiency of up to 65 In addition the HERSA testrig is designed and fabricated in terms of the optimisationresults As a result the experimental validation has beendone and the measured results show good agreement withthe simulated results which verifies the optimisationapproachrsquos effectiveness and reliability
It is worth noting that a HERSA system has many otherobjectives such as comfort and vehicle handing stabilityParticularly feasibility study only needs to focus on thepower regeneration in this paper +e studies about comfortand stability rely on the road test+erefore further researchneeds to pay more attention for the dynamic model in thefuture which can satisfy irregular road profiles And theintegration of HERSA needs to be optimised for the realapplication
Data Availability
+e data used to support the findings of this study are in-cluded within the article
Conflicts of Interest
+e authors declare that there are no conflicts of interestregarding the publication of this paper
Acknowledgments
+is research was sponsored by the Science Foundation ofNational University of Defense Technology (nos ZK17-03-02 and ZK16-03-14) National Key Research and Devel-opment Program of China (no 2017YFB1300900) ChineseNational Natural Science Foundation (no 51605483) andSichuan Science and Technology Program (2019JDRC0081)
References
[1] R Wang Z Chen H Xu K Schmidt F Gu and A D BallldquoModelling and validation of a regenerative shock absorbersystemrdquo in Proceedings of the 2014 20th International Con-ference on Automation and Computing IEEE Cranfield UKSeptember 2014
[2] L Segel and X Lu ldquoVehicular resistance to motion asinfluenced by road roughness and highway alignmentrdquoAustralian Road Research vol 12 no 4 pp 211ndash222 1982
[3] A Browne and J Hamburg ldquoOn road measurement of theenergy dissipated in automotive shock absorbersrdquo in Pro-ceedings of the Symposium on Simulation and Control ofGround Vehicles and Transportation Systems vol 80 no 2Anaheim CA USA 1986
[4] P Hsu ldquoPower recovery property of electrical active sus-pension systemsrdquo in Proceedings of the 31st Intersociety EnergyConversion Engineering Conference (IECEC 96) vol 3 IEEEWashington DC USA August 1996
[5] H Zhang X X Guo and Z G Fang ldquoPotential energyharvesting analysis and test on energy-regenerative suspen-sion systemrdquo Journal of Vibration Measurement amp Diagnosisvol 35 no 2 pp 225ndash230 2015
[6] Y Okada and H Harada ldquoRegenerative control of activevibration damper and suspension systemsrdquo in Proceedings of35th IEEE Conference on Decision and Control vol 4 IEEEKobe Japan December 1996
[7] Y Okada and K Ozawa ldquoEnergy regenerative and activecontrol of electro-dynamic vibration damperrdquo in Proceedingsof the IUTAM Symposium on Vibration Control of NonlinearMechanisms and Structures Springer Munich Germany July2005
[8] K Nakano Y Suda S Nakadai and Y Koike ldquoAnti-rollingsystem for ships with self-powered active controlrdquo JSMEInternational Journal Series C vol 44 no 3 pp 587ndash5932001
[9] K Nakano Y Suda and S Nakadai ldquoSelf-powered activevibration control using a single electric actuatorrdquo Journal ofSound and Vibration vol 260 no 2 pp 213ndash235 2003
[10] W D Jones ldquoEasy ride Bose Corp uses speaker technology togive cars adaptive suspensionrdquo IEEE Spectrum vol 42 no 1p 68 2005
[11] Y Zhang K Huang F Yu Y Gu and D Li ldquoExperimentalverification of energy-regenerative feasibility for an auto-motive electrical suspension systemrdquo in Proceedings of the2007 IEEE International Conference on Vehicular Electronicsand Safety IEEE Beijing China December 2007
Table 11 Key component differences between model and test rig
Parameters Simulation ExperimentHydraulic cylinder 50ndash28 50ndash30Bore rod (mm)Check valve (mm) 635 (14 inch) 635 (14 inch)Accumulator capacity (L) 0635 060Hydraulic motordisplacement (cc) 577 600
Electrical load (Ω) 20 20
0 05 1 15 2 25 3 35 4Time (s)
280
300
320
340
360
380
400
Pow
er (W
)
Regenerated power excitation 1Hz-25mm
Measured average = 326WSimulated average = 331W
Figure 12 +e validation of regenerated power (1Hz-25mm)
12 Shock and Vibration
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
[12] Z Li L Zuo J Kuang and G Luhrs ldquoEnergy-harvestingshock absorber with a mechanical motion rectifierrdquo SmartMaterials and Structures vol 22 no 2 article 025008 2013
