摘要
当前,物联网电力设备故障定位模型多为智能化结构,定位方式较为单一,最终得出的定位结果存在不可控偏差。为解决这一问题,文章开展基于BP神经网络算法的物联网电力设备故障定位方法研究。文章先进行数据预处理及设备故障特征提取,然后采用BP神经网络算法,提升整体的电力故障定位效率,最后构建BP神经网络测算电力设备故障定位模型,采用自适应间隔核验方式实现故障定位处理。测试结果表明:与传统低压脉冲电力设备故障定位方法、传统GRU电力设备故障定位方法相比,所提电力设备故障定位方法定位误判率较低,定位精准度更高,具有一定应用价值。
The current IoT power equipment fault location models are mostly intelligent structures with relatively single positioning methods,resulting in uncontrollable deviations in the final positioning results.To address this issue,the article conducts research on fault location methods for IoT power equipment based on BP neural network algorithm.The article first performs data preprocessing and equipment fault feature extraction,then uses the BP neural network algorithm to improve the overall efficiency of power fault localization.Finally,a BP neural network is constructed to calculate the power equipment fault localization model,and an adaptive interval verification method is used to achieve fault localization processing.The test results show that compared with traditional low-voltage pulse power equipment fault positioning methods and traditional GRU power equipment fault positioning methods,the proposed power equipment fault positioning method has a lower misjudgment rate and higher positioning accuracy,which has certain application value.
作者
张舜
程晓通
李琼
ZHANG Shun;CHENG Xiaotong;LI Qiong(State Grid Xinjiang Electric Power Co.,Ltd.,Bazhou Power Supply Company,Bazhoukuerle 841000,China)
出处
《无线互联科技》
2024年第19期86-89,共4页
Wireless Internet Science and Technology
关键词
BP神经网络
物联网
电力设备
故障识别
远程异常感应
BP neural network
Internet of Things
power equipment
fault identification
remote abnormal sensing
