基于Tsallis熵与层次化混合分类器的含未知故障断路器机械故障诊断
|
摘要
为提高断路器机械振动信号特征提取效率,避免将无训练样本未知类型故障误识别为正常样本或错误已知故障,提出一种断路器机械状态监测与故障诊断新方法。首先,将断路器机械操动机构振动信号进行时域分割,对分割后的各段信号分别直接提取7种特征,构成特征向量;通过散布矩阵分析特征分类能力,确定以Tsallis熵特征分析断路器机械故障。然后,将特征向量输入到基于单类支持向量机(OCSVM)与极限学习机(ELM)的层次化混合分类器中开展故障诊断。在混合分类器中,首先由OCSVM区分正常与故障状态;如为故障状态,则使用ELM识别故障类型,之后再以OCSVM校正ELM识别结果。通过实际断路器振动数据开展实验证明,散布矩阵能够有效分析特征的类可分性,时域分割提取特征效率高,层次化混合分类器不仅能够准确识别断路器机械状态与故障类型,而且可有效识别无训练样本未知故障类型数据。
To improve the efficiency of feature extraction from mechanical vibration signal of circuit breaker, and to avoid the problems that no-training samples of unknown type are recognized as the normal samples or wrong types of known fault, we propose a new method for monitoring and diagnosing the mechanical state fault of circuit breaker. Firstly, the vibration signal of circuit breaker's mechanism operating system is temporal segmented, and 7 features are extracted from each part of the divided signal to construct different kinds of feature vector. The Tsallis entropy is used to analyze the mechanical fault of circuit breaker by scatter matrix. Then, the feature vector is input to the hierarchical hybrid classifier based on one-class support vector machine(OCSVM) and extreme learning machine(ELM) for fault diagnosis. In the hybrid classifier, the state of normal and fault is recognized by OCSVM at first. If the sample is identified as fault state,the fault types of training samples are identified using ELM classifier. And then, the recognition results are corrected by OCSVM. The experiments carried out on the real circuit breaker prove that the scatter matrix can effectively analyze the class separability of features and the temporal segmented method is efficient to extract features. Hierarchical hybrid classifier can not only identify the mechanical state and fault type of circuit breaker accurately, but also identify the unknown fault type without training samples effectively.
To improve the efficiency of feature extraction from mechanical vibration signal of circuit breaker, and to avoid the problems that no-training samples of unknown type are recognized as the normal samples or wrong types of known fault, we propose a new method for monitoring and diagnosing the mechanical state fault of circuit breaker. Firstly, the vibration signal of circuit breaker's mechanism operating system is temporal segmented, and 7 features are extracted from each part of the divided signal to construct different kinds of feature vector. The Tsallis entropy is used to analyze the mechanical fault of circuit breaker by scatter matrix. Then, the feature vector is input to the hierarchical hybrid classifier based on one-class support vector machine(OCSVM) and extreme learning machine(ELM) for fault diagnosis. In the hybrid classifier, the state of normal and fault is recognized by OCSVM at first. If the sample is identified as fault state,the fault types of training samples are identified using ELM classifier. And then, the recognition results are corrected by OCSVM. The experiments carried out on the real circuit breaker prove that the scatter matrix can effectively analyze the class separability of features and the temporal segmented method is efficient to extract features. Hierarchical hybrid classifier can not only identify the mechanical state and fault type of circuit breaker accurately, but also identify the unknown fault type without training samples effectively.
引文
[1]黄建,胡晓光,巩玉楠.基于经验模态分解的高压断路器机械故障诊断方法[J].中国电机工程学报,2011,31(12):108-113.HUANG Jian,HU Xiaoguang,GONG Yunan.Machinery fault diagnosis of high voltage circuit breaker based on empirical mode decomposition[J].Proceedings of the CSEE,2011,31(12):108-113.
[2]赵莉华,付荣荣,荣强,等.基于自适应神经模糊推理系统的高压断路器操作机构状态评估[J].高电压技术,2017,43(6):2007-2015.ZHAO Lihua,FU Rongrong,RONG Qiang,et al.State evaluation of high voltage circuit breaker operating mechanism based on adaptive neural fuzzy inference system[J].High Voltage Engineering,2017,43(6):2007-2015.
[3]LEE D S,LITHGOW B,MORRISON R E.New fault diagnosis of circuit breakers[J].IEEE Transactions on Power Delivery,2003,18(2):454-459.
[4]孙鹏,曹雨晨,刘洋,等.采用二进制蚁群模糊神经网络的配电网故障分类方法[J].高电压技术,2016,42(7):2063-2072.SUN Peng,CAO Yuchen,LIU Yang,et al.Fault classification technique for power distribution network using binary ant colony algorithm and fuzzy neural network[J].High Voltage Engineering,2016,42(7):2063-2072.
