改进型卷尾猴搜索算法及光伏电池参数辨识  

Improved capuchin search algorithm and its application to photovoltaic cell parameter identification

作　　者：田小情 张著洪 TIAN Xiaoqing;ZHANG Zhuhong(College of Big Data and Information Engineering,Guizhou University,Guiyang 550025,China)

机构地区：[1]贵州大学大数据与信息工程学院,贵阳550025

出　　处：《智能计算机与应用》2023年第4期135-141,146,共8页Intelligent Computer and Applications

基　　金：国家自然科学基金(62063002)。

摘　　要：针对新型卷尾猴搜索算法存在初始卷尾猴分布、静态惯性系数及跟随者位置更新策略制约其全局开采与局部勘测能力的问题,探讨改进型算法及其在光伏电池参数辨识中的应用。算法设计中,利用Logistic混沌映射初始化卷尾猴种群;引入灰狼优化的捕食策略和柯西变异扩大卷尾猴觅食范围,且借助S型函数自适应调节惯性权重,增强全局搜索能力。模型设计中,将光伏电池双二极管结构模型扩展为四二极管模型,进而获得多参数待定的二极管参数辨识模型。比较性的数值实验表明,改进型卷尾猴搜索算法求解基准函数优化及参数辨识问题时,在最优解的搜索能力、寻优效率以及参数辨识效果等方面具有明显优势,且对复杂优化问题的解决具有较好潜力。For the problem that the global exploration and local exploitation capabilities of the new-type capuchin search algorithm are greatly influenced by initial capuchin distribution,static inertia weight,and each capuchin follower′s position update strategy,this work develops an improved capuchin search algorithm,while exploring its application to solar diode identification.In the design of the algorithm,the capuchin population is initialized by the Logistic chaotic mapping,and later,the predation strategy of gray wolf optimization and the conventional Cauchy mutation operator are adopted to enlarge the capuchin′s predation scope.To avoid to come to stagnation,the algorithm adopts the S-shaped function to ensure that the static inertia weight changes adaptively.The solar double-diode model is extended into a multi-parameter four-diode one,furtherly the multi-parameter undetermined diode parameter identification model is obtained.The numerically comparative results have validated that,when solving the problems of benchmark function optimization and parameter identification,the algorithm has a significant advantage over the original capuchin search algorithm and other compared approaches in the aspects of solution quality,efficiency and parameter identification effect,and meanwhile is potential to complex optimization problems.

关 键 词：卷尾猴搜索算法 S型惯性权重 柯西变异 四二极管模型 参数辨识 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]