基于SOM的变压器绕组和铁芯故障诊断
|
摘要
为了实现变压器故障的直观分类和故障识别,在分析变压器振动机理的基础上,提出一种基于自组织特征映射(SOM)神经网络的变压器故障检测方法。首先利用集合经验模式分解(EEMD)方法提取变压器不同运行状态下振动信号的特征矢量,以其表示变压器运行状态,降低故障分类和识别时运算量。然后通过采用SOM网络自组织学习算法,不断学习样本的特征矢量确定故障隶属函数,从而可以快速有效地诊断变压器的故障类型。试验结果表明,该方法可实现对变压器正常状态、绕组轴向变形、绕组径向变形、铁芯故障4种状态分类,并对测试样本进行快速的模式识别。
In order to achieve the fault classification and fault identification of transformer,a transformer fault detection method based on self-organizing feature map(SOM)neural network is proposed under the consideration of the transformer vibration mechanism.Firstly,the ensemble empirical mode decomposition(EEMD)is utilized to obtain the characteristic vector of vibration signals in different operation states which represents the operative state for the purpose of reducing the amount of calculation in classification and recognition.Next,the SOM neural network integrated the self-organizing learning algorithm ensures the fault membership functions via continuously learning of the sample characteristic vector.In this way,the type of transformer faults is diagnosed quickly and efficiently.The simulation results show that the fault classification is implemented for a normal state of transformer,winding radial deformation,the winding axial deformation and the core fault respectively to realize automatically pattern recognition.
In order to achieve the fault classification and fault identification of transformer,a transformer fault detection method based on self-organizing feature map(SOM)neural network is proposed under the consideration of the transformer vibration mechanism.Firstly,the ensemble empirical mode decomposition(EEMD)is utilized to obtain the characteristic vector of vibration signals in different operation states which represents the operative state for the purpose of reducing the amount of calculation in classification and recognition.Next,the SOM neural network integrated the self-organizing learning algorithm ensures the fault membership functions via continuously learning of the sample characteristic vector.In this way,the type of transformer faults is diagnosed quickly and efficiently.The simulation results show that the fault classification is implemented for a normal state of transformer,winding radial deformation,the winding axial deformation and the core fault respectively to realize automatically pattern recognition.
引文
[1]刘秀军,曲建绪,袁超,等.基于变压器铁芯接地线的电网故障暂态行波信号检测新方法[J].电力科学与技术学报,2016,31(3):57-64.LIU Xiu-jun,QU Jian-xu,YUAN Chao,et al.A novel method for power grid fault traveling wave detected based on the grounding line of transformer core[J].Journal of Electric Power Science and Technology,2016,31(3):57-64.
[2]刘勇,杨帆,张凡,等.检测电力变压器绕组变形的扫频阻抗法研究[J].中国电机工程学报.2015,35(17):4 505-4 517.LIU Yong,YANG Fan,ZHANG Fan,et al.Study on sweep frequency impedance to detect winding deformation within power transformer[J].Proceedings of CSEE,2015,35(17):4 505-4 517.
[3]仇剑东,林鹤云.基于能量法的配电变压器短路阻抗计算[J].电气应用.2006,25(11):26-29.QIU Jian-dong,LIN He-yun.Calculation of transformer’s short impedance based on energy method[J].Electrical Application,2006,25(11):26-29.
[4]王建全,陈捷.基于短路电抗在线监测法的变压器绕组变形分析[D].杭州:浙江大学,2011.
[5]孙峰,陈志国.频响法绕组变形试验相关标准的比较[J].变压器,2016,52(2):36-39.SUN Feng,CHEN Zhi-guo.Comparison of winding deformation test with FRA in relevant standards[J].Transformer,2016,52(2):36-39.
[6]曹小龙,胡春梅,曹小虎,等.变压器频响应法测试结果的影响因素分析及改善[J].高压电器,2012,48(7):81-87.CAO Xiao-long,HU Chun-mei,CAO Xiao-hu,et al.Influencing factors on deformation diagnosis of transformer windings[J].High Voltage Apparatus,2012,48(7):81-87.
