1. Introduction

Today’s world sees rapid industry development and this makes us more energy dependent. Fossil sources occupy the largest share of electricity production [1], [2]. The use of conventional energies (around 87% of global energy consumption) has an undesirable impact on the environment in terms of greenhouse gas (GHG) emissions and safety (nuclear accidents). 36.5 billion tons of CO₂  is emitted in 2020 [3]. To overcome these problems, it is necessary to resort to alternative energies [4–9]. Among them is Photovoltaic Modules (PVM) solar energy. Its annual growth rate over the last ten years is estimated at more than 40% [10]. Despite their low efficiency, PVM still have a high price. To get around this efficiency problem, techniques for optimizing the power generated by the PVM are proposed [11], [12–16]. This technique, makes it possible to run after the maximum power that the module is capable of supplying [17]. The generated power depends on weather conditions irradiation and temperature [18–20]. Under these latter conditions, the electrical characteristics of the PVM have only one optimal. So not too much trouble for the technique to converge the system to the MPP. Under non-uniform conditions of irradiation and temperature, the electrical characteristics of the PVM present several optimal points. The algorithm used must therefore be able to distinguish the global optimum from the local optima [21]. It therefore requires a sophisticated algorithm which will be able to make a global exploration of the search space in order to make the system converge towards the global MPP [22].

The most used of the classical methods are the P&O method and the InC method [5], [23–26]. Several researchers have used the P&O technique in their work. This method, based on the voltage disturbance and the observation of the variation of the power, is very widespread in the literature. Originally, it was designed to exceed the limits of other types of deterministic controls such as Hill Climbing, Open Circuit Voltage Fraction, etc. It also presents limits linked mainly to the response time and the numerous oscillations around the MPP. The InC method is proposed [3, 18, 25]. In [27], according to the results, the incremental conductance algorithm performs better than the perturb and observe algorithm. To overcome the problems linked to the limits of the methods mentioned above under variable of climatic conditions, Soft-Computing methods based on Meta-heuristic algorithms and Artificial Intelligence (AI) algorithms are proposed. MPPT techniques are based almost exclusively on these techniques. They are very numerous and diverse. They range from Evolutionary Algorithms (Genetic Algorithm), Meta-heuristic algorithms (Optimization by Particle Swarms) and AI algorithms (Artificial Neural Networks, Fuzzy Logic, ANFIS) [21], [10, 11, 28–31].

Artificial Neural Networks (ANN) have been used in several works to optimize the power delivered by a PVM. Its operating principle is inspired by that of the human brain. With their great generalization capacity, they are used for solving complex optimization problems [32], [33]. This is the case in [34] where MPPT method based on ANN is compared to MPPT method based on P&O. The results show that ANN method is more robust than P&O method, regardless of the operating conditions of the PVM. Its limits lie in its lack of interpretation and the difficulty of determining the appropriate number of Layer/Neurons. In addition, these latter limits represent strong points for the Fuzzy Logic (FL) algorithm. The interpretive inability encountered with ANN is resolved by the use of LF. It uses its linguistic variables to overcome this problem [35], [36]. Using the classical logic process, it has facility for extension and interaction [37], [38].

FL also has limits which can be circumvented. Which makes them two complementary techniques and the combination of which gives the Neuro-Fuzzy technique including ANFIS.

In reference [39], two ANFIS models are proposed for grid-connected PV system current injection and battery control. The results show that the proposed the models offered allow the battery charging/discharging process to be supervised and at the same time injecting good quality energy into the grid.

 In [40] two MPPT techniques based on  ANFIS and P&O are compared. Comparison results show that the technique based on the ANFIS algorithm is more robust.

In [5] Five MPPT techniques are proposed. A comparative study is carried out between them. The simulation results show that the best performances are obtain with ANFIS with an efficiency of 99.4%, against 98.1% and 97.5% respectively for FL and P&O.

The power optimization technique using the ANFIS algorithm is more robust than other techniques such as FL, ANN, InC and P&O. It overcomes the problems encountered with ANN and FL as it is a complementary technique linking the two [28, 38, 41].

Other types of techniques based on algorithms whose principle is inspired by the evolution of nature are presented in the literature [11, 42]. Among them there are methods based on the Particle Swarms Optimization (PSO) algorithm  [43, 44].

