开关柜局部放电模式识别
|
摘要
针对高压开关柜局部放电模式分类中样本数较少,常规的分类方法识别率较低,提出了一种基于改进果蝇算法(FOA)优化的概率神经网络模型(PNN)的局部放电模式识别方法。作为一种新型的群体搜索随机优化算法,FOA算法具有原理简单、参数少,收敛快等优点。实验采集的局部放电信号进行处理并归一化,作为FOA-PNN神经网络的输入样本进行识别,并与传统算法进行比较。仿真实验证明:该算法可以有效地运用到局部放电模式识别中,为开关柜局部放电诊断供了一条新途径,具有良好的研究价值和发展前景。
In view of the small number of samples and the low recognition rate of conventional classification methods in high-voltage switchgear partial discharge pattern classification,this paper proposes a method of partial discharge pattern recognition based on improved Drosophila algorithm(FOA)optimized probabilistic neural network(PNN)model.As a new group search stochastic optimization algorithm,FOA algorithm has the advantages of simple principle,few parameters and fast convergence.The experimental partial discharge signals are processed and normalized,and identified as input samples of FOA-PNN neural network,and compared with traditional algorithms.The simulation results show that the algorithm can be effectively applied to partial discharge pattern recognition,which provides a new way for the diagnosis of partial discharge in switchgear,and has good research value and development prospects.
In view of the small number of samples and the low recognition rate of conventional classification methods in high-voltage switchgear partial discharge pattern classification,this paper proposes a method of partial discharge pattern recognition based on improved Drosophila algorithm(FOA)optimized probabilistic neural network(PNN)model.As a new group search stochastic optimization algorithm,FOA algorithm has the advantages of simple principle,few parameters and fast convergence.The experimental partial discharge signals are processed and normalized,and identified as input samples of FOA-PNN neural network,and compared with traditional algorithms.The simulation results show that the algorithm can be effectively applied to partial discharge pattern recognition,which provides a new way for the diagnosis of partial discharge in switchgear,and has good research value and development prospects.
引文
[1]朱名强.高压开关柜局部放电检测技术研究[J].科技创新与应用,2019(3):145-146.
[2]余力行.一起应用PD Monitor检测开关柜局放的分析和处理[J].电工技术,2018(10):54-55,58.
[3]张显龙,李通.高压开关柜局放检测技术的现场应用[J].电工电气,2018(5):59-63.
[4]李正明,钱露先,李加彬.基于统计特征与概率神经网络的变压器局部放电类型识别[J].电力系统保护与控制,2018,46(13):55-60.
[5]刘通,薛永刚,赵煦,等.变压器内单一局放PRPD谱图统计参量的主成分分析[J].南方电网技术,2014,8(5):33-37.
[6]黄新波,宋桐,王娅娜,等.基于IGSO优化LM网络的变压器故障诊断方法[J].中国电力,2014,47(9):60-65.
[7]王震宇,黄海波,张兵,等.基于故障树理论的高压开关柜状态检修[J].上海电力学院学报,2014,30(Suppl 1):21-24,31.
[8]宋人杰,杨爽,孟凡奇.基于多核学习整合GIS局部放电多类特征的分类研究[J].北京交通大学学报,2018,42(5):93-98,140.
[9]朱沛恒.基于果蝇算法优化的概率神经网络在变压器故障诊断中的应用[J].电力大数据,2018,21(6):37-43.
[10]吴小文,李擎.果蝇算法和5种群智能算法的寻优性能研究[J].火力与指挥控制,2013,38(4):17-20,25.
[11]毛长庚,梁伟峰.开关柜带电检测及局放定位技术的应用[J].大众用电,2017,32(9):26-27.
[12]林伟,王昕,郑益慧,等.基于概率神经网络的局部放电信号特征提取与识别研究[J].电气自动化,2018,40(3):60-63.
[13]任堂正,孙俊,沈晓峰,等.基于声电联合定位法的开关柜局部放电定位分析[J].电力与能源,2018,39(2):151-154.
[14]赵广振,张翠肖,武辉林,等.基于主成分分析和概率神经网络的入侵检测方法[J].石家庄铁道大学学报(自然科学版),2018,31(1):91-95.
[15]胡浩.一种基于概率神经网络建立电气系统故障的分类方法[J].科技创新与应用,2018(8):11-13.
[16]LI F H,LI R,TIAN L L,et al.Data-driven timefrequency analysis method based on variational mode decomposition and its application to gear fault diagnosis in variable working conditions[J].Mech Syst Signal Process,2019,116:462-479.
