firefly ann based upqc-s pv array system for power …

Journal of Green Engineering (JGE)

Volume-9, Issue-4, December 2019

FireFly ANN Based UPQC-S PV Array System for Power Enhancement

1Madhavi Dasari and

2V.S. Bharath

1Research Scholar, Department of EEE, Oxford College of Engineering, VTU,

Karnataka, India. E-mail: [email protected] 2Professor, Department of EEE, Oxford College of Engineering, VTU, Karnataka,

India.

Abstract

In this research work, the action of power outcome from the PV array is

controlled and enhanced. This work is proposed to controll the scheme

applied on variable irradiance with altering the power connections by two-

level inverter controlling strategy. Firefly ANN approach is applied to

analyse the PQ theory and fundamental harmonics in voltages. It is done

using (Firefly Based Neural Networks-Anfis Controller)-FANN-AC scheme.

This contribution is used to develop the two stages controlling approach that

adjust the DC voltages. 3 phase non-linear load conditions and its effects on

source voltages and currents are considered.This work is implemented in

MATLAB -15b, Simulink design is implemented to observe the variable

irradiance changes and non-linear load conditions, limit harmonic distortion

in the current.

Keywords: Controllers, Neural Networks,Power Quality analysis, PV array,

UPQC.

1 Introduction

As day by day, growth in electricity utilisation leads to the identification

of the alternate energy sources such as renewable resource energy systems

along with existing solar and advanced solar power source units. UPQC

assisted to minimising the hysteresis losses by accessing both

Journal of Green Engineering, Vol. 9_4, 638–657. Alpha Publishers

This is an Open Access publication. © 2019 the Author(s). All rights reserved

mailto:[email protected]

639 Madhavi Dasari et al

shunt and series controlling circuits. This controlling strategy is accessed by

many researchers, and we followed the same in our approach with a

limitation and to compensate the voltage sag and swell. Firefly based ANN

approach is applied to PO method and this enhanced the voltage levels and

minimised the harmonics at different sections. However, in our approach, we

followed our principle to work on a series active filter rather than the shunt,

which helps to regulate the voltages and minimises the sag levels occurred in

the PV array system. In power electronics systems, power quality (PQ) is

considered as an essential concern in the present era. It is fundamental,

especially with the presentation of best in class gadgets, whose general

execution is defenceless to the top-notch of intensity provided. Power quality

issue is an event concerning shifting voltages, current or recurrence those

results in a disappointment of end-use gadget [1]. The fundamental objective

of conceiving UPQC is the consolidated utilisation of gathering vivacious

and shunt-enthusiastic channels mainly to repay poor-arrangement current

and sounds because the SCR oversaw capacitor banks make up for

responsive power in power recurrence terms [2]. Displaying and recreation of

custom power conditioners appear at be inescapable as vitality hardware-

based absolutely framework used for expanding the quality top of the line in

circulation systems [3].

2 Literature Review

In [3] authors proposed a control methodology of vector control for the

lattice associated single-stage VSI in the photovoltaic framework the goal of

the matrix associated inverter control is to keep up the DC-interface voltage

and autonomous dynamic and responsive power stream.

In [4] suggested that the principle framework benefits that microgrids can

offer to the central system, just as the future patterns in the advancement of

their task and control for the following future, are displayed and discussed.

Using the savvy fuzzy rationale control for DC connect adjustment

dependent on the evenness property, we proposed a straightforward answer

for the quick reaction and adjustment issues in the nonlinear power electronic

framework.

Some authors proposed MPPT calculations (P&O and INC) make the

yield intensity of the PV framework to be preserved at the most extreme

while the illumination also temperature differed also changes the PV

producer to the heap.

The all-inclusive advantage of the UPQC incorporated the result: it has

an equivalent trademark to SCR controlled capacitor banks of achieving load

pay following in illustration the consistent sinusoidal flows inside the current

control mode [4, 5]. Additionally the most extreme power quality

improvement apparatus for vulnerable nonlinear burdens, which need

genuine sinusoidal information supply [6].

Firefly ANN Based UPQC-S PV Array System for Power Enhancement 640

UPQC worked in unmistakable potential setups for unmarried-stage (2-

wire). Also, three-stage (three-twine and four-wire) systems, different

repayment methods, and current patterns additionally situated in the field [7].

