摘要
光伏阵列长期暴露在恶劣的环境中,导致光伏组件易发生故障,从而影响光伏阵列的发电效率。在实际运行过程中,光伏阵列除发生单一故障之外,还会出现多类型的复合故障,给故障诊断加大了难度。提出了一种基于哈里斯鹰(HHO)算法优化极限学习机(ELM)的光伏阵列多类型复合故障诊断方法。用HHO算法优化ELM的权值和阈值,建立HHO-ELM故障诊断模型,并与ELM、粒子群优化算法(PSO)-ELM、正余弦优化算法(SCA)-ELM以及鲸鱼优化算法(WOA)-ELM算法进行对比。实验结果表明,对于复合故障类型,HHO-ELM模型具有更高的诊断准确率,提高了光伏阵列复合故障的识别精度。
The photovoltaic array is exposed to harsh environment for a long time,leading to easy failure of the photovoltaic array,which affects the power generation efficiency of the photovoltaic array.In the actual operation process,the photovoltaic arrays can not only have a single fault,but also multiple types of compound faults,which increases the difficulty of fault diagnosis.A multi-type composite fault diagnosis method for PV arrays based on the harris hawks optimization(HHO)algorithm optimized the extreme learning machine(ELM)was proposed.The weights and thresholds of ELM were optimized by HHO algorithm,the fault diagnosis model of HHO-ELM was established,and compared with ELM,PSO-ELM,SCA-ELM and WOA-ELM algorithm.The experimental results show that the HHO-ELM model has higher diagnosis accuracy for the composite faults,and improves the identification accuracy of the photovoltaic composite faults.
作者
钱亮
黄伟
杨建卫
QIAN Liang;HUANG Wei;YANG Jianwei(School of Automation Engineering,Shanghai University of Electric Power,Shanghai 200090,China;Zhongdian Huachuang(Suzhou)Electric Power Technology Research Co.,Ltd.,Suzhou Jiangsu 215123,China;Zhongdian Huachuang Electric Power Technology Research Co.,Ltd.,Shanghai 200086,China)
出处
《电源技术》
CAS
北大核心
2024年第2期345-350,共6页
Chinese Journal of Power Sources
基金
国家电网公司华东分部科技项目(H2021-111)。
关键词
光伏阵列
故障诊断
哈里斯鹰算法
极限学习机
多类型复合故障
photovoltaic array
fault diagnosis
harris hawks optimization algorithm
extreme learning machine
multiple types of complex faults
