technical review no.86 [contribution] strategic …...strategic vision for smart machining tool and...
Post on 09-Apr-2020
0 Views
Preview:
TRANSCRIPT
-2-
NTN TECHNICAL REVIEW No.86(2018)
[ Contribution ]
1. Introduction
The paradigm shift of the socio-economic structurefrom a "society of equilibrium" to a "society of non-equilibrium" which proliferates and penetrates on aglobal scale and its challenges can be interpreted asthe transition from the "age of tactics" to "age ofstrategy" as shown in Fig. 1. With the change of thenew value system and socio-economic environment,such as a sustainable society, environmentalconsiderations, and global standards, manufacturingis also changing. Its status is termed the change from"linear society/age of railway" to "non-linearsociety/age of great voyage."1) Attention is paid to theunpredictable destination (market) and the process ofgetting there, rather than the foreseeable andpredictable destination (product), and focus is placedon fundamental thoughts (strategy) rather than thepursuit of higher added value through improvementand enhancement (tactics). Along with these trends,the challenges that cannot be solved by extending theconventional approaches are increasing. The pace ofholistic transformation in many fields from science tosociety and economy is steady and fast, even moreaccelerated in recent years. This paper will identify thepotential power of strategic measurement as one ofthe above trends which drives the manufacturingtransformation, namely, the 4th Industrial Revolution,and explore the future development of the predictiveproduction system and smartmachining/measurement.
Yasuhiro TAKAYAProfessor, Department of Mechanical Engineering, Graduate School of Engineering, Osaka University
Establishment of the innovation driven manufacturing, which enables new valuecreation such as customized products and innovative products with services, is expectedin the age of non-equilibrium and strategy. The holistic measurement based on integratingdifferent measurement principles and multi-scale data fusion as well as smart in-processand on-machine measurement technologies will play an important role in the predictivecontrol type manufacturing system. The connected smart in-process and on-machine
measurement systems that give self-recognition ability to the manufacturing system is also the key technology inIndustry 4.0 to establish the innovation driven manufacturing.
Strategic Vision for Smart Machining Tool andMeasuring Instrument
2. Innovation driven manufacturing
In the efficiency driven manufacturing of the"equilibrium/age of strategy," focus was given to thestrategy to achieve addition of higher product valuethrough high precision and efficient production systemsand the tactics to realize it. The tactics were to rapidlypromote a highly developed production system.Therefore, efforts were made to breakdown themanufacturing technology into elemental technology(production tools) such as machining technology,measurement technology, and production technology(CAD/CAE/CAM/CAT) that compose of productionsystems and solve challenges of each element. Thatresulted in a significant outcome. As such, the role ofmeasurement technology was weighed heavily as oneof the production tools (measurement=testing) in Fig. 2and used as a tactic for achieving the objectives givento the "thing" (product) such as high quality, low cost,and short delivery time. When the manufacturingprocess was in the "equilibrium state" for a prolongedtime, "measurement in manufacturing" as a productiontool played a satisfactory role.
Alternatively, in the innovation driven manufacturingof "non-equilibrium/age of strategy," strategy isweighed more heavily than tactics. Strategy in thiscase is new value creation, such as customizedmanufacturing, creation of innovative products, and theassociated services in addition to the new approach ofhigh value addition. As the tactics to respond to thestrategy change over time, achieving predictive
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 1
Strategic Vision for Smart Machining Tool and Measuring Instrument
production systems is necessary. Therefore, the role ofthe measurement technology is significantly differentfrom the past, in addition to applying a predictiveproduction system, as shown in Fig. 3. Themeasurement technology will be required to assumethe strategy itself for achieving the diversifiedobjectives that are heavily engaged with the entire"event" processing, such as market, product, andprocess throughout the "things" (product), and their lifecycle, tightly connected to information. Today, when
the manufacturing process is transitioned to a "non-equilibrium state," the measurement, which is thesource of information (resource), is required to makeinnovative progress as a strategic tool for"measurement for 'event' creation."
-3-
Fig. 1 Railway era to the age of discovery
[Change of structure and challenges]
SocietyEconomy
Business
Engineering
Measurement
Science
Production(Manufacturing)
21th centuryNon-equilibrium
society[Age of strategy]
20th centuryEquilibrium
society[Age of tactics] (Paradigm shift)
· Situation is always stable· Main attention is on the destination (Product)
· How to build? · What to build?· Value as the "Purpose"· Product quality
· Algebraic approach· Knowledge· Solutions are derived when boundary condition is given
· Situation is always changing· Main attention is on the destination (Market) and how to get there (Process)· Why such activities?· Value as the "Means"· Production quality
· Large-scale three dimensional/ multi-dimensional measurement· On-machine/in-process measurement· Holistic measurement
· One-dimensional, two-dimensional, low-scale three-dimensional measurement· Offline measurement· Pre-process/post-process measurement
· Combinational approach· Wisdom· Structured knowledge does not work
· Dominant equation is clear· Predictable given the initial condition· Equilibrium physics· Deterministic model
· Combinational knowledge/wisdom is critical· Non-equilibrium physics/complex system science· Network model
[Age of railway] [Age of great voyage]
Attention ondestination· Increase speed for arriving early· Increase capacity and frequency for transporting more people and cargoPrediction is easy(Linear Society)
Prediction is difficult(Nonlinear Society)
Attention onmode of transportation
· Condition of sea changes time to time· Diversification and complexity, impact from various factors
Fig. 2 Equilibrium era: Measurement technology asproduction tools
Efficiency driven manufacturing
"Thing"
ProductProcess
Production process
Equilibrium state
Market
TacticsMeasurement in manufacturing
Time
(high quality, low cost, short delivery time)[Objective]
Production tool: measurement = inspection
Fig. 3 The age of non-equilibrium: Measurementtechnology as strategic tools
Innovation driven manufacturing Non-equilibrium stateProcess
Market
Product
Life cycle
Product
ProductProduct
Life cycle
Forecast
ForecastLife cycle
Process
Process
Process
Market A Market B
Market C
[Objective]
Measurement in"event" creation
Time
Strategy
"Thing"Information
"Thing"Information
"Thing"Information
(diversification, value creation)
Strategic tool: measurement
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 2
NTN TECHNICAL REVIEW No.86(2018)
3. Concept of new production quality
3.1 Evaluation of machining accuracy andprocess capacity
Taniguchi predicted that the machining limit ofprecision machining technology would reach 0.1μmin 2000, in his 1983 paper. 2) The progress ofmechanical machining technology has been significantand it is now reaching the area of general machininglimits. The machining process is a general technologyinvolving many elements, such as machine tools,workpieces, working conditions, working environment,and products. The machining error is caused by theinfluence of various factors, such as positioningaccuracy and thermal displacement of machine tools,geometric accuracy and wear property of tools,material and hardness of workpieces, machiningconditions, such as dry machining and processing skillset, working environment, such as temperature andvibration, and high precision and complexity ofproducts. Based on the law of error propagation,accuracy of the entire machining process cannot beimproved by compensating unstable factors by stablefactors. That is, if there is even one unstable factor inthe machining process, the overall accuracy isdecreased, as it is determined by that factor.