Along these lines, this conditioner can procure sensible quality top-notch

advancement, diminishing the vitality aggravations which are given to the

clients through the mains the utilisation of the arrangement unit. Other places

for PQ (i.e., mains vitality interferences) can be presented to the clients

(custom quality) commencing the shunt units [8]. To repeat the essential

guideline of UPQC is to make specific magnificent supply voltage and

payload present-day unsettling influences, explicitly, droops, swell,

irregularity, music, responsive flows, and forefront unbalance delivered by

utilising the nonlinear burdens [9, 10].

The control sign given to the UPQC plays a massive job in sorting out a

superior task of the apparatus. Regular control plans comprehensively

utilised withstanding the control plans got from engineered knowledge which

additionally recorded inside the writing. Utilizations of a couple of cutting

edge numerical riggings in like and wavelet revamp particularly, in power

astounding additionally are completed. A vast arrangement of writing

ensuring uses of fuzzy good judgment, proficient structures, neural systems,

and hereditary calculations in vitality quality is in like manner developing

[11]. The ANN-based controller intends for the advanced administrator of the

shunt-lively power channel and talented disconnected the utilisation of

certainties from the conventional corresponding crucial controller [12].

Accordingly, a virtual controller-based at the TMS320F2812 DSP stage that

did for the reference period for not holding control schemes proposed [13].

3 Problem Identification

Overall writing clarifies about sun-powered related information.

Likewise, PV frameworks are planned yet need to demonstrate and tried with

the proposed controller, to give the most extreme power yield. The control

plan of Grid-associated PV framework investigated for giving the ideal PV

power also in elevation calibre of present infused hooked on the lattice and,

hence, elevated power quality; this was neglected to clarify in above

literature [1-26]. The second thing is the essential factors to be considered in

PQ measurement and tested the PV system with variable solar irradiance.

Power quality control is possible with the proposed method not there in

existed methods. Along with the enhancement of PQ, the transient response

and stability of the grid associated PV scheme examined with proposed

research work. Using the above identifications to analyse the characteristics

of PV systems and mitigate the PQ issues and behaviour of voltage and

current loops for regulating the dc connection voltage also the output inverter

current [15]. At last to infuse vitality hooked on the lattice at solidarity power

factor also current through short consonant twisting.


4 Methodology

In this selection, PV array collects the voltage from solar panels

depending on the radiance intensity from the sun, this may vary for different

seasons and time. Below figure.1 and 2.explained that ANFIS controller-

based work with the help of firefly algorithm using neural networks

phenomenon. With this above diagram, we are reducing the power alterations

and irradiance problem. The proposed approach will help the system to get

better efficiency and unity power factor.

Figure 1 Existing Methodology

In another research paper[14] prescribed the use of fuzzy decision-

making ability (FL) approach inside microgrid power framework based

absolutely at the most extreme current vitality moulding framework

contraptions which incorporate Unified/Power/Quality Conditioner (UPQC).

Thus [15, 16], a straight quadratic controller (LQR) oversee method

implanted through the ANN is utilised to organise the activity of the

arrangement also shunt VSIs of the UPQC. In some other advancement, a

regulator set of standards dependent on wavelet revise for UPQC to stifle

present-day music and voltage hangs is proposed [17]. Along these lines, in

[18], the developers stress the upgrade of vitality palatable with the guide of

the use of UPQC with fuzzy appropriate judgment controller (FLC) as well

as counterfeit neural network organiser on the customary relative principal

(PI) regulator. In [19], the assurance of voltage references for arrangement

fiery power get out is cultivated dependent on a stable III-section advanced

fragment bolted circle (DSLL) device utilising a fuzzy controller.


Figure 2 Existing Work -Without Neural Networks and FireFly Optimization

4.1 Proposed Work

It discovered that the dynamic steadiness of the power framework can

upgrade by enhancing a transient balancing out sign got from deviation in

speed, terminal recurrence, quickening power, rotor point and so forth., on

the ordinary voltage blunder sign of AVR. Thus, Power System Stabilizer

(PSS) is created to deliver these transient balancing out the sign to help in

damping the annoyance motions utilising the balance of generator excitation

signal by applying PSS yield as a strengthening control sign to the AVR this

work is conceivable with FFNN-AC.


Figure 3 proposed FFNN-AC method

In this section, proposed work PV array is collected the voltage from

solar panels depending on the radiance intensity from the sun, this does not

vary for different seasons and time. Above fig.3 and 4a explained that

ANFIS controller-based work with the help of firefly algorithm using neural

networks phenomenon. With this above diagram, we are reduced the power

alterations and irradiance problem. Current and voltage controller assisted

the system to get better efficiency and unity power factor.

