Hybrid Method for Maximum Power Point Tracking in Photovoltaic Systems

抄録

Partial shading and prolonged search time for determining maximum power point (MPP) are primary factors reducing power conversion efficiency in photovoltaic (PV) systems. This paperproposes a hybrid method that combines an improved horse racing algorithm (HRA) for global search with a gradient descent (GD) algorithm for precise MPP identification within the localized region. Simulation results demonstrate that the proposed approach achieves three key objectives: accurately location of the MPP; elimination of the steady-state oscillation amplitude; and significant reduction in the convergence time. Comparative analysis with incremental conductance (InC), perturbation and observation (P&O) methods confirms the superior performance of the hybrid approach under both partial shading and uniform irradiance conditions.