Machining accuracy can be tackled separatelybetween precision and accuracy. In addition,machining requires capability to satisfy the standardsize and allowance. Fig. 4 shows how to evaluate themachining accuracy and process capacity. If the upperand lower limits of the machining specification are LSLand USL, respectively, and the allowance is T, thenwhen the average machining value (machining size,etc.) is the midpoint value of T and the standarddeviation is σ, the limit machining accuracy p andmachining capability index Cpk or Cp are calculated bythe evaluation equation indicated in Fig. 4,
respectively. For example, when Cp≧1.33, 99.1% ofworkpieces fall within the allowance of 1/1.33=0.75.
Machining accuracy (error) cannot be directly"observed." Therefore, it is necessary to understandhow to "visualize" it through the "window ofmeasurement." Fig. 5 3) shows how the machiningcapability index is evaluated depending on themagnitude of "uncertainty" 4) of measurement. Whenthe allowance is T, LSL and USL shows the lower andupper limits of the machining specification,respectively. For simplicity, assume the averagemachining value (e.g. machining size) to be the centervalue of T and the standard deviation that shows themachining variance to be σ. For example, if the truemachining capability index Cp (or Cpk) = 1.33, then Cpis observed through the "uncertain window ofmeasurement", and when the machining capability isevaluated with the uncertainty U = 0.1 T (Fig. 5 (a)),then Cp (Cpk) is evaluated as 1.24. Alternatively, whenthe uncertainty U = 0.2T (Fig. 5 (b)), Cp (Cpk) = 1.04and it is evaluated as the lower machining capability. Italso reveals that the entire uncertainty U', whichincludes the uncertainty of measurement, is largerthan σ. Therefore, the magnitude of uncertainty asthe index to show that the reliability of measurementdetermines the reliability against assurance ofmachining accuracy. The uncertainty of measurementis an important index for highly efficient machiningprocesses based on the evaluation of machiningcapability. In order to assure high precision/highaccuracy machining by the strategic measurement "tocreate value," it is necessary to use measurementswith higher reliability (lower uncertainty). As seenabove, improvement of process capability is requiredby reducing machining errors from the machiningprocess, that is, bias errors and variation errors fromthe absolute values, for improving the machiningaccuracy. To do so, it is most effective to examine all
-4-
Fig. 4 Process capability index (Cpk, Cp) for evaluating the machining accuracy
Probabilistic density (frequency)
Bias error d Averageμ
USLLSL
*Errors from the machining process include bias errors and variation errors. The following value is evaluated to determine if the machine tool has machining capability to satisfy required accuracy of parts.
【 Process capability index Cpk 】
*When average,μ, is the center value of T
(*LSL: Lower Specification Limit, USL: Upper Specification Limit)
p=d+σ
Machiningaccuracy limit p:
ε
3σ
σ
Allowance T
Standard sizeSize value X
Variation errorσ
Cpk = min
Cp = =
USL-μ
6σUSL-LSL
6σT
3σμ-LSL,
Allowablemachining errorε
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 3
-5-
the elements of the entire machining processconsidering all the factors in the machining process assubject to measurement and apply improvement to themost unstable factor.
3.2 From product (quality) to production qualityMachining accuracy is determined by the complex
superposition of machining errors that vary withvarious unpredictable factors, such as the propertiesof the machine tools and workpieces, machiningenvironment, and machining procedures. Theaccidental errors included in the machining errorscannot be corrected, as their cause cannot bedetermined. However, they can be estimated withrepetitive measurements and, therefore, it is effectivein mass production, where the same machining isrepeated for the same components (many probabilistictrials). Machining accuracy can be considered as thestatistically fluctuated quantity, that is, probabilisticallyvariable quantity. Therefore, the conventional qualitycontrol approach used in mass production of theefficiency-driven manufacturing is based on themachining accuracy evaluation of the statistical model.
On the other hand, together with the increase ofdemand of energy related facilities such as powergeneration turbines and customized products, thedemand for aircraft parts is increasing globally, inrecent years as shown in Fig. 6 5). Domesticproduction of large precision parts which requirerigorous quality control is also increasing. For small-lotor one-unit production or customized productionwhere new quality control concepts and strategicmachining measurements as the base for production
system are required, not only products but the processwill also drastically change, along with the agingprocess of individual products in their life cycle andchange of strategy. In these cases, carefulconsideration is required for applying quality controlbased on the conventional statistical concept as thesystematic error in machining errors change in theenvironment where variation of temperature and agingof machines cause change in accuracy. In addition,when the number of units for machining is very smalland the process tends to change, verification ofappropriateness of the statistical assumptions shownin Fig. 4 will be difficult.