Figure 4a ANFIS-NN-FireFly Model


Figure 4 Simulink model of FFNN-AC method

MPPT is a versatile framework used to the regulator a fixed converter

amongst the heap also the PV board (Fig. 4). This converter is intended to fit

each time the evident resistance of the heap to the “impedance” of the PV

meadow relating to the most extreme powerpoint. This technique depends on

the utilisation of an inquiry calculation of the most extreme intensity of the

photovoltaic board bend [8]. There is a wide range of MPPT procedures

accessible in writing; the most to a great extent utilised calculations depicted

in the accompanying areas for example "FFNN-AC".

4.1.1 UPQC

Our approach comprised of PV array with variable irradiance, shunt and

series active filter with Firefly trained NN, non-linear load and source unit

with RC filters.


Figure 5 Block diagram

UPQC controlling strategy: Our approach is dependent on the phasor

diagram for PV-array based UPQC shown in figure 5.

Figure 6 PV-UPQC phasor diagram

Here Inverter shunt currents given by Ish, current from PV array denoted

by Ipv, load current represented by Il. Vs, Vl, Vse given as Source voltage,

Load voltage, Series voltage respectively shown in fig.6 From fig. 7 source

voltage of the circuit is given by

√ ( )( ) (1)


UPQC approach used in this approach helps in injecting the voltages to

load and gets out from voltage sag condition. The derivation is carried out

from [14]. Then this is followed by using Firefly based Neural network

optimisation for controlling the sag voltages in the series compensator.

4.1.2 Firefly Based NN

The idealised behaviour of these fireflies could summarise as follows:

fireflies are unisex, so they attract others regardless of their sex. In this

algorithm, it assumed that the brightness of a firefly defines its attractiveness.

This brightness is then encoded to objective functions f (x) for x ∈ S ⊂ ℝn,

considering as a continuous constrained optimisation problem.

Firefly will train by controlling the weights parameter at each step and

also compares with the positions this relatively reduces the PLL utility and

minimises the sag voltages due to the position acquisition. Our approach

limits the values-based maximum light penetration by each firefly and its

position with respective error weight option. The following steps help us for

training ANN using Firefly algorithm and make us learn how optimisation

carried.

Figure 7 Source circuit

Fig.7 represents the three-phase source and measurement system which

helps the microgrid system developed in this paper as an external

compensator.

ANN OPTIMIZED BY FIREFLY represented in figure 9

Input: Object related to the source

STEP 1: Assign neural network with input, hidden and output layers.

STEP 2: Based Light maximisation principle update weights of neurons.

Step 3: Repeat step 2 until the best light penetrating limit acquired.

Step 4: Verify the best firefly was selected or not

Step 5: If not selected repeat Step 2

Step 6: If best obtained

Step 6.1: Update the weights and generate the best-minimised error value.

Step 7: Repeat step 2 to Step 6 until the best samples obtained.

Step 8: Compare originals limits with this Firefly neural network.

Output: The output of FIREFLY optimiser compensates the sag voltages in

the series compensator.


(a)

Figure 8 a) Transformers circuit b) shunt controller circuit

Fig .8a explained that transformer adjustment of our research work here

shunt connection established for efficiency and power accuracy.Fig.8b shows

that shunt controller in the proposed system this will decreases the time and

increases the PQ with respected to objectives.


Figure 9 Flow chart for ANN with Firefly optimisation.


Figure 10 Series controller circuit

Figure 9 shows that Flow chart for ANN with Firefly optimisation.Fig.10

explains that series controller of circuit here system power quality is more

compared to the shunt system, but the time factor is a concern.

4.1.3 Neural Networks Role

Figure 11 a) Neural Networks Role b) input voltage


5 Results and Discussions

Fig.11a. explained that neural networks data processing from controller

here using layers phenomena analysing the performance, training state,

regression.

Figure 12 Input voltage from grid source and analysis 13 b) validation of echo’s

13 c) target Vs output variation

Fig.12 gives the training performance limits and validating the data error

minimisation using gradient the best-minimised outcome from neural

networks will help in compensating the deflections in source voltages in both

sag and swell modes.


Figure 13 d) Source currents e)Load Voltages at non-linear conditions

Fig.13d) shows that source current from grid this current have so many

disturbances depending on sun intensity. Fig.13e)explains that load voltage

at conditions of nonlinearity here we apply FFNN-AC method. Such that

reducing the sag and swings.