[7]Garcia B,Burgos J C,Alonso A M.Transformer tank vibration modeling as a method of detecting deformations Part-I:Theoretical foundation[J].IEEE Transactions on Power Delivery,2006,21(1),158-163.
[8]王丰华,李清,金之俭,等.振动法在线检测突发短路时变压器绕组状态[J].控制工程,2011,18(4):596-599.WANG Feng-hua,LI Qing,JIN Zhi-jian,et al.On-line monitoring the winding condition of power transformer under sudden short-circuit based on the vibration analysis method[J].Control Engineering of China,2011,18(4):596-599.
[9]Garcia B,Burgos J C,Alonso A M.Transformer tank vibration modeling as a method of detection winding deformations partⅡ:Experimental verification[J].IEEE Transaction on Power delivery,2006,21(1):164-169.
[10]谢坡岸.振动分析法在电力变压器绕组状态检测中的应用研究[D].上海:上海交通大学,2008.
[11]熊卫华,赵光宙.基于希尔伯特—黄变换的变压器铁心振动特性分析[J].电工技术学报,2006,21(8):9-13.XIONG Wei-hua,ZHAO Guang-zhou.Analysis of transformer core vibration characteristics using Hilbert-Huang transformation[J].Transactions of China Electrotechnical Society,2006,21(8):9-13.
[12]徐艳春,陈国训,李振兴,等.基于HHT的电能质量多扰动信号检测方法[J].电力科学与技术学报,2016,31(1):55-61.XU Yan-chun,CHEN Guo-xun,LI Zhen-xing,et al.Power quality disturbance signal detection method based on HHT[J].Journal of Electric Power Science and Technology,2016,31(1):55-61.
[13]张宇辉,武东斌,吴家明,等.基于品质因子可调小波变换与排列熵的电能质量信号检测方法[J].电力科学与技术学报,2018,33(1):75-80.ZHANG Yu-hui,WU Dong-bin,WU Jia-ming,et al.Power quality signal detection method with tunable Qfactor wavelet transform and permutation entropy[J].Journal of Electric Power Science and Technology,2018,33(1):75-80.
[14]李莉,朱永利.变压器绕组多故障条件下的振动信号特征提取[J].电力自动化设备,2014,34(8):140-146.LI Li,ZHU Yong-li.Feature extraction for vibration signal of transformer winding with multiple faults[J].Electric Power Automation Equipment,2014,34(8):140-146.
[15]高宇.基于SOM神经网络的风电电子装置故障诊断[J].电力系统及其自动化学报,2010,22(3):142-145.GAO Yu.Fault diagnosis of wind turbine power electronic devices based on SOM neural network[J].Proceedings of the CSU-EPSA,2010,22(3):142-145.
[16]尹柏强,何怡刚,朱彦卿.一种广义S变换及模糊SOM网络的电能质量多扰动检测和识别方法[J].中国电机工程学报,2015,35(4):866-872.YIN Bai-qiang,HE Yi-gang,ZHU Yan-qing.Detection and classification of power quality multi-disturbances based on generalized S-transform and fuzzy SOM neural network[J].Proceedings of CSEE,2015,35(4):866-872.
[2]刘勇,杨帆,张凡,等.检测电力变压器绕组变形的扫频阻抗法研究[J].中国电机工程学报.2015,35(17):4 505-4 517.LIU Yong,YANG Fan,ZHANG Fan,et al.Study on sweep frequency impedance to detect winding deformation within power transformer[J].Proceedings of CSEE,2015,35(17):4 505-4 517.
[3]仇剑东,林鹤云.基于能量法的配电变压器短路阻抗计算[J].电气应用.2006,25(11):26-29.QIU Jian-dong,LIN He-yun.Calculation of transformer’s short impedance based on energy method[J].Electrical Application,2006,25(11):26-29.
