Application of a Chaotic Quantum Bee Colony and Support Vector Regression to Multipeak Maximum Power Point Tracking Control Method Under Partial Shading Conditions

Abstract

In view of the multipeak characteristics of a photovoltaic (PV) array P–V curve under local shadow conditions and that the traditional maximum power point tracking (MPPT) algorithm cannot effectively track the maximum power point of the curve, a multipeak MPPT algorithm based on a chaotic quantum bee colony and support vector regression (SVR) is proposed. By constructing and analyzing the mathematical model of a photovoltaic array under a local shadow, the P–V characteristic equation of the photovoltaic array is obtained. The improved strategy of the artificial bee colony algorithm is studied, and the improved chaotic quantum bee colony algorithm (CQABC) is applied to the optimization of SVR parameters; this application improves the accuracy and generalization performance of the maximum power point prediction model based on SVR. The calculation process of the multipeak MPPT algorithm based on CQABC-SVR is given, and the effectiveness of the algorithm is verified by simulation and testing. The experimental results show that the algorithm can accurately track the global maximum power point under uniform illumination or local shadow conditions, effectively overcoming the problem of traditional MPPT algorithms easily falling into local extrema.