Figure 14 a) Load Currents at non-linear conditions b)THD (threshold) for source

voltages


Fig.14 a) explains that load current at conditions of non-linearity here we

apply FFNN-AC method. Such that reducing the sag and swings. Fig.14 b)

source voltage from different sources here fundamental frequency is 50Hz

THD=0.01%

Figure15a) THD for load voltages 15b) THD at a source current

Fig.15a) source voltage from different loads and the fundamental

frequency is 50Hz THD=0.01% . Fig. 15b) source voltage from different

source current and fundamental frequency is 50Hz THD=0.01%


Figure 16 Harmonic distortion for load current

Fig.16 source voltage from different load current and the fundamental

frequency is 50Hz THD=0.01%

Figure 17. pdc, vdc , idc model


Figure 18 final comparison surface graphs

Here, pdc, vdc , idc model is depicted in Fig 17 and Fig.18 explained that final

comparison graph between irradiance, Pmax Existed work, Pmax proposed

work we achieved good improvement in power quality, stability and

efficiency.

5.CONCLUSION In this Research, the displaying and the “Simulink_models” of the entire

made out of PV “generator”, support conversation and M_PPT calculations

(FFNN-AC) exhibited. As per the consequences of the recreation, we reason

that The PV generator execution breaks down with expanding HEAT,

diminishing of sunlight based illumination and variety of the electrical

burden. The MPPT calculation FFNN-AC make the yield intensity of the PV

framework to reserved at the most extreme when the illumination and

“temperature” changed and modifies the PV generator to the heap.

6 References

[1] Tahiri, F. E., Chikh, K., Khafallah, M., &Saad, “Comparative study

between two Maximum Power Point Tracking techniques for

photovoltaic system” International Conference on Electrical and

Information Technologies (ICEIT) 978-1-4673-8469-8/16.

[2] Joan Rocabert, Alvaro Luna, FredeBlaabjerg and Pedro Rodriguez,

"Control of Power Converters in AC Microgrids," IEEE Transactions on

Power electronics, vol. 27, no. 11, pp.4734-49, 2012.


[3] Thounthong, Phatiphat, ArkhomLuksanasakul, PoolsakKoseeyaporn, and

Bernard Davat, "Intelligent model-based control of a standalone

photovoltaic/fuel cell power plant with supercapacitor energy storage,"

IEEE Transactions on Sustainable Energy, vol.4, no. 1, pp.240-249,

2013.

[4] S.Samerchur, S. Premrudeepreechacharn, Y.Kumsuwun and K.Higuchi.

“Power control of single-phase voltage source inverter for grid-connected

photovoltaic systems” in Power Systems Conference and Exposition

(PSCE), pp. 1-6, 2011.

[5] Dong Cao, Shuai Jiang, Xianhao Yu, and Fang ZhengPeng, "Low-cost

semi-Z-source inverter for single-phase photovoltaic systems," IEEE

Transactions on Power Electronics, vol.26, no. 12, pp.3514-3523, 2011.

[6] Soumyadeep Ray, MadichettySreedhar and AbhijitDasgupta, "ZVCS

based high-frequency link grid-connected SVPWM applied three phases

three-level diode clamped inverter for photovoltaic applications," in

Power and Energy Systems Conference: Towards Sustainable Energy,

pp. 1-6, 2014.

[7] Matthias Klatt, Alicia Dorado, Jan Meyer, Peter Scheiner, Jürgen Backes,

Ran Li, and Dresden–Germany Dresden–Germany Stuttgart–Germany,

"Power quality aspects of rural grids with high penetration of

microgeneration, mainly PV-installations," Proceedings of the 21st

International Conference on Electricity Distribution, pp.1-4, 2011.

[8] Chandani.Chovatia, Narayan P. Gupta, and Preeti N. Gupta, "Power

Quality Improvement in a PV Panel connected Grid System using Shunt

Active Filter," International Journal of Computer Technology and

Electronics Engineering (IJCTEE), vol.2, no. 4, pp.41-45, 2012.

[9] A.Hari Prasad, Y.Rajasekhar Reddy and P.V. Kishore, "Photovoltaic Cell

as Power Quality conditioner for Grid-connected system."International

Journal of Scientific and Engineering Research, vol.2, no.10, pp.1-8,

2011

[10]RiadKadri, Jean-Paul Gaubert, and Gerard Champenois. "An improved

maximum power point tracking for photovoltaic grid-connected inverter

based on voltage-oriented control," IEEE Transactions on Industrial

Electronics, vol. 58, no.1, pp. 66-75, 2011.

[11]S.A.Lakshmanan, B. S. Rajpour hit, and Amit Jain, "A novel current-

controlled SVPWM technique for the grid-connected solar PV system."

In PES General Meeting Conference & Exposition, pp.1-5, 2014.