In [45], a comparative study between techniques using PSO, InC and P&O is carried out. Results of simulation show that the PSO technique is the more robust with the most low response time compared to the two others techniques (InC and P&O). In [46], authors proposed a comparative study between PSO, P&O and FL techniques to optimize the MPP of the PVM. Results show that the MPPT technique based on PSO outperformed FL and P&O techniques.

The work in this paper is consisting to do a comparative study between MPPT techniques based on ANFIS, PSO and InC algorithms put under the same conditions in order to indicate the most efficient and the most robust for power optimization problems. A validation of these three commands is done using an experimental database. The scientific contribution of this paper is to design a MPPT technique based on ANFIS algorithm using database collected to solar power plant in tropical zone.

The rest of the article is organized as follows. Section II presents the MPPT proposed approaches. Section III the proposed approach for the MPPT techniques. Section IV presents the results of simulations and discussions. Finally, a conclusion is made in section V.

2. Proposed approach

In solar PV system, power delivered by the PVM is not always the maximum. This is due to the phenomenon of intermittence. As a result, the operating point of the PVM is not even the MPP. It so requires a technique which is able to extract the MPP of the PVM. This technique, called MPPT optimizes the power through the generation of a duty cycle for controlling the static converter.

The block diagram of the studied system is given in Figure 1.

Figure  1: Schematic diagram of the MPPT technique

2.1. Incremental Conductance method based MPPT

The InC algorithm is among the most efficient of the classical algorithms. It is based on the cycle of the change in pressure to the change in voltage of the PVM (equation 1). At MPP, the slope is zero. The tracking is done according to the position of the operational point (or slope) (dPpv / dVpv) relative to the MPP (equation (3)). The latter depends on the value of the conductance (Ipv / Vpv). The sign of the latter indicates whether the MPP is reached or not (equation (2)). It is compared to its increment. This amounts to saying that the MPP depends on the voltage variation and that of the current [18,26],[47,48]. Thus, the algorithm increments or decrements the duty cycle of the static converter to continue the MPP. The flowchart is shown in Figure 2.

Figure 2: Incremental Conductance flowchart based MPPT

2.2. PSO method based MPPT

Particle Swarm Optimization is an evolutionary meta-heuristic approach. It is used for solving optimization problems. Its principle is based on the behavior of particles (individuals) [49]. In this paper we use a swarm of birds (Figure 3). In this type of swarm, collective intelligence is involved. Particles converge towards those with the best performance [50].

Figure 3: Movement of a particle in the swarm of birds

The duty cycle α of the boost converter represents the particles. Velocity and position are initialized to begin and the movement of the particles are described by equations (4) and (5).  For evaluating the best position of the particles, fitness function is calculated according to the equation (6). Vpv and Ipv are calculated for each particle i with a fixed position in the search space [αmin, αmax]. The algorithm converges the system towards the global optimum. For this the duty cycle is initiated. This ratio, depending on P_besti and G_best, is corrected if it deviates from the best overall duty cycle as in the principle of Figure 3. The particles are then evaluated in terms of position and velocity according to equations (7) and (8). Updates are made to re-evaluate the optimum duty cycle for controlling the boost converter. Table 1 gives the parameters of the PSO. These parameters are defined for the PSO simulations. The flowchart is shown in Figure 4.

Table 1: PSO implementation  parameters

Parameters	Values
Q1	1,2000
Q2	2
m	0,4000
Iterations	25

F is the fitness function,

α is the duty cycle of the boost converter,

t is the time of simulation

r₁ and r₂: uniformly pull in [0,1].

w: coefficient of inertia.

Qi: acceleration coefficient

P_besti is the personal best position of particle i

G_best is the best position of the particles in the entire population.The MPPT technique based on PSO algorithm is described by equations (4) to (8).

Figure  4: PSO flowchart based MPPT

2.3. ANFIS method based MPPT

ANFIS is an adaptive neuro-fuzzy inference system that has five layers in its structure. These layers refine the fuzzy rules already established by human experts and readjust the overlap between the different fuzzy subsets [51–54].

The neural structure replaces the hidden layers with fuzzy rules. This further simplifies learning and interpreting the results obtained. The structure proposed in this work receives the voltage and current from the PVM as inputs and supplies the output with an optimal power comparable to that of the PVM.