[4]王建全,陈捷.基于短路电抗在线监测法的变压器绕组变形分析[D].杭州:浙江大学,2011.
[5]孙峰,陈志国.频响法绕组变形试验相关标准的比较[J].变压器,2016,52(2):36-39.SUN Feng,CHEN Zhi-guo.Comparison of winding deformation test with FRA in relevant standards[J].Transformer,2016,52(2):36-39.
[6]曹小龙,胡春梅,曹小虎,等.变压器频响应法测试结果的影响因素分析及改善[J].高压电器,2012,48(7):81-87.CAO Xiao-long,HU Chun-mei,CAO Xiao-hu,et al.Influencing factors on deformation diagnosis of transformer windings[J].High Voltage Apparatus,2012,48(7):81-87.
[7]Garcia B,Burgos J C,Alonso A M.Transformer tank vibration modeling as a method of detecting deformations Part-I:Theoretical foundation[J].IEEE Transactions on Power Delivery,2006,21(1),158-163.
[8]王丰华,李清,金之俭,等.振动法在线检测突发短路时变压器绕组状态[J].控制工程,2011,18(4):596-599.WANG Feng-hua,LI Qing,JIN Zhi-jian,et al.On-line monitoring the winding condition of power transformer under sudden short-circuit based on the vibration analysis method[J].Control Engineering of China,2011,18(4):596-599.
[9]Garcia B,Burgos J C,Alonso A M.Transformer tank vibration modeling as a method of detection winding deformations partⅡ:Experimental verification[J].IEEE Transaction on Power delivery,2006,21(1):164-169.
[10]谢坡岸.振动分析法在电力变压器绕组状态检测中的应用研究[D].上海:上海交通大学,2008.
[11]熊卫华,赵光宙.基于希尔伯特—黄变换的变压器铁心振动特性分析[J].电工技术学报,2006,21(8):9-13.XIONG Wei-hua,ZHAO Guang-zhou.Analysis of transformer core vibration characteristics using Hilbert-Huang transformation[J].Transactions of China Electrotechnical Society,2006,21(8):9-13.
[12]徐艳春,陈国训,李振兴,等.基于HHT的电能质量多扰动信号检测方法[J].电力科学与技术学报,2016,31(1):55-61.XU Yan-chun,CHEN Guo-xun,LI Zhen-xing,et al.Power quality disturbance signal detection method based on HHT[J].Journal of Electric Power Science and Technology,2016,31(1):55-61.
[13]张宇辉,武东斌,吴家明,等.基于品质因子可调小波变换与排列熵的电能质量信号检测方法[J].电力科学与技术学报,2018,33(1):75-80.ZHANG Yu-hui,WU Dong-bin,WU Jia-ming,et al.Power quality signal detection method with tunable Qfactor wavelet transform and permutation entropy[J].Journal of Electric Power Science and Technology,2018,33(1):75-80.
[14]李莉,朱永利.变压器绕组多故障条件下的振动信号特征提取[J].电力自动化设备,2014,34(8):140-146.LI Li,ZHU Yong-li.Feature extraction for vibration signal of transformer winding with multiple faults[J].Electric Power Automation Equipment,2014,34(8):140-146.
[15]高宇.基于SOM神经网络的风电电子装置故障诊断[J].电力系统及其自动化学报,2010,22(3):142-145.GAO Yu.Fault diagnosis of wind turbine power electronic devices based on SOM neural network[J].Proceedings of the CSU-EPSA,2010,22(3):142-145.
[16]尹柏强,何怡刚,朱彦卿.一种广义S变换及模糊SOM网络的电能质量多扰动检测和识别方法[J].中国电机工程学报,2015,35(4):866-872.YIN Bai-qiang,HE Yi-gang,ZHU Yan-qing.Detection and classification of power quality multi-disturbances based on generalized S-transform and fuzzy SOM neural network[J].Proceedings of CSEE,2015,35(4):866-872.