基于KPCA-ISSA-KELM的光伏阵列故障诊断方法  

Fault Diagnosis Method of PV Array Based on KPCA-ISSA-KELM

作　　者：武文栋 施保华[1,3] 郑传良 郭茜婷 陈峥 WU Wendong;SHI Baohua;ZHENG Chuanliang;GUO Qianting;CHEN Zheng(College of Electrical Engineering&New Energy,China Three Gorges University,Yichang 443002,China;State Grid Fujian Ningde Power Supply Company,Ningde 352100,China;Hubei Microgrid Engineering Technology Research Center,Yichang 443002,China)

机构地区：[1]三峡大学电气与新能源学院,湖北宜昌443002 [2]国网福建宁德供电公司,福建宁德352100 [3]湖北省微电网工程技术研究中心,湖北宜昌443002

出　　处：《智慧电力》2022年第11期69-76,共8页Smart Power

基　　金：国家自然科学基金资助项目(52007103);国网福建省电力有限公司科技项目(52139020006V)。

摘　　要：为提高光伏阵列故障诊断的精度,提出一种基于核主成分分析(KPCA)和改进麻雀搜索算法(ISSA)优化核极限学习机(KELM)的光伏故障诊断方法。利用KPCA降维提取故障数据的非线性特征,减少外界条件产生的冗余数据,有效提高复杂故障识别准确率。通过融入Levy飞行和自适应权重t对麻雀搜索算法进行改进,并利用ISSA对KELM中的核参数γ和正则化系数C进行优化,建立了基于KPCA-ISSA-KELM的光伏阵列故障诊断模型。实验结果表明,经ISSA优化KELM的光伏阵列故障诊断模型与其他光伏阵列诊断模型相比,在故障诊断精度上达到97%,验证了该模型的准确性和有效性。In order to improve the accuracy of photovoltaic array fault diagnosis,a novel photovoltaic fault diagnosis method based on kernel principal component analysis and improved sparrow search algorithm is proposed to optimize the kernel extreme learning machine.KPCA is used to extract nonlinear features of fault data and can reduce redundant data generated by external conditions,and effectively improve the accuracy of complex fault identification.In addition,the ISSA is obtained by incorporating Levy flight and adaptive weight t,and the ISSA is used to optimize the kernel function parameterγand regularization coefficient C of the KELM,and the PV array fault diagnosis model based on KPCA-ISSA-KELM is established.Experimental results show that KELM’s fault diagnosis model optimized by ISSA has a fault diagnosis accuracy of 97%compared with other PV array diagnosis models,which verifies the accuracy and validity of the proposed model.

关 键 词：光伏阵列 核主成分分析 核极限学习机 改进麻雀搜索算法 故障诊断 

分 类 号：TM615[电气工程—电力系统及自动化]