The architecture of the MPPT ANFIS technique is given in Figure 5.

Figure  5. Architecture of the MPPT ANFIS technique

Layer 1: Each Neuron calculates the degree of truth of a fuzzy subset by its transfer function. It is called fuzzification layer. Numeric input values ​​are converted to linguistic variables ​​(equations 9 and 10) for high interpretability.

Layer 2: It calculates the degree of activation of antecedents (premises). This is the layer of fuzzy rules (equation 11).

Layer 3: It normalizes the degree of activation of the rules: it is the normalization layer (equation 12).

Layer 4: It determines the parameters of the consequence of the fuzzy rules. Previous linguistic variables will be translated again into numeric values ​​before being sent to the last layer. It is defuzzification (equation13).

Layer 5: It calculates the overall output of the system: it is the output layer. Equation (14) gives the expression of the optimal power generated by the PVM with the ANFIS technique for three fuzzy subsets.

where ai and bi are the parameters of the premise of the MsF, and Popt is the optimal power delivered by the ANFIS controller.

With nit the number of iterations when learning the ANFIS algorithm. Figure 6 gives the flowchart of the technique implemented in simulink and table 2 the learning parameters. The matlab “anfisedit” interface is used for learning the MPPT ANFIS command with a real database made up of two inputs (Vpv and Ipv) and one output. The number and type of MsF are fixed as well as the number of iterations. The hybrid learning algorithm is used and an error tolerance of 1e-4 is arbitrarily set.

Table 2 : ANFIS learning parameters

Parameters	Values
FIS	Takagui-Sugeno
MsF (Type)	Gaussian
MsF (Number)	3   3
Fuzzy rules	9
Epochs	10
RMSE (Training)	0.000123
RMSE (Checking)	0.002012

3. Simulation Results and Discussions

In this section, the results of simulations under matlab/simulink are presented. These simulations are performed with a real database. The latter first allowed to digitally characterize the PVM before being used for the validation of the three techniques presented in previous sections. The PVM consists of two PVs in series. The photovoltaic platform shown in Figure 7 is used in this study. The characteristics of the photovoltaic system being identified (Sharp Module) are given in Table 3. This platform is located at the Polytechnic high school of Cheikh Anta Diop University, Dakar, Senegal. This country is a tropical zone with an adequate rate of sunshine (5.7 kWh/m²/day) for the installation of solar PV plant.

Figure  6: ANFIS flowchart based MPPT

Table 3: Characteristics of the PV

Parameters	Values
VCO	20 V
VMPP	16 V
ISC	2.5600 A
IMPP	2.4300 A
PMPP	38.3800 W

Figure 7: Experimental bench

Figure  8: Experimental database of irradiation and temperature

In this subsection, the results obtained by the three MPPT methods are visualized. A comparative study is then carried out in order to detect the best order. Table 4 shows the simulation parameters and the Figure 9 gives the Simulink model.

Figure 9: System simulated under matlab/simulink

Table 4: System electrical parameters

Parameters	Values
Input capacity	200.6230 µF
Inductance	1 mH
Output capacity	480 µF
Load	34.8000 Ω
Switching frequence	15 kHz

The RMSE and MAPE criteria are evaluated at the level of equations (17) and (18). They characterize the difference between the power generated by the control and the real power (Figure 14). The efficiency of the MPPT technique is given by equation (19).

Figure 12 and Figure 13 respectively show the currents and voltages of the PVM obtained with the different MPPT techniques. They follow changes in weather conditions (Figure 8) which have a considerable influence on the point of operation of the PVM. In Figure  12, we see that the currents obtained with the PSO and ANFIS techniques are almost identical while that obtained with the InC technique is much lower. The opposite phenomenon is observed on the voltage curves in Figure 13. Indeed, the MPPT InC technique pursues the MPP first by comparing the voltage and the current of the PVM. Then, as the principle is based on conductance, the control performs compensation to achieve the desired MPP. Only, as it evaluates each time the variation of the power compared to the voltage, it often diverges with an enormous overshoot of the voltage. In addition, being also a static control, it takes a relatively long time to adapt to a climatic disturbance. This induces a small and slow variation in its duty cycle (Figure 10). As a result, it has difficulty extracting maximum power for these non-uniform irradiation and temperature conditions. Figure 11 represents the power curves of the three controls with respect to the measured power (reference). It reveals overruns for the PSO and InC techniques while the ANFIS technique follows the set-point with an accuracy of 93.46%.

