一种基于卷积神经网络和长短期记忆网络的光伏系统故障辨识方法  

A photovoltaic system fault identification method based on convolutional neural network and long short-term memory network

作　　者：涂彦昭 高伟[1] 杨耿杰[1] TU Yanzhao;GAO Wei;YANG Gengjie(College of Electrical Engineering and Automation,Fuzhou University,Fuzhou 350108)

机构地区：[1]福州大学电气工程与自动化学院,福州350108

出　　处：《电气技术》2022年第2期48-54,共7页Electrical Engineering

基　　金：福建省自然科学基金资助项目(2021J01633)。

摘　　要：随着光伏发电装机容量的不断上升,如何及时检测并解决光伏组件故障和异常,减少组件能量损失,提高光伏系统的发电效率成为一项重要任务。本文通过研究光伏阵列处于不同故障状态下的I-V曲线之间的特征差异性,直接以I-V曲线作为故障诊断的输入量,提出一种融合卷积神经网络(CNN)与长短期记忆(LSTM)网络的光伏系统故障辨识方法。实验结果表明,该方法不仅能识别出单一故障,如短路、遮阴、老化等,而且能有效识别出双重故障同时存在的情况。As the installed capacity of photovoltaic power generation continues to rise,how to detect and solve the faults and abnormalities of the photovoltaic modules in time to reduce energy loss and improve the power generation efficiency of photovoltaic systems has become a significant task.The characteristic differences between the I-V curves of photovoltaic arrays in different fault states are studied in this paper.The I-V curves are directly used as the input for fault diagnosis.On these grounds,a photovoltaic system fault identification method based on convolutional neural network(CNN)and long short-term memory(LSTM)network is proposed in this paper.Experimental results show that this method can not only identify single faults like short circuit,partial shading,abnormal aging and so on,but also effectively identify the simultaneous existence of hybrid faults.

关 键 词：光伏系统 故障诊断 I-V曲线 卷积神经网络(CNN) 长短期记忆(LSTM)网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TM615[自动化与计算机技术—控制科学与工程]