基于IGCSSA-SVM的变压器故障诊断
Fault Diagnosis Method of Transformer Based on IGCSSA-SVM
张珊珊1
作者信息
- 1. 安徽理工大学, 安徽 淮南 232001
- 折叠
摘要
针对变压器故障诊断精确度较低的问题,基于变压器油中溶解气体分析技术,本文提出一种改进麻雀搜索算法(IGCSSA)优化支持向量机(SVM)的变压器故障诊断模型.首先,利用精英反向和高斯柯西变异策略改进传统麻雀算法,然后利用改进麻雀搜索算法(IGCSSA)对SVM算法优化.最后对比麻雀算法-支持向量机(SSA-SVM),灰狼算法-支持向量机(GWO-SVM),粒子群算法-支持向量机(PSO-SVM)诊断结果.测试结果表明,IGCSSA-SVM模型准确率达到了 94%,相比SSA-SVM、GWO-SVM和PSO-SVM模型分别提高了 9%、19%和 18%,能够有效提高诊断准确率.
Abstract
In response to the problem of low accuracy in transformer fault diagnosis,based on dissolved gas analysis technology in trans-former oil,this paper proposes an improved sparrow search algorithm(IGCSSA)optimized support vector machine(SVM)transformer fault diagnosis model.Firstly,the traditional Sparrow algorithm is improved using elite reverse and Gaussian Cauchy mutation strategies,and then the SVM algorithm is optimized using the Improved Sparrow Search Algorithm(IGCSSA).Finally,compare the diagnostic results of Sparrow Algorithm Support Vector Machine(SSA-SVM),Grey Wolf Algorithm Support Vector Machine(GWO-SVM),and Particle Swarm Optimization Support Vector Machine(PSO-SVM).The test results show that the IGCSSA-SVM model has an accuracy of 94%,which is 9%,19%,and 18%higher than SSA-SVM,GWO-SVM,and PSO-SVM models,respectively,effectively improving diagnos-tic accuracy.
关键词
变压器/故障诊断/支持向量机/麻雀搜索算法Key words
Transformer/Fault diagnosis/Support vector machine/Sparrow search algorithm引用本文复制引用
出版年
2024
NETL
NSTL
维普
万方数据