矿区光伏发电系统逆变器故障自动检测方法  

Automatic Detection Method for Inverter Fault in Photovoltaic Power Generation System in Mining Area

作　　者：马千里 MA Qianli(Huangling Mining New Energy Development Co.,Ltd.,Yan’an 727307,China)

机构地区：[1]黄陵矿业新能源开发有限公司,陕西延安727307

出　　处：《通信电源技术》2025年第4期199-202,共4页Telecom Power Technology

摘　　要：矿区环境中存在大量粉尘颗粒,这些粉尘容易在逆变器表面和内部堆积,影响逆变器的散热性能和电气连接,在故障自动检测中很难准确区分是正常的性能波动还是由粉尘引起的故障早期征兆。基于此,提出矿区光伏发电系统逆变器故障自动检测方法。先提取矿区光伏发电系统逆变器故障特征,然后构建一个时间卷积网络(Temporal Convolutional Network,TCN)模型,输入提取特征,学习故障信号特征序列中的时序依赖关系,输出逆变器故障类型自动检测结果。实验结果表明,设计方法自动检测矿区光伏发电系统逆变器故障时,在检测精度与检测效率方面均表现出良好的性能。There are a large number of dust particles in the mining environment.These dust particles are prone to accumulate on the surface and inside of the inverter,affecting its heat dissipation performance and electrical connections.It is difficult to accurately distinguish between normal performance fluctuations and early signs of faults caused by dust in automatic fault detection.Based on this,a method for automatic detection of inverter faults in photovoltaic power generation systems in mining areas is proposed.Firstly,extract the fault characteristics of the photovoltaic power generation system inverter in the mining area,and then construct a Temporal Convolutional Network(TCN)model.Input the extracted features,learn the temporal dependencies in the fault signal feature sequence,and output the automatic detection results of the inverter fault type.The experimental results show that the design method exhibits good performance in both detection accuracy and efficiency when automatically detecting inverter faults in mining photovoltaic power generation systems.

关 键 词：矿区 光伏发电系统 逆变器 故障检测 自动检测方法 

分 类 号：TM615[电气工程—电力系统及自动化] TM464