线路故障信号监测分解提取及诊断建模分析  

Line fault signal monitoring decomposition extraction diagnosis modeling and analysis

作　　者：吕欣玥 张力强 LV Xinyue;ZHANG Liqiang(Xi'an Technological University,Xi'an 710021,China;Datong Power Supply Company,Shanxi Electric Power Company,State Grid,Datong 037008,Shanxi China)

机构地区：[1]西安工业大学,陕西西安710021 [2]国网山西省电力公司大同供电公司,山西大同037008

出　　处：《粘接》2024年第5期181-184,共4页Adhesion

摘　　要：针对传统电气线路故障检测存在的问题,提出了一种新的电气线路故障检测技术。对实测电气线路电流信号采用奇异值分解进行降噪处理,采用经验模态分解对降噪后的电流信号分解得到本征模态函数,运用排列熵提出本征模态函数所包含的故障特征。提取的故障特征作为极限学习机的输入,电气线路故障类型作为输出,构建了电气线路故障检测的极限学习机模型。将提出的电力线路故障检测模型和BP神经网络模型、YOLOv4模型进行对比,结果表明,提出的电气线路故障检测模型具有更高的检测效率和检测精度。In view of the problems existing in the traditional electrical circuit fault detection,a new fault detection technology for electrical lines was proposed.Singular value decomposition(SVD)was used to denoise the actual electric circuit current signal,and the intrinsic mode function was obtained by empirical mode decomposition of the denoised signal,and the fault features contained in the eigenmode function were proposed by using permutation entropy.Taking the extracted fault features as the input of the extreme learning machine and the fault type of electrical line as the output,an extreme learning machine model of electrical line fault detection was constructed.The proposed power line fault detection model was compared with BP neural network model and YOLOv4 model,and the results showed that the proposed electrical line fault detection model had higher detection efficiency and detection accuracy.

关 键 词：电气线路 故障检测 故障特征提取 极限学习机 

分 类 号：TM726[电气工程—电力系统及自动化] TP391[自动化与计算机技术—计算机应用技术]