Therefore, the concept of production quality is beingproposed 6), as a new production control and qualitycontrol approach, as shown in Fig. 7, which can beapplied to small-lot or one-unit production, orcustomized production solving the challenges of theconventional statistical process control approach. Thisapproach emphasizes quality/function assurancefocused on the process as the basic concept by"informatizing every process of manufacturing", notonly geometric quantity of workpieces as the index tomeasure the quality. It establishes production qualitycontrol by defining innovative and integratedproduction quality that uses enormous amount of data(industrial big data), designing an entire product lifecycle from production logistics to maintenance basedon such definition, introducing new management /control methods for big data with advancedtechnology to achieve them. The technologicalinnovation required for that includes product testingtechnology, machining monitoring technology, and
Fig. 5 Importance of measurement uncertainty to evaluate process capability index(Reconfigured and summarized reference 3) Weckenmann et al., Precision Engineering (2000))
Probabilistic density (frequency)
Machining accuracy Measurement accuracy Process capability evaluation+
+
=
=
+ =
U = 0.1T
U = 0.2T
LSL
LSL
USL
USL
LSL USL
Measuredquantity
Measuredquantity
Measuredquantity
Measuredquantity
Measurementaccuracy
Measurementaccuracy
AllowanceT
Average
Average
Measured quantityU : Uncertainty
(a) In the case that uncertainty U = 0.1T
(b) In the case that uncertainty U = 0.2T
U U
U U U
T
U
Cp = 1.24(or Cpk)
Cp = 1.04(or Cpk)
Cp = 1.33(or Cpk)
U
T
U
U’ = σ 2+U 2
Process capabilityevaluation
Process capabilityevaluation
Machiningaccuracy
Machiningaccuracy
Strategic Vision for Smart Machining Tool and Measuring Instrument
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 4
customized production or one-unit production willincrease in the "non-equilibrium/age of strategy,"where the adopted strategy is to generate new valueby creating innovative products. Therefore, productionquality control based on in-process/on-machinemeasurement is indispensable, which is also the keytechnology for achieving predictive production system.
-6-
Fig. 7 Fundamental scheme of production quality control and the required innovative technologies to establish it
Source: Japan Aircraft Development Corporation
Source: Japan Aircraft Development Corporation
(a) Joint development of Boeing 787 (b) Historical trend of aircraft production volume
(c) Forecast of jet passenger aircraft share in operation
Source: "FY2014 outlook of production/export/booking of aircraft (revised)" The Society of Japanese Aerospace Companies (August 2014)
H20 H21 H22 H23 H24 H250%0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
25%
50%
75%
100%
H26(Forecast)
Boeing/Spirit (U.S.) 35%
Vought (U.S.)/Alenia (Italy) 26%Japan 35%
Others 4%
(100 million yen)
40000
34359
26814
18590
13802
10234
DC10,MD11L1011
DC8,707
A38074735000
30000
25000
20000
15000
10000
5000
01992 1997 2002 2007 2012 2017 2022 2027 2032
# of aircraft
Actual
Demand
Remainingaircraft
Forecast400 seats or more
310-399 seats
230-309 seats
170-229 seats
120-169 seats
100-119 seats
60-99 seats
20-59 seats
Forecast of jet passenger aircraft share in operation
777
787
737-900
767
747787
757
A340
A350
A321
A321A330
A320,MD80/MD90727-200,737-300/400
717,727-100,737-100/200/500,TRIDENT,DC80
CS100/300737-600,A318ARJ21,SSJ100,MRJ
ERJ170/190,CRJ700/900/1000328JET,ERJ130/145,ORJ200BAC111/F28/F70/F100,DC8
A310A300
A319/A320
737-700/800
A330
Accessories
Engine parts
Engine body
Aircraft parts
Aircraft body
Fig. 6 Increasing demand for high-precision parts in aerospace industry
multi-sensor data integration technology, as shown inFig. 7 (a) to (h). This technology, in other words,implies that the in-process/on-machine measurementtechnology encompassing a broad range of elementsthat directly determine machining accuracy, such astools, machine tools, machining environment, andmachining/measurement operation, is critical. Inaddition, it is expected that small-lot production and
NTN TECHNICAL REVIEW No.86(2018)
【Concept of conventional quality control】 【Concept of new production quality】(1) Mass production(2) Statistical process control(3) Yield management
(1) Small batch, Customized, or even one-of-a-kind products) (2) New paradigm beyond six sigma approach(3) Zero defect manufacturing
【Key elements of comprehensive production quality control】Innovative and integrated
qualityProduction logistics and
maintenance designManagement and control
methodsAdvanced technological
enablers
【Technological innovation required for establishing production quality】(a) Product inspection technology(b) Process monitoring technology(c) Multi-sensor data fusion technology(d) Learning techniques and cognitive computing methods
(e) E-maintenance technology(f) Product traceability technology(g) Production monitoring technology(h) ICT and digital manufacturing technology
(*ICT: Information and Communication Technology)
(Reconfigured and summarized reference 7) Colledani et al. CIRP annals (2014))
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 5
-7-
4. Production quality control byholistic measurement
4.1 Current status of in-process/on-machinemeasurement
The role of the machining measurement forproduction control is "informatization" of manufacturing.In the advanced production system using diversifiedand vast production information, machiningmeasurement plays an important role as the keyinformation source on the products and machiningstates. The most essential purpose of machiningmeasurement is production control, productionautomation, and test automation. In-process/on-machine measurement is the most efficient andeffective machining measurement technology engageddirectly in monitoring and controlling of machining. Withthe machine tools with reproducibility, the machiningerrors that vary over time during machining operationdue to factors such as deformation and displacementof machine due to change in temperature, thermalexpansion and deflection of tools, and wear may beevaluated and corrected by re-machining.
Recently, machine tools with in-process/on-machinemeasurement capability are being commercializedand the importance of their role is increasing. Thesetrends are driven by diversified purposes of themeasurement. In-process/on-machine measurementplays a central role for production/quality control aswell. Fig. 8 shows the fundamental measurementproperties required in in-process/on-machinemeasurement technology and the unique properties
and relationship between machining factors acrossmachine tools, workpieces, and machining environmentand in-process/on-machine measurement. Of particularimportance as a fundamental measurement property isaccuracy and precision. For example, considerationfor questions such as if accuracy and precision forevaluating tolerance range as the base for processcapability are satisfactory and if absolutemeasurement is required due to machining bias errorsis critical. In addition, the in-process measurementevaluation is affected by dynamic characteristics suchas transient response and frequency response.Furthermore, properties specific to in-process/on-machine measurement are also required. Theseinclude environment factors required for installingmeasurement devices to machine tools andhardware/software factors. For example, in the case ofon-machine 3D profile measurement using an opticalcomparator, vibration property, compactness, andswitching process are required. The advantage of in-process/on-machine measurement is not limited toimprovement and stabilization of machining accuracybut extended to higher efficiency ofmachining/measurement setup. As such, the scope ofmeasurement extends from workpieces to machiningstate in space and time domain realizing "visualizationof machining process" by quantifying machiningfactors.
