基于改进自适应粒子群算法的光伏电池参数识别

Parameters identification of photovoltaic cells based on an improved adaptive particle swarm algorithm

为解决传统粒子群算法辨识光伏电池参数易陷入局部极值及辨识准确性低的问题,通过分析单二极管光伏电池电路拓扑,提出一种改进自适应粒子群优化算法的参数辨识模型,采用自适应策略调节传统粒子群算法中的惯性权重因子,引入异步学习因子平衡全局极值和局部极值之间的搜索关系,利用所构建的自适应粒子群优化模型,识别三种不同类型光伏电池的参数,与传统粒子群参数辨识方法进行对比分析。结果表明,所提改进自适应粒子群算法在参数辨识中具有更高的精度,平均相对误差为1.109%~2.505%,总体误差均在3%以下,验证了所提自适应粒子群算法在光伏电池参数辨识中的可行性和有效性。

This paper proposes a parameter identification model of an improved adaptive particle swarm optimization algorithm by analyzing the single-diode photovoltaic cell circuit topology,which is designed to address the low identification accuracy caused by traditional particle swarm algorithm by falling into local extreme value easily.The study involves adjusting the inertial weight factor in traditional particle swarm algorithm with adaptive strategy;introducing the asynchronous learning factor to balance the searching relationship between the global maxima and local maxima;identifying the parameters of the three different types of photovoltaic cells by using the constructed adaptive particle swarm optimization model;comparing and analyzing the obtained parameters with traditional particle swarm parameter identification method.The results show that the proposed improved adaptive particle swarm algorithm has higher accuracy in the process of parameter identification,the average relative error is 1.109%-2.505%,and the overall error is below 3%.The study verifies the feasibility and effectiveness of the proposed adaptive particle swarm algorithm in photovoltaic cell parameter identification.