[12]Yongheng Yang and Frede Blaabjerg, "Low-Voltage Ride-Through

Capability of a Single-Stage Single-Phase Photovoltaic System

Connected to the Low-Voltage Grid", International Journal of Photo

energy, pp.1-9, 2013.

[13]A.Mahmud, H.R Pota and M.J. Hossain, “Nonlinear Current Control

Scheme for a Single-Phase Grid-Connected Photovoltaic System.” IEEE

Transactions on Sustainable Energy, vol.5, no.1, pp.218-27, 2014.


[14]Masoud Farhoodnea, Azah Mohamed, Hussain Shareef, and

HadiZayandehroodi, "Power Quality Analysis of Grid-Connected

Photovoltaic Systems in Distribution Networks," Przeglad Elektro

techniczny, pp.208-13, 2013.

[15]N. Kumarasabapathy and P. S. Manoharan, “MATLAB Simulation of

UPQC for Power Quality Mitigation Using an Ant Colony Based Fuzzy

Control Technique,” The Scientific World Journal, vol. 2015, Article ID

304165, 9 pages, 2015.

[16]J. Bratt, “Grid connected PV inverters: Modeling and simulation”, Thesis

Presented to the Faculty of San Diego State University, 2011.

[17]A. Nordin and A. Omar “Modeling and Simulation of Photovoltaic (PV)

Array and Maximum Power Point Tracker (MPPT) for Grid-Connected

PV System”, 3rd International Symposium & Exhibition in Sustainable

Energy & Environment, 2011.

[18]Q. Chunqing, Y. Yong and S. Ji, “Deadbeat Decoupling Control of

Three-phase Photovoltaic Grid-connected Inverters”, IEEE International

Conference on Mechatronics and Automation. 2009.

[19]R. Benadli, B. Khiari and A. Sellami, “Three-Phase Grid-Connected

Photovoltaic System with Maximum Power Point Tracking Technique

Based On Voltage-Oriented Control and Using Sliding Mode Controller”,

6th International Renewable Energy Congress, 2015.

[20]F. Ding, P. Li, B. Huang, F. Gao, C. Ding and C. Wang, “Modeling

and Simulation of Grid-connected Hybrid Photovoltaic/Battery

Distributed Generation System”, International Conference on Electricity

Distribution, 2010.

[21]M. Makhlouf, F. Messai, H. Benalla, “Modeling and simulation of

gridconnected photovoltaic distributed generation system”, Journal of

Theoretical and Applied Information Technology, vol. 45 no.2, 2012.

[22]F.E. Tahiri, K. Chikh, M. Khafallah and A. Saad, “Comparative study

between two Maximum Power Point Tracking techniques for

Photovoltaic System”, 2nd International Conference on Electrical and

Information Technologies ICEIT, 2016.

[23]J. Jiang, T. Huang, Y. Hsiao and C. Chen, “Maximum Power Tracking

for Photovoltaic Power Systems”, Tamkang Journal of Science and

Engineering, vol. 8, no 2, pp. 147-153, 2005.

[24]L. Abderezak, B. Aissa and S. Hamza, “Comparative study of three

MPPT algorithms for a photovoltaic system control”, World Congress on

Information Technology and Computer Applications (WCITCA), 2015.

[25]D. Dera, T. Kerekes, R. Teodorescu and F. Blaadjerg, “Improved MPPT

algorithms for rapidly changing environmental conditions”, Power

Electronics and Motion Control Conference, 2006.

[26]S. Mohammed, D. Devaraj, T. P. Ahamed, “Maximum Power Point

Tracking System for Stand-Alone Solar PV Power System Using

Adaptive Neuro-Fuzzy Inference System”, Biennial International


Conference on Power and Energy Systems: Towards Sustainable Energy

(PESTSE), 2016.

Biographies

Ms. Madhavi Dasari has obtained her B.E degree from Visvesvaraya

Technological University, Belagavi in the year 2004. She obtained her

M.Tech degree from JNTU, Kakinada in the year 2011. Her research

include power quality, photovoltaic systems, artificial intelligence and

renewable energy.

Mr. V. S. Bharath, has obtained his B.E degree from Madras University,

Chennai in the year 1998. He obtained his M.E degree from Annamallai

University, Chidambram in the year 2002. He completed his Ph.D at Bharath

University, Chennai in the year 2015. He has published over 20 Technical

papers in National and International Conference proceeding/Journals. His

area of interest is Inverter fed AC drives.

firefly ann based upqc-s pv array system for power …

Documents