The machining error, which value is constantlymaintained during the repetitive machining operation,provides the same effect not influencing the accidentalerror. Therefore, it provides a constant bias to the
[Properties specific to in-process/ on-machine measurement]
[Environmental factors]· Temperature, humidity· Vibration· Dust, etc.
[Hardware factors]· Measurement criteria (independent/dependent)· Non-contact/contact· Measurement speed· Compactness[Software factors]· Measurement data processing (error analysis model, correction data generation, stitching, etc.)· Measurement strategy (sequential three-point method, measurement path generation, swing-round method, etc.)
[Machine tool]
[Tool]
[Workpiece]
· Correction of machine coordinates (geometric error)· Evaluation of motion accuracy· Accurate matching of workpiece reference point and machining origin point
· Measurement of length, diameter, and tool edge· Detection of relative coordinate distance between measurement probe and tool edge· Evaluation of tool wear· Evaluation of tool deflection and forming load· Dynamic balance adjustment for grindstone
· Real time verification of machining accuracy· Higher accuracy of machining correction· Higher efficiency of machining process· Evaluation of workpiece deformation
[Working environment][Machining/measurement operation]
· Real time correction of machining state changes due to changes in temperature and external vibration· Pre-inspection of machining/measurement· Higher efficiency of machining/measurement setup (positioning of tools and workpieces)· Higher efficiency of measurement operation (standardization of tool path probe path software, etc.)
[Basic measurement properties]
Absolute criteria(Master value)
Exactness
Average of data pool
Measurement device
Machine tool
Upper limitLower limit
Precision
Machining (product)allowance
Measuredquantity
[Static properties]· Exactness (absolute accuracy)· Precision (repeatability)· Range· Linearity· Traceability[Dynamic properties]· Transient response· Frequency response
Fig. 8 Fundamental properties of measuring technology in machining process and roles of on-machine measurement
Strategic Vision for Smart Machining Tool and Measuring Instrument
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 6
machining accuracy. In general, evaluation ofsystematic error which cannot be found in theserepetitive measurements is difficult and often itsexistence cannot be detected. However, if the factorfor systematic error is predetermined and itsmagnitude is known, it can be eliminated withcorrection. Therefore, a smart machine tool which candetermine the factors of machining errors in real timewith multiple in-process/on-machine measurements isan important concept as an elemental technology ofthe production system in conventional efficiencydriven manufacturing. The basic concept of the smartmachine tool that holistically provides machiningprocess monitoring, recognition, and prediction usingmultiple sensors was presented around 1993 byMoriwaki 7). Since then, active research anddevelopment has been conducted on in-process/on-machine measurement and machining control aimedat application to practical systems, resulting inincreased development in application fields such asadvanced sensors for tool condition monitoring (TCM),signal processing methods, decision making strategy,etc. These trends of research activities on thechallenges of in-process/on-machine measurementindicate that attention is paid to the solution ofindividual challenges for each element oftools/workpieces/machine tools which are the mainelements of the machining process, and thechallenges in the boundary area from interaction of
-8-
・On-machine measurement of tool wear (1976): direct/indirect in-process measurement method
・On-machine measurement for process monitoring (1983): AE sensor, accelerometer, spindle motor output, etc.
・Tool state monitoring TCM (1995): Optical surface measurement sensor, accelerometer, AE sensor, etc.・Measurement of workpiece-tool distance (1998): In-process measurement machining accuracy correctionMeasurement of cylindrical surface micro profile (2006): Optical angle sensor・Motion accuracy measurement of 5-axis machine tool (2008): New measurement principle (two-axis scale type)・On-machine measurement of aspheric surface work (2009): High accuracy contact probe・On-machine measurement of tool cutting edge shape (2009): Three-dimensional measurement by AFM・In-process evaluation of edge motion and workpiece shape (2009): Laser displacement sensor・In-process evaluation of tool position (2010): Optical imaging measurement of tool length and diameter・Evaluation of motion accuracy of 5-axis machine tool (2010): Geometric error analysis by improved DBB・On-machine measurement of workpiece shape (2010): Laser displacement sensor
・Measurement of tool deflection and cutting load (2010): Camera imaging measurement・Direct evaluation of cutting force (2011): Development of cutting force detection spindle・On-machine measurement of machined surface roughness (2014): Contact probe (surface roughness gauge)
・On-machine measurement of mirror finished surface (1992-1999): Fizeau interference, zone-plate interference, radial shearing interference, etc.
Evaluation of machiningdynamic characteristics
[Workpiece][size, shape,
surfacecharacteristics,
stress, temperature, etc.] [On-machine system]
Machining atmosphere (environmental change)
Dynamicresponse
Thermal
deformatio
n
Holding/
binding
Drive
mechanism
Dynamic
deformatio
n
Stat
icde
form
atio
nW
ear
Complex vibration
behavior
Dynamic characteristicsof cutting and grinding
Hol
ding
/bi
ndin
g
Drive
mechan
ism
Measurement
operation factors
Machini
ng
opera
tion f
actors
[Tool][size, position,
vibration, forming load,
wear, temperature, etc.]
Phenomena thatchange over amedium period
Abrupt change, phenomena thatchange in a shortperiod
Phenomena thatchange over along period
[Machine tool][space error,
vibration, motion accuracy, thermal
displacement, wear, etc.]
Machining statemonitoringGeometric quantitymeasurement (size, shape, dimension, etc.)
Fig. 9 Spreading roles of on-machine measuring technologies in machining process
these elements. In other words, adaptation toadvanced and complex requirements of high-speed,multi-functional, and intelligence is sought after inaddition to higher accuracy of machining processes,from the viewpoint of practical application of in-process/on-machine measurement to the real system.As the complexity of machining process increases,interaction of tools, workpieces, and machine toolshas become more complex and the challenges of in-process/on-machine measurement in the boundaryarea have also become multifaceted and advanced.
