基于改进QPSO算法的光伏发电最大功率点跟踪  

Maximum power point tracking for photovoltaic generation based on improved QPSO algorithm

作　　者：方胜利[1] 杨峰 朱晓亮 马春艳[1] 侯贸军[1] FANG Shengli;YANG Feng;ZHU Xiaoliang;MA Chunyan;HOU Maojun(College of Electrical and Information Engineering,Hubei University of Automotive Technology,Shiyan 442002,China;Shiyan Juneng Power Design Co.,Ltd.,Shiyan 442000,China)

机构地区：[1]湖北汽车工业学院电气与信息工程学院,湖北十堰442002 [2]十堰巨能电力设计有限公司,湖北十堰442000

出　　处：《安徽大学学报（自然科学版）》2023年第4期57-66,共10页Journal of Anhui University(Natural Science Edition)

基　　金：湖北省教育厅科学技术研究中青年人才基金资助项目(Q20171802)。

摘　　要：光伏阵列输出在不同工况下具有单峰或多峰特性.针对因最大功率点跟踪(maximum power point tracking,简称MPPT)精度不高、跟踪时间较长而导致光伏发电效率低下的问题,提出一种改进的量子粒子群优化(quantum particle swarm optimization,简称QPSO)算法.采用Logistic混沌映射初始化粒子种群;在种群进化前期将反向学习策略引入惯性权重自适应调整的量子粒子群优化(dynamically changing weights quantum-behaved particle swarm optimization,简称DCWQPSO),扩大种群搜索范围,提高种群的全局搜索能力;在种群进化后期将模拟退火机制引入DCWQPSO,提高种群收敛速度,并对粒子群进行柯西变异,增强粒子的多样性,提升局部搜索能力.Matlab仿真结果表明:相对其他4种算法,该文提出的改进QPSO算法的跟踪时间更短、跟踪精度更高.因此,该文算法具有优越性.The output of photovoltaic array has the characteristics of single peak or multi-peak under different working conditions.Aiming at the low efficiency of photovoltaic power generation due to the low precision and long tracking time of maximum power point tracking(MPPT),an improved quantum particle swarm optimization(QPSO)algorithm was proposed.The particle population was initialized by Logistic chaotic mapping.The reverse learning strategy was introduced into the dynamically changing weights quantum-behaved particle swarm optimization(DCWQPSO)to expand the search range of the population and improve the global search ability of the population in the early stage of population evolution.The simulated annealing mechanism was introduced into the DCWQPSO algorithm to improve the convergence rate of the population and the Cauchy mutation was carried out to increase particle diversity and enhance the local search ability in the later stage of population evolution.The Matlab simulation results showed that the improved QPSO algorithm proposed in this paper had shorter tracking time and higher tracking accuracy compared with the other four algorithms.Therefore,the algorithm proposed in this paper was superior.

关 键 词：最大功率点跟踪 改进量子粒子群优化 LOGISTIC混沌映射 反向学习策略 模拟退火 柯西变异 

分 类 号：TM933[电气工程—电力电子与电力传动]