Furthermore, the performance criteria presented in Figure 14 show that the ANFIS technique is better with an extremely low RMSE (0.0194), no overshoot (D=0) and a higher efficiency of around 99.9984%. It is followed by the PSO technique with a RMSE of 1.7235 and an efficiency of 94.8748%.

Figure  10: Variation in the duty cycle

Figure  11: PVM powers

Figure  12: Currents of the PVM

Figure 13: PVM voltages

Figure 14: MPPT controller’s performance parameters

These results corroborate those found in the literature which highlight the oscillating nature of the power obtained with the InC technique due to their inability to detect with precision the overall maximum in a situation of non-uniform climatic conditions [3]. They also confirm the results presented in [38] where the authors made a comparison between ANFIS and InC.

4. Conclusion

MPPT techniques are used to optimize power generated by PVM. In this work, a comparative studied between PSO, ANFIS and InC is done. An experimental validation of these MPPT techniques is also done using real database. The simulation results show that the MPPT technique based on ANFIS algorithm has the best performances. However, if we have to choose between these methods, the two parameters to consider are the complexity of the implementation and the performance of the method. For the first criterion, the choice will be relatively focused on the InC because of its great ease of implementation. For the second criterion, ANFIS will be chosen because of its best performance in terms of response time and accuracy. But we can conclude that the ANFIS and PSO methods give good results compared to the InC method.

Conflict of Interest

The authors declared that there is no conflict of interest.

Acknowledgments

The authors thank the efficiency and energy system research team of Alioune Diop University of Bambey (Senegal) and the Laboratory of Water, Energy, Environment and Industrial Processes, ESP (Dakar-Senegal).

