Conception and Hardware Implementation of MPPT Controller for Partially Shaded Photovoltaic Panels using Backstepping and Neural Network based Particle Swarm Optimization

Abstract

In this paper, in order to track the maximum power point under partial shading condition, An Artificial Neural Network based Particle Swarm Optimization (ANNPSO) combined with backstepping controller has been employed. The proposed controller relay to the ability of the PSO to find the global maximum power point and to the ANN to generate a reference voltage according to the PSO data without oscillations. The backstepping controller has been employed in order to track the reference voltage generated by the ANN by adjusting the duty cycle of the SEPIC converter. The simulation was carried out using MATLAB software. On one hand, the results affirms the ability of the proposed controller to find and to track the global maximum power point, and on the other hand the present method demonstrates high efficiency against previous works, The results shows that the proposed method tracks the reference voltage within 20 ms and it is able to tracks the GMPP with high performance. Furthermore, the experimental study shows that the proposed controller can be easily implemented using low coast materials, and the obtained results show a gain of 5W compared to the controller that combine the incremental conductance with backstepping.