[17]李宗岳,陈志军,李名远.基于改进FOA-LSSVM的变压器故障诊断[J].水电能源科学,2016,34(4):194-197.
[18]CHEN A Y,THOMS R,KURFESS.Signal processing techniques for rolling element bearing spall size estimation[J].Mech Syst Signal Process,2019,117:16-32.
[19]贾勇勇,邓敏,李玉杰,等.基于深度残差网络的GIS局部放电模式识别技术研究[J].高压电器,2018,54(11):123-129.
[20]万书亭,豆龙江,李聪,等.基于VMD和样本熵的高压断路器故障特征提取及分类[J].振动与冲击,2018,37(20):32-38.
[21]张雪英,刘秀丽,栾忠权.基于峭度准则VMD样本熵与PNN的齿轮故障诊断[J].组合机床与自动化加工技术,2018(9):61-64.
[22]杨大为,赵永东,冯辅周,等.基于参数优化变分模态分解和多尺度熵偏均值的行星变速箱故障特征提取[J].兵工学报,2018,39(9):1683-1691.
[2]余力行.一起应用PD Monitor检测开关柜局放的分析和处理[J].电工技术,2018(10):54-55,58.
[3]张显龙,李通.高压开关柜局放检测技术的现场应用[J].电工电气,2018(5):59-63.
[4]李正明,钱露先,李加彬.基于统计特征与概率神经网络的变压器局部放电类型识别[J].电力系统保护与控制,2018,46(13):55-60.
[5]刘通,薛永刚,赵煦,等.变压器内单一局放PRPD谱图统计参量的主成分分析[J].南方电网技术,2014,8(5):33-37.
[6]黄新波,宋桐,王娅娜,等.基于IGSO优化LM网络的变压器故障诊断方法[J].中国电力,2014,47(9):60-65.
[7]王震宇,黄海波,张兵,等.基于故障树理论的高压开关柜状态检修[J].上海电力学院学报,2014,30(Suppl 1):21-24,31.
[8]宋人杰,杨爽,孟凡奇.基于多核学习整合GIS局部放电多类特征的分类研究[J].北京交通大学学报,2018,42(5):93-98,140.
[9]朱沛恒.基于果蝇算法优化的概率神经网络在变压器故障诊断中的应用[J].电力大数据,2018,21(6):37-43.
[10]吴小文,李擎.果蝇算法和5种群智能算法的寻优性能研究[J].火力与指挥控制,2013,38(4):17-20,25.
[11]毛长庚,梁伟峰.开关柜带电检测及局放定位技术的应用[J].大众用电,2017,32(9):26-27.
[12]林伟,王昕,郑益慧,等.基于概率神经网络的局部放电信号特征提取与识别研究[J].电气自动化,2018,40(3):60-63.
[13]任堂正,孙俊,沈晓峰,等.基于声电联合定位法的开关柜局部放电定位分析[J].电力与能源,2018,39(2):151-154.
[14]赵广振,张翠肖,武辉林,等.基于主成分分析和概率神经网络的入侵检测方法[J].石家庄铁道大学学报(自然科学版),2018,31(1):91-95.
[15]胡浩.一种基于概率神经网络建立电气系统故障的分类方法[J].科技创新与应用,2018(8):11-13.
[16]LI F H,LI R,TIAN L L,et al.Data-driven timefrequency analysis method based on variational mode decomposition and its application to gear fault diagnosis in variable working conditions[J].Mech Syst Signal Process,2019,116:462-479.
[17]李宗岳,陈志军,李名远.基于改进FOA-LSSVM的变压器故障诊断[J].水电能源科学,2016,34(4):194-197.
[18]CHEN A Y,THOMS R,KURFESS.Signal processing techniques for rolling element bearing spall size estimation[J].Mech Syst Signal Process,2019,117:16-32.
[19]贾勇勇,邓敏,李玉杰,等.基于深度残差网络的GIS局部放电模式识别技术研究[J].高压电器,2018,54(11):123-129.
[20]万书亭,豆龙江,李聪,等.基于VMD和样本熵的高压断路器故障特征提取及分类[J].振动与冲击,2018,37(20):32-38.
[21]张雪英,刘秀丽,栾忠权.基于峭度准则VMD样本熵与PNN的齿轮故障诊断[J].组合机床与自动化加工技术,2018(9):61-64.
[22]杨大为,赵永东,冯辅周,等.基于参数优化变分模态分解和多尺度熵偏均值的行星变速箱故障特征提取[J].兵工学报,2018,39(9):1683-1691.