1. W.H. Ali, P. Cofie, J.H. Fuller, S. Lokesh, E.S. Kolawole, “Performance and Efficiency Simulation Study of a Smart-Grid Connected Photovoltaic System,” Energy and Power Engineering, 71–85, 2017, doi:10.4236/epe.2017.92006.
2. E. hadji M. Ndiaye, A. Ndiaye, M. Faye, S. Gueye, “Intelligent Control of a Photovoltaic Generator for Charging and Discharging Battery Using Adaptive Neuro-Fuzzy Inference System,” International Journal of Photoenergy, 2020, 2020.
3. S. Shabaan, M.I.A. El-sebah, P. Bekhit, “Maximum power point tracking for photovoltaic solar pump based on ANFIS tuning system,” Journal of Electrical Systems and Information Technology, 5(1), 11–22, 2018.
4. N. Vela, F. Chenlo, “Early degradation of silicon PV modules and guaranty conditions,” 85, 2264–2274, 2011, doi:10.1016/j.solener.2011.06.011.
5. M.A. Enany, M.A. Farahat, A. Nasr, “Modeling and evaluation of main maximum power point tracking algorithms for photovoltaics systems,” Renewable and Sustainable Energy Reviews, 58, 1578–1586, 2016.
6. F. Chekired, A. Mahrane, M. Chikh, Z. Smara, “Optimization of energy management of a photovoltaic system by the fuzzy logic technique,” Energy Procedia, 6, 513–521, 2011, doi:10.1016/j.egypro.2011.05.059.
7. V. Indragandhi, R. Logesh, V. Subramaniyaswamy, V. Vijayakumar, P. Siarry, L. Uden, “Multi-objective optimization and energy management in renewable based AC/DC microgrid,” Computers and Electrical Engineering, 70, 179–198, 2018, doi:10.1016/j.compeleceng.2018.01.023.
8. A. Ndiaye, “Etude et conception de commandes intelligentes dédiées aux systèmes d’injection d’énergie photovoltaïque dans le réseau électrique de distribution,” Thèse de Doctorat de Ecole Supérieur Polytechnique de Dakar, 2014.
9. A.A. Fröhlich, E.A. Bezerra, L.K. Slongo, “Experimental analysis of solar energy harvesting circuits efficiency for low power applications,” Computers and Electrical Engineering, 45, 143–154, 2015, doi:10.1016/j.compeleceng.2014.09.004.
10. F.K. Slimane Hadji, Jean-Paul Gaubert, “Theoretical and experimental analysis of genetic algorithms based MPPT for PV systems,” Energy Procedia, 74, 772–787, 2015, doi:10.1016/j.egypro.2015.07.813.
11. M.K. M. Zagrouba, M. Bouaïcha, A. Sellami, “Optimisation par les algorithmes génétiques et modélisation par la méthode LPV d’un système photovoltaïque,” Vème Congrès International Sur Les Energies Renouvelables et l’Environnement, 1–6, 2010.
12. S. Titri, C. Larbes, K. Youcef, K. Benatchba, “A new MPPT controller based on the Ant colony optimization algorithm for Photovoltaic systems under partial shading conditions,” Applied Soft Computing Journal, 58, 465–479, 2017.
13. E. hadji M. Ndiaye, A. Ndiaye, M. Faye, “Experimental Validation of PSO and Neuro-Fuzzy Soft-Computing Methods for Power Optimization of PV installations,” in 8th IEEE International Conference on Smart Grid, IEEE Xplore, Elhadjimbaye2020: 189–197, 2020, doi:10.1109/icsmartgrid49881.2020.9144790.
14. B. Long, L. Huang, H. Sun, Y. Chen, F. Victor, K. To, “An intelligent dc current minimization method for transformerless grid-connected photovoltaic inverters,” ISA Transactions, (xxxx), 2018, doi:10.1016/j.isatra.2018.12.005.
15. M. Farhat, “Advanced ANFIS-MPPT Control Algorithm for Sunshine Photovoltaic Pumping Advanced ANFIS-MPPT Control Algorithm for Sunshine Photovoltaic Pumping Systems,” 2012 First International Conference on Renewable Energies and Vehicular Technology Advanced, (April 2012), 2014.
16. H. Bounechba, A. Bouzid, H. Snani, A. Lashab, “Electrical Power and Energy Systems Real time simulation of MPPT algorithms for PV energy system,” International Journal of Electrical Power and Energy Systems, 83, 67–78, 2016, doi:10.1016/j.ijepes.2016.03.041.
17. S. Hadji, J. Gaubert, F. Krim, “Real-time Genetic Algorithms-based MPPT: Study and comparison ( Theoretical an Experimental ) with conventional methods,” Energies, (Ic), 2018, doi:10.3390/en11020459.
18. R.O. Nabil Karami, Nazih Moubayed, “General reviews and classification of different MPPT techniques,” Renewable and Sustainable Energy Reviews, 68(September 2016), 1–18, 2017.
