Search

The choice and the dimensioning of the controller for the maximum power point tracking (MPPT) are determined for the ideal energy efficiency of the photovoltaic (PV) systems. Many works have been developed in the field of MPPT methods, especially fuzzy logic controllers. However, these are robust if the parameters of the membership functions have been well designed. In this paper, the performances of an intelligent fuzzy logic controller (FLC)-based MPPT method have been optimized by an evolutionary genetic algorithm (GA). The works presented in the literature have shown the efficiency of the proposed method compared to classical methods. In our paper, the validation of the experimental results obtained is given with respect to a reference signal. The control of the simulated PV source and the proposed method are built using the Simulink/Matlab environment and implemented on the dSPACE DS1104 controller to validate the practical execution of the suggested method. The standalone PV system has been tested in an emulated test bench experimentation. Experimental results confirm the efficiency of the proposed method and its high accuracy in handling fast varying load conditions.

This paper presents a study on the enhancement of the Maximum Power Point Tracking algorithm named: Perturb and Observe (P&O). The introduced techniques deal with the weaknesses of this algorithm, especially when high changes or oscillations occur around the Maximum Power Point (MPP). To this end, a proportional-integral-derivative (PID) controller optimized by the particle swarm algorithm (PSO) is used to conceive an automatic and self-adaptive P&O algorithm for maximum power extraction from the HZ-20 Thermo Electrical Generator (TEG) module feeding a resistive load via a DC-DC converter. The introduced work is developed in the MatLab/Simulink environment. The obtained results are presented in terms of a comparative study between the two most known hybridization structures of the PID controller and the P&O algorithm.

Fuzzy logic controller (FLC)-based maximum power point tracking (MPPT) or simply fuzzy MPPT technique has attracted a large interest because it allows a more human approach to control and shows good performance compared to conventional techniques. However, in spite of the many advantages of the fuzzy MPPT, the design of this controller remains a difficult task due to the large number of parameters and fuzzy rules that needs an optimal tuning. This paper describes a new design method of fuzzy MPPT for photovoltaic conversion system using hierarchical genetic algorithms (HGAs). The chromosomes of HGAs are organized in hierarchical structure and able to optimize all parameters of fuzzy MPPT like the number of fuzzy rules and its consequents, and the membership functions parameters. The optimized fuzzy MPPT is tested in simulation and showed optimal responses even under rapid variations in irradiance. The proposed approach is compared to other fuzzy MPPT design methods currently in use, and the results confirm the superiority of our approach.

International audience; With the exception of a few simple applications, all photovoltaic (PV) conversion systems have a maximum power point tracking (MPPT) unit which allows an optimal extraction of power. In this paper, the fuzzy logic controller and evolutionary methods are combined to obtain an efficient MPPT unit with a fast response in transient state and minimal errors in steady state. The proposed combination shows a good performance in simulation and offers a varied set of MPPT controllers.

In recent years, we assist to growing up interest of PSO algorithm in engineering applications and especially in machine learning. However, the PSO algorithm stills suffer from some weakness as a global search method harming its use in these fields. Many authors have introduced a lot of strategies to improve this algorithm, where the use of multiple swarms seems more promoting. Even if this strategy has significantly improved the performance of the PSO algorithm, it stills not yet enough. In this paper, we introduce a particular environment based on a hierarchical society coordinating the exploration and exploitation mechanisms to enhance the multi-swarm approach. Subjected to well-known optimization benchmark functions, the introduced strategy has proved its efficiency and effectiveness among the most popular versions to improve the potential of the PSO algorithm as a global optimization technique.

International audience; This paper presents an intelligent approach to optimize the performances of photovoltaic systems. The system consists of a PV panel, a DC–DC boost converter, a maximum power point tracker controller and a resistive load. The key idea of the proposed approach is the use of a fuzzy controller with an adaptive gain as a maximum power point tracker. The proposed controller integrates two different rule bases. The first is used to adjust the duty cycle of the boost converter as in the case of a conventional fuzzy controller while the second rule base is designed for an online adjusting of the controller’s gain. The performances of the adaptive fuzzy controller are compared with those obtained using a conventional fuzzy controllers with different gains and in each case, the proposed controller outperforms its conventional counterpart.

International audience; This paper proposes a new method to design a fuzzy logic controller by genetic algorithms (GAs). The basic idea of this new methodology is to use a new way to mutate the chromosome. Bio-process control problem is used to compare the proposed mutation to traditional mutation. The results show that the proposed mutation efficiently improves GA's performance.

International audience; In this paper we propose a hybrid algorithm to optimize the structure of TSK type fuzzy model using backpropagation (BP) learning algorithm and non-dominated sorting genetic algorithm (NSGA-II). In a first step, BP algorithm is used to optimize the parameters of the model (parameters of membership functions and fuzzy rules). NSGA-II is used in a second phase, to optimize the number of fuzzy rules and to fine tune the parameters. A well known benchmark is used to evaluate performances of the proposed modelling approach, and compare it with other modelling approaches.

This paper deals with the optimization of fuzzy controllers using neural networks and hierarchical genetic algorithms. The method combines the training advantage of neural networks, and the aptitude to find a global optimum offered by genetic algorithms. The fuzzy controller is implemented as a neural network where each layer represents a part of the fuzzy controller. The training process consists in optimizing the connection weights which code the various parameters of the controller. Once the training is finished, the parameters coded chromosomes take part in the evolution process using selection, crossover and mutation. This hybridization is applied to nonlinear system.