自适应果蝇算法多峰值光伏阵列最大功率点跟踪  

Maximum Power Point Tracking of Multi-peak PV Array Using Adaptive Fruit Fly Optimization Algorithm

作　　者：原琳[1] 程海军[1] 赵凤贤[1] YUAN Lin;CHENG Haijun;ZHAO Fengxian(School of Electrical Engineering,Liaoning University of Technology,Jinzhou 121001,China)

机构地区：[1]辽宁工业大学电气工程学院,锦州121001

出　　处：《电源学报》2020年第3期168-174,共7页Journal of Power Supply

基　　金：辽宁省教育厅资助项目(JQL201715408,JL201615407)。

摘　　要：光伏PV(photovoltaic)阵列在实际应用中常存在遮挡现象,与光照均匀时的单峰特性不同,遮挡情况下的PV输出曲线呈多峰特性,常规的最大功率点跟踪方法大多寻找到第1个峰值点即停止搜索,易使光伏阵列因陷入局部极值点的跟踪而失效。提出一种基于自适应的果蝇优化算法AFOA(adaptive fruit fly optimization algorithm),对原有果蝇算法的初始值设定及寻优步长进行改进,并定时与扰动观察法P&O(perturb&observe)相结合,增强寻优算法的实时性。通过Matlab仿真,分别在光照均匀和遮挡情况下,与扰动观察法和粒子群优化PSO(particle swarm optimization)算法的跟踪效果进行了比较,仿真结果表明,无论有无遮挡现象,AFOA算法都可准确跟踪到系统的全局最大功率点,提高了系统输出功率的稳定性及发电效率。In practical applications,there is often a shaded phenomenon for a photovoltaic(PV)array,and the cor-responding PV output curve has multi-peak characteristics,which is different from those under uniform illumination.Mo-st of the conventional maximum power point tracking(MPPT)methods will stop after they have found the first peak,which will probably cause the PV array to fall into the local extreme point and further fail.An adaptive fruit fly optimization algorithm(AFOA)is proposed to improve the initial value set by the original fruit fly algorithm and the optimal step size as well.In addition,the perturb&observe(P&O)method is combined to improve the algorithm’s real-time capability.Thro-ugh Matlab simulations,the tracking results using AFOA are compared with those using the P&O and particle swarm optimization(PSO)algorithms under uniform illumination and occlusion,respectively.Simulation results show that the AFOA algorithm can accurately track the global maximum power point with or without occlusion,thus improving the system’s stability of output power and the power generation efficiency.

关 键 词：遮挡 最大功率点跟踪 自适应果蝇算法 

分 类 号：TM615[电气工程—电力系统及自动化]