FireFly ANN Based UPQC-S PV Array System for Power …
Transcript of 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
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.
641 Madhavi Dasari et al
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.
Firefly ANN Based UPQC-S PV Array System for Power Enhancement 642
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.
643 Madhavi Dasari et al
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
Firefly ANN Based UPQC-S PV Array System for Power Enhancement 644
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.
645 Madhavi Dasari et al
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)
Firefly ANN Based UPQC-S PV Array System for Power Enhancement 646
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.
647 Madhavi Dasari et al
(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.
Firefly ANN Based UPQC-S PV Array System for Power Enhancement 648
Figure 9 Flow chart for ANN with Firefly optimisation.
649 Madhavi Dasari et al
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
Firefly ANN Based UPQC-S PV Array System for Power Enhancement 650
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.
651 Madhavi Dasari et al
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
Firefly ANN Based UPQC-S PV Array System for Power Enhancement 652
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%
653 Madhavi Dasari et al
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
Firefly ANN Based UPQC-S PV Array System for Power Enhancement 654
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.
655 Madhavi Dasari et al
[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.
Firefly ANN Based UPQC-S PV Array System for Power Enhancement 656
[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
657 Madhavi Dasari et al
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.