19. P. Hunoor, S.R. Savanur, “Design and Analysis of ANFIS Controller to Control Modulation Index of VSI Connected to PV Array,” European Journal of Advances in Engineering and Technology, 2(5), 12–17, 2015.
20. P.F. Torres, A.F.P. Costa, V.L. Chaar Junior, W.L. Monteiro, M.A.B. Galhardo, J.T. Pinho, W.N. Macêdo, “A mobile educational tool designed for teaching and dissemination of grid connected photovoltaic systems,” Computers and Electrical Engineering, 76, 168–182, 2019, doi:10.1016/j.compeleceng.2019.03.017.
21. F. Belhachat, C. Larbes, “Global maximum power point tracking based on ANFIS approach for PV array con fi gurations under partial shading conditions,” Renewable and Sustainable Energy Reviews, 77(February), 875–889, 2017, doi:10.1016/j.rser.2017.02.056.
22. W. Issaadi, S. Issaadi, A. Khireddine, “Comparative study of photovoltaic system optimization techniques: Contribution to the improvement and development of new approaches,” Renewable and Sustainable Energy Reviews, 82(August 2017), 2112–2127, 2018, doi:10.1016/j.rser.2017.08.041.
23. E. Fadil, H. El, A. Yahya, “Maximum Power Point Tracking Algorithm for Photovoltaic Systems under Partial Shaded Conditions,” IFAC-PapersOnLine, 49(13), 217–222, 2016.
24. T.H. Kwan, X. Wu, “High performance P & O based lock-on mechanism MPPT algorithm with smooth tracking,” Solar Energy, 155, 816–828, 2017, doi:10.1016/j.solener.2017.07.026.
25. H. Bounechba, A. Bouzid, K. Nabti, H. Benalla, “Comparison of perturb & observe and fuzzy logic in maximum power point tracker for PV systems,” Energy Procedia, 50, 677–684, 2014.
26. D. Gueye, A. Ndiaye, F. Mactar, “Design Methodology of Novel PID for Efficient Integration of PV Power to Electrical Distributed Network,” International Journal of Smart Grid, 2(1), 2018.
27. I.V. Banu, R.Ă. Beniug, M. Istrate, “Comparative Analysis of the Perturb-and-Observe and Incremental Conductance MPPT Methods,” THE 8th INTERNATIONAL SYMPOSIUM ON ADVANCED TOPICS IN ELECTRICAL ENGINEERING May 23-25, 2013 Bucharest, Romania, (July 2014), 2013, doi:10.1109/ATEE.2013.6563483.
28. Y.K. Semero, J. Zhang, D. Zheng, S. Member, “PV Power Forecasting Using an Integrated GA-PSO-ANFIS Approach and Gaussian Process Regression Based Feature Selection Strategy,” CSEE JOURNAL OF POWER AND ENERGY SYSTEMS, 4(2), 210–218, 2018, doi:10.17775/CSEEJPES.2016.01920.
29. S. Paul, J. Thomas, “Comparison of MPPT using GA optimized ANN employing PI controller for solar PV system with MPPT using Incremental Conductance,” International Conference on Power, Signals, Controls and Computation (EPSCICON), (January), 8–10, 2014.
30. S. Hadji, F. Krim, J. Gaubert, U. De Poitiers, “Development of an algorithm of maximum power point tracking for photovoltaic systems using Genetic Algorithms,” 7th International Workshop on Systems, Signal Processing and Their Applications (WOSSPA), 43–46, 2011.
31. P. Jood, S.K. Aggarwal, V. Chopra, “Performance assessment of a neuro-fuzzy load frequency controller in the presence of system non-linearities and renewable penetration,” Computers and Electrical Engineering, 74, 362–378, 2019, doi:10.1016/j.compeleceng.2019.02.009.
32. A. Harrag, H. Bahri, “Novel neural network IC-based variable step size fuel cell MPPT controller Performance , efficiency and lifetime improvement,” International Journal of Hydrogen Energy, 42(5), 3549–3563, 2016, doi:10.1016/j.ijhydene.2016.12.079.
33. S. Saravanan, R.B. N, “RBFN based MPPT algorithm for PV system with high step up converter,” Energy Conversion and Management, 122, 239–251, 2016.
34. S. Messalti, A. Harrag, A. Loukriz, “A new variable step size neural networks MPPT controler: Review, simulation and hardware implementation.,” Renewable and Sustainable Energy Reviews, 68(August 2015), 221–233, 2017.
35. F. Chekired, Z. Smara, A. Mahrane, M. Chikh, S. Berkane, “An energy flow management algorithm for a photovoltaic solar home,” Energy Procedia, 111(September 2016), 934–943, 2017, doi:10.1016/j.egypro.2017.03.256.
36. M. Nabipour, M. Razaz, S.G.H. Seifossadat, S.S. Mortazavi, “A new MPPT scheme based on a novel fuzzy approach,” Renewable and Sustainable Energy Reviews, 74(February), 1147–1169, 2017, doi:10.1016/j.rser.2017.02.054.