Areas subject to in-process/on-machinemeasurement are expanding from geometric quantity tostate quantity and the challenges are not onlymultifaceted, but expanding with advanced multi-axis/multi-tasking machine tools and intelligentmachining process. In addition, the requirement offurther improvement of accuracy for ultra-precisionmachining is expanding into new dimensions, such as3D free curve profile, micro complex profile, and nano-surface pattern, which are required for commercializationof advanced functional parts. Fig. 9 summarizespositioning and scope of in-process/on-machinemeasurement technology in the on-machine systemwith tools, workpieces, and machine tools as its keyelements. It indicates the boundary areas of eachelement and relation among measurement technologyand breaks down the time-variant phenomena intoshort, medium, and long period and marks their
NTN TECHNICAL REVIEW No.86(2018)
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 7
-9-
positioning in the respective time scale. The relation ofin-process/on-machine measurements is indicated byring bands connecting two elements and a ring bandcircling three elements. The former indicates interactionof the elements, and the latter indicates a complexinteraction area where three elements are involved. Inaddition, the outer ring band that encloses the entirediagram indicates machining atmosphere and manualoperation. The purpose of measurement is classifiedinto evaluation of machining dynamic characteristics,machining state monitoring, and measurement ofgeometric quantity, such as size, shape, anddisplacement, with R&D examples shownchronologically from [a] to [ p]. For example, in the caseof [a], on-machine measurement of tool wear (1976) ispositioned as the on-machine measurement technologywith the purpose of monitoring machining state thatchanges over time in the boundary area between toolsand machine tools. Also, the in-process/on-machinemeasurement technology that corresponds to thesolution to challenges other than the interactionbetween elements are only indicated with factors as thechallenges to be solved in the future.
4.2 New in-process/on-machine measurementstrategy
As the basic concept of new in-process/on-machinemeasurement by multiple sensors, an example of highfunctional measurement information is shown in Fig.10by "multi-sensor fusion" and "multi-sensor cooperation".In the manufacturing process of large precision parts, astrategy for proper installation and placement of in-process/on-machine measurement system on themachine tools at the manufacturing sites is required.
Therefore, a machining monitoring system 8) 9) and on-machine dimensional measurement systems are builtand installed on the machine tools based on theconcept of "multi-sensor fusion" and "multi-sensorcooperation," respectively.Fig. 10 (a) shows the concept of multi-sensor fusion
on-machine measurement for machining statemonitoring. By conducting state estimations combiningfeature amounts from multiple sensors, characteristicsof different physical amounts can be used to improvethe detecting capability of machining monitoring. In themachining process that requires long time,measurement information (physical amount varied overtime) obtained from the respective sensors is integratedin real time. Use of advanced signal processing todetect feature amounts that correlate with abnormalitybased on the time-frequency analysis provides adaptivethreshold management in machining monitoring. Itshigh adaptability to environment and abnormalitysymptom detection capability reduces the risks relatedto production quality, such as chattering and tool failure,which is the most important measurement strategy forachieving a zero-defect process. In addition, in thefuture it is expected to both maintain production qualityand improve process capability by integratingestimation methods and machining monitoringinformation considering multiple analysis technologiesand uncertainty, such as cutting phenomenon, cutterpath, tool/workpiece shapes, and by optimizingmachining conditions based on the process model.Fig. 10 (b) shows the concept of multi-sensor
cooperation on-machine measurement for highprecision measurement of geometric quantity. Higheraccuracy of measurement is pursued by optimizing
Multi-sensor fusion: improved detection capability by usingcharacteristics of different physical quantities
Multi-sensor cooperation: optimized configuration ofcomponents which measure the same geometric quantities
(b) On-machine measurement using multi-sensor cooperation(a) On-machine measurement using multi-sensor fusion
Laserinterferencecomparator
Laserdisplacement
sensor
Positiondetection
probe
Lasertracker
Calibrationgauge
Straingauge
ThermocoupleAccelerometer
AE sensor
AmmeterOpticalsensor
Emissionof light
Stress
TemperatureAcceleration
Vibration
Soundwave
Motor current Thermaldisplacement
Surfaceroughness
stylus
High accuracymeasurement of
geometric quantity[size, shape, position, attitude,
surface characteristics,displacement, motion accuracy,
etc.]
Machining statemonitoring
[Machining abnormality, tool wear/damage,
dynamic characteristics,etc.]
Fig. 10 The smart on-machine measurement system using multi-sensor fusion and cooperation
Strategic Vision for Smart Machining Tool and Measuring Instrument
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 8
and allocating respective roles for multiple componentswith the purpose of measuring the same geometricquantity (e.g. size, shape, attitude, and position) basedon their measurement elements. The example of on-machine dimensional measurement systems integratedwith the large CNC lathe for machining large precisionparts consists of laser interferometer, laser tracker,tough trigger probe, and block gauge to realize highprecision, multi-sensor cooperation for on-machinemeasurement providing absolute length calibration.
In the case of the machining process of steamturbine rotors, the operation of moving the work frommachine tools to the measurement system was difficultand on-machine measurement was manuallyconducted using large micrometers, etc. Therefore, theproblem was that it involved many measurementprocesses, resulting in increased uncertainty. To solvethis problem and achieve efficiency of themanufacturing process and machining qualityassurance, a high precision on-machine measurementsystem was developed. Fig. 11 shows an example ofon-machine measurement system for large structuralparts by "multi-sensor cooperation" targeting steamturbine rotors (max. length of 10m or more, diameterof 1m or more, and dimensional tolerance of 0.1mm orless) of large CNC lathes. It shows configuration of thesystem which conducts absolute measurement ofwheel position and axis diameter of the steam turbinerotor of the large CNC lathe by probing measurementmethod 10). An inline length measuring system usinglaser tracker is introduced as the external coordinatesystem and a wireless touch trigger probe isimplemented on the tool carriage. In addition, as theparts and machine elements are both affected byshrinkage and expansion (about 0.1mm of impact per1m if the rotor temperature changes from 20˚C to10˚C), a calibration method of absolute length isimplemented using a block gauge to correct biaserrors.
