Driving training‐based optimization (DTBO) for global maximum power point tracking for a photovoltaic system under partial shading condition

Abstract The presence of bypass diodes in photovoltaic (PV) arrays can mitigate the negative effects of partial shading conditions (PSCs), which can cause multiple peak characteristics at the output. However, conventional maximum power point tracking (MPPT) methods can develop errors and detect the local maximum power point (LMPP) instead of the global maximum power point (GMPP) under certain circumstances. To address this issue, several artificial intelligence (AI)‐based methods have been proposed, but they result in complicated and unreliable methodologies. This study introduces the driving training‐based optimization (DTBO) method, which aims to address the partial shading (PS) problem quickly and reliably in maximum power point (MPP) detection for PV systems. DTBO improves tracking speed and reduces fluctuations in the power output during the tracking period. The proposed method is extensively verified using the Typhoon hardware‐in‐the‐loop (HIL) 402 emulator and compared to conventional methods such as particle swarm optimization (PSO), and JAYA, as well as the recently proposed adaptive JAYA (AJAYA) method for MPPT in a PV system under similar conditions.