37. O. Kraa, M. Becherif, M.Y. Ayad, R. Saadi, M. Bahri, A. Aboubou, I. Tegani, “A Novel Adaptive Operation Mode based on Fuzzy Logic Control of Electrical Vehicle,” Energy Procedia, 50(0), 194–201, 2014, doi:10.1016/j.egypro.2014.06.024.
38. E.M. Ndiaye, A. Ndiaye, M.A. Tankari, G. Lefebvre, “Adaptive Neuro-Fuzzy Inference System Application for The Identification of a Photovoltaic System and The Forecasting of Its Maximum Power Point,” 7th International IEEE Conference on Renewable Energy Research and Applications, ICRERA 2018, 5, 1–7, 2018.
39. M.M. Ismail, A.F. Bendary, “Smart Battery Controller using ANFIS for Three Phase Grid Connected PV Array System,” Mathematics and Computers in Simulation, 2018.
40. K. Amara, A. Fekik, D. Hocine, M. Lamine, “Improved performance of a PV solar panel with Adaptive Neuro Fuzzy Inference System ANFIS based MPPT,” 7th International IEEE Conference on Renewable Energy Research and Applications, ICRERA 2018, 5, 1098–1101, 2018.
41. B. Kebe, O. Ba, B. Niang, L. Thiaw, “Etude , synthèse et implémentation d ’ un contrôleur neuro- floue pour la commande de groupe turbo-alternateur,” 2(January), 46–53, 2018.
42. M. Elloumi, R. Kallel, G. Boukettaya, “A comparative study of GA and APSO algorithm for an optimal design of a standalone PV/Battery system,” 15th International Multi-Conference on Systems, Signals & Devices (SSD), 1104–1109, 2018.
43. A. El Dor, “Perfectionnement des algorithmes d ’ optimisation par essaim particulaire: applications en segmentation d ’ images et en électronique,” Thèse de Doctorat de l’Unicersité Paris-Est, 2013.
44. K. Khezzane, F. Khoucha, “Application de la Technique PSO pour la Poursuite du PPM d ’ un Système Photovoltaïque,” The 3nd International Seminar on New and Renewable Energies, 1–6, 2014.
45. E.A. Gouda, M.F. Kotb, D.A. Elalfy, “Modelling and Performance Analysis for a PV System Based MPPT Using Advanced Techniques,” EJECE, European Journal of Electrical and Computer Engineering, 3(1), 1–7, 2019.
46. O. Ben Belghith, L. Sbita, F. Bettaher, “MPPT Design Using PSO Technique for Photovoltaic System Control Comparing to Fuzzy Logic and P & O Controllers,” Energy and Power Engineering, 8, 349–366, 2016, doi:10.4236/epe.2016.811031.
47. T.T. Guingane, Z. Koalaga, E. Simonguy, F. Zougmore, D. Bonkoungou, T.T. Guingane, Z. Koalaga, E. Simonguy, F. Zougmore, D.B. Modélisation, “Modélisation et simulation d ’ un champ photovoltaïque utilisant un convertisseur élévateur de tension ( boost ) avec le logiciel MATLAB / SIMULINK,” JOURNAL INTERNATIONAL DE TECHNOLOGIE, DE L’INNOVATION, DE LA PHYSIQUE, DE L’ENERGIE ET DE L’ENVIRONNEMENT, 2(1), 2016.
48. I.V. Banu, M. Istrate, “Modeling of Maximum Power Point Tracking Algorithm for Photovoltaic Systems,” ICEPE, (July 2014), 2012, doi:10.1109/ICEPE.2012.6463577.
49. S. Motahhir, A. El Hammoumi, A. El Ghzizal, “The Most Used MPPT Algorithms: Review and the Suitable Low-cost Embedded Board,” Journal of Cleaner Production, 2019, doi:10.1016/j.jclepro.2019.118983.
50. S. Krishnamurthy, J.P. Ram, “EL-PSO based MPPT for Solar PV under Partial Shaded Condition,” Energy Procedia, 117, 1047–1053, 2017, doi:10.1016/j.egypro.2017.05.227.
51. J.R. Jang, “ANFIS: Adaptive-Network-Based Fuzzy Inference System,” IEEE, 23(3), 1993.
52. A. Fallah, F. Zarei, H. Zarrabi, M.J. Lariche, “ANFIS-GA modeling of dynamic viscosity of N- Alkane in different operational conditions,” Petroleum Science and Technology, 0, 1–7, 2018, doi:10.1080/10916466.2018.1458117.
53. R.K. Kharb, F. Ansari, S.L. Shimi, “Design and Implementation of ANFIS based MPPT Scheme with Open Loop Boost Converter for Solar PV Module,” International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering, 3(1), 6517–6524, 2014.
54. A. Sharifian, M.J. Ghadi, S. Ghavidel, L. Li, J. Zhang, “A new method based on Type-2 fuzzy neural network for accurate wind power forecasting under uncertain data,” Renewable Energy, 120, 220–230, 2018, doi:10.1016/j.renene.2017.12.023.

(Click Here)

(Click Here)