$V_{\text{cp}}$-Decomposition-Based Adaptive Search for Optimal Power Extraction in Photovoltaic Systems

Maintaining optimal power in changing environments presents a significant challenge for photovoltaic systems. This article introduces a novel
$V_{\text{cp}}$
decomposition-based adaptive strategy (VD-AS) aimed at enhancing optimal power extraction in varying environmental conditions. Distinct from the conventional maximum power point tracking (MPPT) technologies, the proposed VD-AS method estimates the peak pattern based on the voltage at the current peak on the power-voltage curve. Utilizing the peak pattern information, a sim-to-real peak identification model is capable of predicting the global maximum power point (GMPP) locus. It not only enables an accurate adaptive search for the GMPP locus, but also substantially minimizes the need for unnecessary global searches when the peak pattern is consistent. Experimental results from a laboratory prototype demonstrate that the proposed method can accurately track the GMPP with a smaller voltage search range and shorter tracking time. Its tracking efficiency is generally 10% higher than that of state-of-the-art GMPP tracking methods.