-10-
Workpiece (protruded cylindrical part: wheel)
Reference plane
Carriage
O
Z
X
Positioning on XZplane with resolutionof 0.001 mm
Maximum distancebetween wheel planes
Block gauge
Tool/reflection mirror/probe
Laser tracker
Fig. 11 The developed on-machine measuring systemfor a steam turbine rotor on CNC turning machine
NTN TECHNICAL REVIEW No.86(2018)
4.3 Concept of holistic measurementFor the predictive production system, which is the
foundation for innovation driven manufacturing,development of smart in-process/on-machinemeasurement as the strategic tool for machiningmeasurement technology innovation is required torealize production quality control. As its foundation,the measurement strategy is to realize "informatizationof every process related to manufacturing" bydeveloping high degree of "multi-sensor fusion" and"multi-sensor cooperation," and with holisticmeasurement 11) based on integration of differentmeasurement principles and multi-scale data fusion.For realizing holistic measurement, new theory forevaluating uncertainty, new comprehensivemeasurement principle to encompass the calibrationtechnology, and conventional measurement principleand broader research regarding the compatibilityamong them are required.
As a specific example, a basic concept of intelligentmulti-sensor coordinate measuring machine (CMM) ispresented 12). The holistic measurement uses newmeasurement technology, such as multi-sensor CMMwith metrology X-ray CT (volumetric measurement).With the advancement of measured data fusiontechnology by integrating CAD/CAE/CAM/CAT data,this builds the foundation for advanced productionquality control while different measurement systemsbecome connected with each other through theInternet. The new strategic machining measurementwhich extends the concept of this measurementfusion technology to in-process/on-machinemeasurement is the smart in-process/on-machinemeasurement that realizes environment adaptivemeasurement with such properties as real-time, widerange, and multi-point simultaneity. As shown in Fig.12, the optical non-contact measurement technologyin a broad sense, including X-ray, is making rapidprogress and technology that can measure surfacemicro-profile, 3D profile, and 3D internal structure hasalready been developed. Metrology X-ray CTtechnology in particular is making significant progressand its incorporation into inline processes andapplications for assembly process inspection arebeing proposed 13). When building multi-sensor CMMusing these measurement technologies, 3Dmeasurement data fusion technology is required formeasurement integration of micro-profiles to complexlarge 3D profiles and volumetric measurement data.Efforts for developing measurement fusiontechnology 14) is already in progress. However, sometechnical challenges still remain, specifically relatedto multi-scale data fusion, such as measurement dataquality including uncertainty, standard and calibration
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 9
methods, the basic quantity of measurement such asone-dimensional length and displacement, and two-dimensional surface micro profile with different anglesand scale.
5. Outlook and challenges of smartmachining/measurement machines
5.1 The 4th Industrial Revolution and predictiveproduction system
The 4th Industrial Revolution is a disruptivetransformation unlike the previous industrialrevolutions, as significant innovation is expected notonly in the manufacturing process but also across theentire product life cycle due to the rapid technologicalinnovation in information technology, as evidenced bythe promotion of Industry 4.0 in Germany andIndustrial Internet in the U.S. The ultimate objective ofthe 4th Industrial Revolution is achievement of a newpredictive production system based on the totaloptimization and efficiency of the manufacturingprocess, development of predictive maintenance, etc.That is, creation of new value by innovative productsthat fill the gap of "invisible value" and shift from massproduction to complete customized production (one-unit production). In the manufacturing process withcompletely different mechanisms from the past,production quality has better adaptability than thequality control approach used in mass production.Furthermore, the basic concept of the 4th IndustrialRevolution requires intelligence with autonomy (e.g.self-recognition, self-maintenance, and self-prediction)as its core property to the members of themanufacturing system (e.g. products and individual
-11-
manufacturing facilities). The key is to build predictiveproduction systems with self-recognition capability.Internet of Things (IoT) which converts everything todata and connects them to the Internet through sensortechnology, smaller and faster processors, andadoption of cloud technology, industrial big data whichis collected by IoT, and cyber physical system (CPS)which use them will play the central role as themechanism for such intelligence 15). In addition, themeasurement technology, which plays the role ofinformation resources, is required to undergotransformation as a strategy tool. Particularly, smartin-process/on-machine measurement technology is animportant and high-quality source of productioninformation together with controllers and networksystems, and requires technology innovation as astrategy tool, such as self-recognition and self-prediction embedded in the measurement systems.
5.2 Smart in-process/on-machine measurementnetwork
The approach of intelligent machine tools by on-machine measurement using multiple sensors is aconcept for standalone machine tools. Its objective isto increase efficiency of individual machining/measurement systems with knowledge base andincorporation into the machines. Therefore, it remainsin the scope of efficiency driven manufacturing. On theother hand, innovation driven manufacturing requiresintelligent environment adaptive in-process/on-machine measurement, i.e. open and intelligentmachine tools through integration with smart in-process/on-machine measurement. Configuration ofmulti-sensor fusion and multi-sensor cooperation
Measurement speed
Measured data quantity● 3D internal structure measurement technology
●X-ray CT(Research stage)
Holistic measurement (fusion of 3D measured data)
(Commercial stage)
◆ 3D shape measurement technology
◆ Moire fringe method◆ Fringe pattern projection method
◆ Triangulation method◆ Optical interferometry
◆ Stereo image method◆ Digital holography method◆ Photogrammetry
◆ Light-section method
※Surface roughness measurement technology
※Light scattering method
※Coherence scanning interferometry
※Confocal microscope※Focal position detecting method※Phase shifting interferometry
※Stripe pattern projection method
Large
Volume
Surface
Line
Point
3-D
2-D
1-D
Small Slow Fast 1/time
Fig. 12 Rapid progress in three dimensional measurement technologies for industrial measurement
Strategic Vision for Smart Machining Tool and Measuring Instrument
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 10
based on holistic measurement changes in real timeaccording to measured content (information onmeasurement environment, measurement condition,measured amount, subject of measurement,machining condition, required accuracy, allowance,uncertainty, etc.) to achieve advanced prediction. Itsmechanism is based on the network of smart in-process/on-machine measurement connected invirtual space, as shown in Fig. 13. That is, themeasurement content and measurement informationare shared among intelligent machine tools connectedto IoT, to drive predictive production systems whichcontrol production quality. The smart in-process/on-machine measurement is required to have thecapability to sufficiently satisfy conditions of itsmeasurement information such as amount, speed,diversification, and reliability. Capabilities such as self-recognition, self-maintenance, and self-prediction areobtained by CPS.
