摘要
为改善综合智能化光缆通信线路故障诊断的识别准确率,提出了一种使用灰狼算法优化支持向量机的综合智能化光缆通信线路故障诊断模型。首先,提取光缆线路故障信号的时域特征、频域特征和样本熵特征,之后建立综合智能化光缆通信线路故障的GWO-SVM诊断模型。与GA-SVM、PSO-SVM和DE-SVM对比发现,使用GWO-SVM模型的综合智能化光缆通信线路故障诊断精度更高和收敛速度更快,从而为综合智能化光缆通信线路故障诊断提供了新的方法。
In order to improve the accuracy of integrated intelligent optical cable line fault diagnosis,an integrated and intelligent fault diagnosis model for optical cable lines is proposed based on the grey Wolf Algorithm to optimize the Support vector machine.Firstly,the time-domain,frequency-domain and sample entropy features of optical cable fault signals are extracted,and then an intelligent GWO-SVM diagnosis model of optical cable fault is established.Compared with GA-SVM,PSO-SVM and DE-SVM,GWO-SVM has higher diagnostic accuracy and faster convergence speed in fault diagnosis of optical cable lines,it provides a new method for comprehensive and intelligent fault diagnosis of optical cable lines.
作者
方明
Fang Ming(Guiyang Bureau,EHV Transmission Company,China Southern Power Grid Co.,Ltd.,Guiyang 550081,China)
出处
《现代科学仪器》
2022年第2期222-227,共6页
Modern Scientific Instruments
关键词
支持向量机
灰狼优化算法
光缆线路故障
时域特征
频域特征
support vector machine
grey wolf optimization algorithm
cable fault
time domain feature
frequency domain feature
