基于IFSA-ELM的GIS局部放电故障诊断研究  

Research on GIS Partial Discharge Fault Diagnosis Based on IFSA-ELM

作　　者：周煜健 李欣港 林玥玥 ZHOU Yujian;LI Xingang;LIN Yueyue(China Energy Investment Corporation Beihai Power Generation Co.,Ltd.,Guangxi Beihai 536000,China;School of Electrical Engineering,Guangxi University,Guangxi Nanning 530000,China)

机构地区：[1]国能广投北海发电有限公司,广西北海536000 [2]广西大学电气工程学院,广西南宁530000

出　　处：《广西电力》2024年第4期10-16,25,共8页Guangxi Electric Power

基　　金：国能广投北海发电有限公司科技创新项目(202400881)。

摘　　要：气体绝缘开关设备(GIS)是电力系统中重要的高压设备,其可靠性对整个系统的安全运行至关重要。局部放电(PD)是GIS中常见的绝缘故障前兆,能够有效监测和诊断局部放电对于避免严重故障具有重要意义。本文提出了一种基于相位分辨局部放电(PRPD)图谱和改进鱼群算法-极限学习机(IFSA-ELM)的GIS局部放电故障诊断方法。首先,通过高频电流传感器(HFCT)采集GIS的局部放电信号,构建PRPD图谱,提取放电特征参数。然后,利用IFSA算法对ELM模型进行优化,调整ELM的输入权重和偏置,提高模型的泛化能力和诊断精度。实验结果表明,该方法能够准确识别不同类型的局部放电模式,其识别准确率高达97.5%,有效提升了GIS局部放电故障诊断的准确率,为电力系统的状态监测与故障预测提供了重要技术支持。Gas-insulated switchgear(GIS)is important high-voltage equipment in power system,and its reliability is crucial to the secure operation of the whole system.Partial discharge(PD)is a common precursor of insulation faults in GIS,and it is important to be able to effectively monitor and diagnose PD to avoid serious faults.In this study,a GIS partial discharge fault diagnosis method based on phase-resolved partial discharge(PRPD)mapping and Improved Fish Swarm Algorithm-Extreme Learning Machine(IFSA-ELM)is proposed.Firstly,the local discharge signals of GIS are collected by high-frequency current transducer(HFCT),and the PRPD maps are constructed to extract the characteristic parameters of the discharge.Then,the IFSA is used to optimize the ELM model,adjust the input weights and biases of ELM,and improve the generalization ability and diagnostic accuracy of the model.The experimental results show that the method can accurately recognize different types of partial discharge patterns,and its recognition accuracy is as high as 97.5%.It effectively improves the accuracy and reliability of GIS partial discharge fault diagnosis,and provides important technical support for power system condition monitoring and fault prediction.

关 键 词：气体绝缘开关设备(GIS) 局部放电(PD) 相位分辨局部放电(PRPD) 改进鱼群算法-极限学习机(IFSA-ELM) 故障诊断 

分 类 号：TM855[电气工程—高电压与绝缘技术]