6. Summary
The most recent "2017 White Paper on ManufacturingIndustries" 16) reports the initiatives and measures thatthe manufacturing industry in Japan is taking for the 4th
Industrial Revolution. The white paper mentions that thevalue of virtual data from activities on the Internet ispredominantly dependent on the "quantity" of datawhich is already dominated by the U.S. IT companies.However, the value of the real data collected bysensors in real life activities, such as operational data ofthe factory facilities, is potentially dependent on the"quality" of data and the companies in themanufacturing industry may be able to take a leadingrole depending on the future course of action. In fact,the survey results regarding data collection of theproduction process in Fig. 14 indicates the trend that
-12-
the activities of "visualizing" operational states acrossthe machines in the individual process, production line,the entire production process, traceability management,collection of production process data from overseasfactories, etc. are all increasing. Therefore, it isconsidered that smart in-process/on-machinemeasurement will be increasingly important as a high-quality information source. In addition, solutions usingrecent digital technology are considered to be the"engineering chain" and "supply chain" indicated in Fig.15. which also shows the mechanism to make thosechains more efficient by quickly feeding back the smartin-process/on-machine measurement data. In contrast,the requirements for smart in-process/on-machinemeasurement, such as improvement of reliability andquality on the measurement data, are becoming higherand higher, as the data is more integrated, complex,and significantly larger. Fig. 16. indicates the transition of machining
measurement technology up to the 4th IndustrialRevolution and transformation to the strategicmachining measurement, as the foundation ofproduction quality control. Machining measurementtechnology has been playing a role in elementtechnology for production systems that satisfy therequirements of automation of inspection, productioncontrol, automation of production, etc. In that process, itunderwent a series of revolutionary improvements inmeasurement capability, such as improvements inmeasurement accuracy and speed, highersophistication from one dimensional to two dimensional,etc. However, the measurement technology required forthe 4th Industrial Revolution is not an extension ofconventional technology, but a fundamentaltransformation. That is, the predictive productionsystem that takes advantage of IoT, industrial big data,and CPS based on real time holistic measurement
NTN TECHNICAL REVIEW No.86(2018)
When and wherewas it measured?
Which machinetool was used? Machiningcondition?
Industrialbig data
VolumeVelocityVarietyVeracity
Product(Virtual space)
Measurementenvironment?
Uncertainty?Allowance? Parts of which
machine?
*RFID (Radio Frequency IDentifier); IC tag
VisibilityValue
Smart in-process/on-machine
measurement
Autonomous operation by cyberphysical system (CPS) (use ofbig data, AI and IoT)
IoT
RFID
Fig. 13 Connecting data of smart in-process/on-machine measurement in cyber physical space
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 11
-13-
[Collecting any data fromthe domestic factories?]
["Visualization" of data collection and use of data for production process improvement such as traceability management]
2016
2015年
(n=4566)
(n=3751)
Work on process improvement by"visualizing" availability of machinesin the individual processes
Work on process improvement by"visualizing" availability of machinesin the lines and entire productionprocesses
Work on process improvement by"visualizing" availability of workforcerelated to lines and production process
Manage traceability of product/materialamong intra-company or inter-companyfactories
Work on collection and use of datarelated to production process inoverseas factories[owns overseas factories]
Already implemented Planned Hope to implement if possible No plan No data collection implementedSource: METI Survey (Dec. 2016)
Source: METI Survey (Dec. 2016)
Source: METI Survey (Dec. 2015) ※ Note that the simple comparison to the survey of the previous year is not appropriate as the number of survey respondents increased by approx. 20%
No33.4% Yes
66.6%
No59.4%
Yes40.6%
【2016 survey】(n=3966)
【2015 survey】(n=3507)
【2016 survey】(n=3980)
【2015 survey】(n=3518)
【2016 survey】(n=3970)
【2015 survey】(n=3510)
【2016 survey】(n=3897)
【2015 survey】(n=3496)
【2016 survey】(n=720)
【2015 survey】(n=626)
15.5%
18.1%
13.9%
16.3%
9.1%
10.2%
16.4%
14.7%
10.0%
11.5% 7.0% 21.9% 8.1% 51.4%
6.8% 38.9% 16.7% 27.6%
3.2% 11.8% 6.6% 63.7%
4.0% 24.1% 16.4% 39.1%
5.6% 17.8% 2.9% 63.5%
7.1% 35.4% 10.0% 38.4%
5.0% 13.7% 63.3%1.7%
7.5% 31.4% 8.9% 38.3%
4.6% 12.3% 1.4% 63.5%
7.2% 30.0% 8.9% 38.4%
Fig. 14 Research results on data collection of production process
Supply chain
Engineering chain
(Quickly reflect customer usage datato supply chain)
Expand and increase speed of data usageusing digital tools
Customer
(Quickly reflect customer usage datato planning and design)
Order process
Production control
Product Design
Product planning
Production(Production association)
MaintenanceFollow-through service, etc.
Logistics/distribution
Optimized production
Joint order process
High mix low volume production
Technology inheritance(Digitalization of expertise)
Design support
Logistics optimization
Sales forecast
Predictive maintenance(customer)
Remote maintenance(customer)
Operation optimization(customer)
Fundamentally new service
Planning support
R&D support
※ Improvement of productivity and value addition by optimization and efficiency within company
※ Direct improvement of customer valueR&D
Expand and increase speed ofdata usage using digital tools
Productivity improvement Creation of new added value
Predictive maintenance (intra-company)
Remote maintenance(intra-company)
Operation optimization(intra-company)
Fig. 15 Effective utilization of IOT in Industry 4.0
Strategic Vision for Smart Machining Tool and Measuring Instrument
that uses intelligent in-process/on-machinemeasurement is considered to achieve the ultimateproduction quality control with self-recognition of theintelligent production elements.
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 12
References1) Shuichi Fukuda, "On Value Creation" Maruzen,
20052) Taniguchi, N., Current Status in, and Future Trends
of, Ultraprecision Machining and Ultrafine MaterialsProcessing, Annals of CIRP, Vol. 32/2, 1983), 2.
3) A. Weckenmann, M. Rinnagl, “Acceptance ofprocesses: do we need decision rules?”, PrecisionEngineering, Vol.24, (2000), 264-269.
