基于三比值特征量与改进鸡群算法优化支持向量机的变压器故障模型
|
摘要
提出一种基于三比值特征量与改进鸡群算法优化支持向量机变压器故障诊断方法。以油中溶解气体(DGA)三比值作为特征量,利用改进鸡群算法对支持向量机参数进行优化,从而构建基于SVM的变压器故障诊断模型。基于国内117组变压器故障数据的故障诊断实例表明,以三比值为特征量的ICSO-SVM的诊断准确率高于IEC三比值法、标准SVM法、PSO-SVM和以DGA全数据为特征量的ICSO-SVM,证明了本文所提方法能提高变压器故障诊断准确率。
A transformer fault diagnosis method based on the three DGA ratios and chicken swarm optimization(ICSO)support vector machine(SVM) is proposed. The DGA three ratiosis used as the feature. Then, the improvedchicken swarm optimization algorithmis used to optimize the parameters of SVM. Finally,a transformer fault diagnosis model based on SVM is constructed.The resultbased on 117 sets of transformer fault data in China shows that the accuracy of the ICSO-SVM with the three ratiosas the feature is higher than that of the IEC three-ratio method, standard SVM, PSO-SVM,and ICSO-SVM with DGA full data as the feature.It is proved that the proposed method can improve the accuracy of transformer fault diagnosis.
A transformer fault diagnosis method based on the three DGA ratios and chicken swarm optimization(ICSO)support vector machine(SVM) is proposed. The DGA three ratiosis used as the feature. Then, the improvedchicken swarm optimization algorithmis used to optimize the parameters of SVM. Finally,a transformer fault diagnosis model based on SVM is constructed.The resultbased on 117 sets of transformer fault data in China shows that the accuracy of the ICSO-SVM with the three ratiosas the feature is higher than that of the IEC three-ratio method, standard SVM, PSO-SVM,and ICSO-SVM with DGA full data as the feature.It is proved that the proposed method can improve the accuracy of transformer fault diagnosis.
引文
[1]刘勇辉.基于DGA的电力变压器故障诊断[J].高电压技术,2005(6):87-88.
[2]汪可,李金忠,张书琦,等.变压器故障诊断用油中溶解气体新特征参量[J].中国电机工程学报,2016,36(23):6570-6578.
[3]IEEE C57.104-1991,IEEE guide for the interpretation of gases generated in oil-immersed transformers[S].
[4]DUVAL M,DEPABLA A.Interpretation of gas-inoil analysis using new IEC publication 60599 and IEC TC 10 databases[J].IEEE Electrical Insulation Magazine,2001,17(2):31-41.
[5]陈金强.基于RIMER专家系统和DGA的变压器故障诊断[J].高压电器,2013,49(11):76-81.
[6]张珂斐,郭江,肖志怀.模糊综合评价和改进证据理论相融合的变压器状态评价方法研究[J].武汉大学学报:工学版,2016,49(2):279-284,314.
[7]张珂斐,郭江,聂德鑫,等.基于化学反应优化神经网络与融合DGA算法的油浸式变压器模型研究[J].高电压技术,2016(4):1275-1281.
[8]郑含博,王伟,李晓纲,等.基于多分类最小二乘支持向量机和改进粒子群优化算法的电力变压器故障诊断方法[J].高电压技术,2014,40(11):3424-3429.
[9]李振璧,王康,姜媛媛.基于模拟退火的改进鸡群优化算法[J].微电子学与计算机,2017,34(2):30-33,38.
[2]汪可,李金忠,张书琦,等.变压器故障诊断用油中溶解气体新特征参量[J].中国电机工程学报,2016,36(23):6570-6578.
[3]IEEE C57.104-1991,IEEE guide for the interpretation of gases generated in oil-immersed transformers[S].
[4]DUVAL M,DEPABLA A.Interpretation of gas-inoil analysis using new IEC publication 60599 and IEC TC 10 databases[J].IEEE Electrical Insulation Magazine,2001,17(2):31-41.
[5]陈金强.基于RIMER专家系统和DGA的变压器故障诊断[J].高压电器,2013,49(11):76-81.
[6]张珂斐,郭江,肖志怀.模糊综合评价和改进证据理论相融合的变压器状态评价方法研究[J].武汉大学学报:工学版,2016,49(2):279-284,314.
[7]张珂斐,郭江,聂德鑫,等.基于化学反应优化神经网络与融合DGA算法的油浸式变压器模型研究[J].高电压技术,2016(4):1275-1281.
[8]郑含博,王伟,李晓纲,等.基于多分类最小二乘支持向量机和改进粒子群优化算法的电力变压器故障诊断方法[J].高电压技术,2014,40(11):3424-3429.
[9]李振璧,王康,姜媛媛.基于模拟退火的改进鸡群优化算法[J].微电子学与计算机,2017,34(2):30-33,38.