Abstract
The power against voltage curve for PV power system during shadow condition contains number of local maximum power points (MPPs) and only one global. The classical maximum power point tracking (MPPT) algorithms are designed to follow the global MPP, but they stuck around local MPPs such as fuzzy logic controller (FLC). Therefore, A global MPPT based on teaching-learning-based optimization (TLBO) algorithm has been presented in this paper. The performance of PV system under abnormal conditions such as partial shading has been improved. TLBO algorithm is simple computational steps and faster convergence to optimal solution. A comprehensive assessment of TLBO-based tracker is carried out against FLC and particle swarm optimization (PSO) techniques for same conditions. Six different partial shading patterns have been employed to investigate TLBO performance using MATLAB/Simulink. The parameters of comparison include the tracking speed and overall tracking efficiency. The results confirm that TLBO-based tracker exactly convergence to global MPP under different studied cases. TLBO has best performance compared to the other studied techniques. The tracking speed is increased using TLBO-based tracker; the average tracking time of global MPP is reduced by more than 23.8 % compared with PSO in all studied different partial shading patterns.