Improved particle swarm optimization for global maximum power point tracking of partially shaded PV array

Abstract

This paper presents an improved particle swarm optimization (PSO) algorithm for determining the global maximum power point tracking (GMPP) of photovoltaic (PV) array under partially shaded conditions (PSC). Under PSC, the power–voltage characteristics have a more complex shape with several local peaks and one global peak. Most of the conventional techniques that are applied in the maximum power-tracking control unit of PV stations do not provide reliable tracking of the GMPP under PSC, which leads to decrease the reliability and the performance of the PV power plant. The performances of the proposed PSO algorithm and the conventional perturb and observe algorithms are evaluated using simulations in MATLAB/Simulink. Eight different partial shading patterns have been selected to prove the robustness of the proposed algorithm. A (step-up) DC–DC boost converter is interfaced with the proposed model. The results indicate that the modified PSO algorithm can very fast track the GMPP within 150–280 ms for different shading conditions furthermore the quality of the tracked power is very high as compared with the previous studies in the literature. Also, the average tracking efficiency of the proposed PSO algorithm is higher than 99.8%, which provides good prospects to apply this algorithm in the control search unit for the GMPP in PV stations.