4) International Organization for Standardization, “Guideto Expression of Uncertainty of Measurement”, JCGM100, (2008).
5) Ministry of Economy, Trade and Industry, Ministryof Health, Labor and Welfare, Ministry ofEducation, Culture, Sports, Science andTechnology, 2015 White Paper on ManufacturingIndustries Overview; FY2014 Promotion measuresof core manufacturing technology (Overview),(2015)
6) Marcello Colledani, Tullio Tolio, Anath Fischer, BenoitIung, Gisela, Lanza, Robert Schmitt, Józef Váncza,Design and management of manufacturing systemsfor production quality,CIRP Annals - ManufacturingTechnology, Vol.63/2, (2014), 773-796.
7) Byrne G, Dornfeld D, Inasaki I, Ketteler G, König W,Teti R, Tool Condition Monitoring (TCM) — TheStatus of Research and Industrial Application, Annalsof CIRP, Vol. 44/2, (1995), 541–567.
8) Masahiro Uekita, Yasuhiro Takaya, Tool conditionmonitoring for from milling of large parts by combiningspindle motor current and acoustic emission signals,International Journal of Advanced ManufacturingTechnology, Vol.89, Combined 1-4, (2017), 65-75.
9) Masahiro Uekita, Yasuhiro Takaya, Tool conditionmonitoring technique for deep-hole drilling of largecomponents based on chatter identification intime–frequency domain, Measurement, Vol.103,(2017), 199-207.
10) Masahiro Uekita, Yasuhiro Takaya, On-machinedimensional measurement of large parts bycompensating for volumetric errors of machine tools,Precision Engineering, 43, (2016), 200-210.
11) A. Weckenmann, THE ROLE OF METROLOGY INTHE EXCHANGE OF GOODS AND IN THEDEVELOPMENT OF MANUFACTURINGTECHNOLOGIES, Proceedings of the10th
International Symposium on Measurement andQuality Control 2010, (2010), 1-4.
12) A. Weckenmann, X. Jiang, K. –D. Sommer, U.Neuschaefer-Rube, J. Seewig, L. Shaw, T. Estler,Multisensor Data Fusion in Dimensional Metrology,CIRP Annals - Manufacturing Technology, Vol.58/2,(2009), 701-721.
-14-
NTN TECHNICAL REVIEW No.86(2018)
Mea
sure
men
t dat
a am
ount
/mea
sure
men
tac
cura
cy/m
easu
rem
ent s
peed
(1980s and later) 200019701900
Industry1.0
First industrial revolution
Block gauge (1896)・International Meter Convention (1875)
Laser interference comparatorVernier, micrometer
Optical 3D measurementdevice3D coordinate measuringmachine: CMM (1970)
Industrial X-ray CTMulti-sensor CMM
· Manual measurement· SI-unit (1960)
· Coordinate measurement, point group measurement· Automated measurement, computer control· High degree of freedom, universal measurement· Offline measurement
· CAD/CAM/CAE integrated measurement· Combined measurement (multi-sensor fusion, multi-sensor cooperation)· In-process/on-machine measurement
Second industrial revolution Third industrial revolution Fourth industrial revolutionLatter half of18th century
Beginning of20th century
Latter half of20th century Present
Industry2.0Industry3.0
Industry4.0
2D characterization (contour form)
3D characterization (3D form) Volume measurement (3D volume)
Holistic measurement (fusion of 3D measured data)
1D characterization (dimension)
Revolutionary improvement of measurement capability (machining measurement): standardization, higher accuracy, 3D/automation
Fig. 16 Revolution of strategic measurement technology in production inspired by Industry 4.0
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 13
-15-
13) L. De Chiffre, S. Carmignato, J.-P. Kruth, R. Schmitt,A. Weckenmann, Industrial applications of computedtomography, CIRP Annals - ManufacturingTechnology, Vol.63/2, (2014), 655-677.
14) A. Weckenmann, X. Jiang, K. –D. Sommer, U.Neuschaefer-Rube, J. Seewig, L. Shaw, T. Estler,Multisensor Data Fusion in Dimensional Metrology,Annals of the CIRP, Vol.58/2, to be published, 2009.
15) Jay Lee, "Industrial big data; challenges ofmanufacturing industry toward Industry 4.0",Nikkan Kogyo Shimbun, Ltd., 2016
16) Ministry of Economy, Trade and Industry, Ministryof Health, Labor and Welfare, Ministry ofEducation, Culture, Sports, Science andTechnology, 2017 White Paper on ManufacturingIndustries Overview; FY2016 Promotion measuresof core manufacturing technology (Overview),(2017)
<Author biography>
Yasuhiro TAKAYAProfessor, Doctor (Engineering), Department of Mechanical Engineering,
Graduate School of Engineering, Osaka University
March, 1992 Engineering Doctor, Doctor Course Completed, Precision Engineering Department, Graduate School of Engineering, Hokkaido University
April, 1992 - June, 1995 Assistant, Industrial Mechanical Engineering, School of Engineering, Osaka UniversityJuly, 1995 - June, 1997 Instructor, Industrial Mechanical Engineering, School of Engineering, Osaka UniversityJuly, 1997 - March, 2006 Assistant Professor, Department of Mechanical System Engineering,
Graduate School of Engineering, Osaka UniversityApril, 2006 - present Professor, Department of Mechanical Engineering,
Graduate School of Engineering, Osaka University
[Specialty]Precision machining measurement, optical applied nano-measurement, optical applied nano/micro machining
[Academic society and committee affiliations]The Japan Society for Precision Engineering: Intelligent Nano-Measure Expert Committee ChairThe Japan Society of Mechanical Engineers: Fellow, 1st Planning Committee Chair, Manufacturing and Machine Tool DivisionThe Japan Society for Abrasive TechnologyThe International Academy for Production Engineering (CIRP): FellowInternational Measurement Confederation (IMEKO): TC14 ChairThe Japan Society for Die and Mould Technology: Director (Vice President)Mitsutoyo Association for Science and Technology (MAST): Advisor and many others
Strategic Vision for Smart Machining Tool and Measuring Instrument
*02-寄稿文E_new_*02/寄稿文 19/07/03 10:26 ページ 14
top related