Application of partial shading effect on mppt for photovoltaic systems in different ambient conditions with intelligent algorithms

Maximum power point tracking (MPPT) aims to increase the efficiency of Photovoltaic (PV) systems by enabling PV panels to operate at maximum power point. For this purpose, many methods have been developed for MPPT in the literature, including traditional, intelligent and hybrid. One of the traditional methods of Perturb & Observe (P&O) performs well in MPPT under equal irradiation conditions. However, in partial shading conditions, it cannot find the global maximum power point between multiple peaks, it sticks to the local maximum power points and fails. As an alternative to traditional methods, metaheuristic methods such as Particle Swarm Optimization (PSO) and Cuckoo Search Optimization (CSO) algorithms have been developed. Metaheuristic methods can overcome the problems of capturing local maximum power points because their formulations contain probabilistic parameters that allow them to escape from local maximum power points. For this reason, these algorithms provide superior performance compared to traditional methods. In this thesis, a model consisting of PV array, DC-DC boost converter and MPPT algorithms has been developed in MATLAB / SIMULINK. In this model, MPPT operation has been performed with P&O, PSO and CSO algorithms under partial shading conditions and compared in terms of tracking speed and accuracy of these algorithms. As a result, PSO and CSO algorithms were successful in finding the global maximum power point, while D&G algorithm stuck to a local point in other configurations except for a single shading configuration. In addition, it has been seen that CSO algorithm reaches global maximum power point faster than PSO algorithm.