Maximum Power Point Tracking of Photovoltaic Generation System Using Improved Quantum-Behavior Particle Swarm Optimization

Abstract

This study introduces an improved quantum-behavior particle swarm optimization (IQPSO), tailored for the task of maximum power point tracking (MPPT) within photovoltaic generation systems (PGSs). The power stage of the MPPT system comprises a series of buck-boost converters, while the control stage contains a microprocessor executing the biomimetic algorithm. Leveraging the series buck-boost converter, the MPPT system achieves optimal operation at the maximum power point under both ideal ambient conditions and partial shade conditions (PSCs). The proposed IQPSO addresses the premature convergence issue of QPSO, enhancing tracking accuracy and reducing tracking time by estimating the maximum power point and adjusting the probability distribution. Employing exponential decay, IQPSO facilitates a reduction in tracking time, consequently enhancing convergence efficiency and search capability. Through single-peak experiments, multi-peak experiments, irradiance-changing experiments, and full-day experiments, it is demonstrated that the tracking accuracy and tracking time of IQPSO outperform existing biomimetic algorithms, such as the QPSO, firefly algorithm (FA), and PSO.