[13] Z Fang X Guo L Xu and H Zhang ldquoExperimental study ofdamping and energy regeneration characteristics of a hy-draulic electromagnetic shock absorberrdquo Advances in Me-chanical Engineering vol 5 article 943528 2013
[14] Z G Fang X X Guo L Xu and J Zhang ldquoResearching onvalve system of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mechanics and Mate-rials vol 157-158 pp 911ndash914 2012
[15] Z Fang X Guo L Xu and H Zhang ldquoAn optimal algorithmfor energy recovery of hydraulic electromagnetic energy-regenerative shock absorberrdquo Applied Mathematics amp In-formation Sciences vol 7 no 6 pp 2207ndash2214 2013
[16] Z Fang X Guo L Zuo et al ldquo+eory and experiment ofdamping characteristics of hydraulic electromagnetic energy-regenerative shock absorberrdquo Journal of Jilin University(Engineering and Technology Edition) vol 44 no 4pp 939ndash945 2014
[17] C Li R Zhu M Liang and S Yang ldquoIntegration of shockabsorption and energy harvesting using a hydraulic rectifierrdquoJournal of Sound and Vibration vol 333 no 17 pp 3904ndash3916 2014
[18] P Zheng R Wang and J Gao ldquoA comprehensive review onregenerative shock absorber systemsrdquo Journal of VibrationEngineering amp Technologies pp 1ndash22 2019
[19] L Xu and X Guo ldquoHydraulic transmission electromagneticenergy-regenerative active suspension and its working prin-ciplerdquo in Proceedings of the 2010 2nd International Workshopon Intelligent Systems and Applications IEEE France October2010
[20] R Wang F Gu R Cattley and A Ball ldquoModelling testingand analysis of a regenerative hydraulic shock absorbersystemrdquo Energies vol 9 no 5 p 386 2016
[21] K Ahmad and M Alam ldquoDesign and simulated analysis ofregenerative suspension system with hydraulic cylindermotor and dynamordquo in Proceedings of the SAE TechnicalPaper Series (No 2017-01-1284) USA March 2017
[22] H Zhang G Li Y Wang Y Gu X Wang and X GuoldquoSimulation analysis on hydraulic-electrical energy re-generative semi-active suspension control characteristic andenergy recovery validation testrdquo Transactions of the ChineseSociety of Agricultural Engineering vol 33 no 16 pp 64ndash712017
[23] X Guo H Liu B Wang and Z Xu ldquoRoad surface roughnessmeasurement and its reconstructionrdquo Vehicle amp PowerTechnology vol 4 pp 14ndash18 2010
[24] B Christoph Hydraulische achsantriebe im digitalen regelk-reis PhD thesis RWTH Aachen University AachenGermany 1995
[25] L I Wenjing and A N Gongchang Application of KirchhoffrsquosVoltage Law in Circuit Analysis University of ElectronicScience and Technology Chengdu China 2013
[26] ISO 3320-2013 Fluid Power Systems and ComponentsmdashCy-linder Bores and Piston Rod Diameters and Area Ratios MetricSeries International Organization for Standardization (ISO)Geneva Switzerland 2013
[27] Z A Xu T B Wang and C Y Li Brief Introduction to theOrthogonal Test Design China Building Materials Science ampTechnology Zhaoqing China 2011
[28] R H Dong B H Xiao and Y S Fang ldquo+e theoreticalanalysis of orthogonal test designsrdquo Journal of Anhui Instituteof Architecture vol 6 p 29 2004
Shock and Vibration 13
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom
International Journal of
AerospaceEngineeringHindawiwwwhindawicom Volume 2018
RoboticsJournal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Active and Passive Electronic Components
VLSI Design
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Shock and Vibration
Hindawiwwwhindawicom Volume 2018
Civil EngineeringAdvances in
Acoustics and VibrationAdvances in
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Electrical and Computer Engineering
Journal of
Advances inOptoElectronics
Hindawiwwwhindawicom
Volume 2018
Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom
The Scientific World Journal
Volume 2018
Control Scienceand Engineering
Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom
Journal ofEngineeringVolume 2018
SensorsJournal of
Hindawiwwwhindawicom Volume 2018
International Journal of
RotatingMachinery
Hindawiwwwhindawicom Volume 2018
Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Chemical EngineeringInternational Journal of Antennas and
Propagation
International Journal of
Hindawiwwwhindawicom Volume 2018
Hindawiwwwhindawicom Volume 2018
Navigation and Observation
International Journal of
Hindawi
wwwhindawicom Volume 2018
Advances in
Multimedia
Submit your manuscripts atwwwhindawicom