[5]姚亚夫,张星.基于瞬时能量熵和SVM的滚动轴承故障诊断[J].电子测量与仪器学报,2013,27(10):957-962.YAO Yafu,ZHANG Xing.Fault diagnosis approach for roller bearing based on EMD momentary energy entropy and SVM[J].Journal of Electronic Measurement and Instrument,2013,27(10):957-962.
[6]孙来军,胡晓光,纪延超.改进的小波包-特征熵在高压断路器故障诊断中的应用[J].中国电机工程学报,2007,27(12):103-108.SUN Laijun,HU Xiaoguang,JI Yanchao.Fault diagnosis for high voltage circuit breakers with improved characteristic entropy of wavelet packet[J].Proceedings of the CSEE,2007,27(12):103-108.
[7]LANDRY M,LéONARD F,LANDRY C,et al.An improved vibration analysis algorithm as a diagnostic tool for detecting mechanical anomalies on power circuit breakers[J].IEEE Transactions on Power Delivery,2008,23(4):1986-1994.
[8]马强,荣命哲,贾申利.基于振动信号小波包提取和短时能量分析的高压断路器合闸同期性的研究[J].中国电机工程学报,2005,25(13):149-154.MA Qiang,RONG Mingzhe,JIA Shenli.Study of switching synchronization of high voltage breakers based on the wavelet packets extraction algorithm and short time analysis method[J].Proceedings of the CSEE,2005,25(13):149-154.
[9]吴广宁,袁海满,宋臻杰,等.基于粗糙集与多类支持向量机的电力变压器故障诊断[J].高电压技术,2017,43(11):202-208.WU Guangning,YUAN Haiman,SONG Zhenjie,et al.Fault diagnosis for power transformer based on rough set and multi-class support vector machine[J].High Voltage Engineering,2017,43(11):202-208.
[10]龙嘉川,王先培,代荡荡,等.基于改进S变换的超高频局部放电信号特征提取及分类[J].高电压技术,2018,44(11):3649-3656.LONG Jiachuan,WANG Xianpei,DAI Dangdang,et al.Feature extraction and classification of UHF PD signals based on improved S-transform[J].High Voltage Engineering,2018,44(11):3649-3656.
[11]游子跃,王宁,李明明,等.基于EEMD和BP神经网络的风机齿轮箱故障诊断方法[J].东北电力大学学报,2015,35(1):64-72.YOU Ziyue,WANG Ning,LI Mingming,et al.Method of fault diagnosis of gearbox based on EEMD and BP neural network[J].Journal of Northeast Electric Power University,2015,35(1):64-72.
[12]LIU H H,HAN M H.A fault diagnosis method based on local mean decomposition and multi-scale entropy for roller bearings[J].Mechanism&Machine Theory,2014,75(5):67-78.
[13]缪希仁,林瑞聪.基于关联维数与极端学习机的高压输电线路雷击过电压故障识别[J].高电压技术,2016,42(5):1519-1526.MIAO Xiren,LIN Ruicong.Lightning over-voltage fault identification of high-voltage transmission line based on correlation dimension and extreme learning machine[J].High Voltage Engineering,2016,42(5):1519-1526.
[14]何星,王宏利,陆敬辉,等.基于优选小波包和ELM的模拟电路故障诊断[J].仪器仪表学报,2013,34(11):2614-2619.HE Xing,WANG Hongli,LU Jinghui,et al.Analog circuit fault diagnosis method based on preferred wavelet packet and ELM[J].Chinese Journal of Scientific Instrument,2013,34(11):2614-2619.
[15]黄南天,张书鑫,蔡国伟,等.采用EWT和OCSVM的高压断路器机械故障诊断[J].仪器仪表学报,2015,36(12):2774-2781.HUANG Nantian,ZHANG Shuxin,CAI Guowei,et al.Mechanical fault diagnosis of high voltage circuit breakers utilizing EWT and OCSVM[J].Chinese Journal of Scientific Instrument,2015,36(12):2774-2781.
[16]YOON J,HE D,BRANDON VAN H B.On the use of a single piezoelectric strain sensor for wind turbin planetary gearbox fault diagnosis[J].IEEE Transactions on Industrial Electronics,2015,62(10):6585-6592.
[17]THEODORIDIS S,KOUTROUMBAS K.Pttern recognition[M].7th edition.Singapore:Academic Press,2008.
[18]黄南天,方立华,王玉强,等.基于局域均值分解和支持向量数据描述的高压断路器机械状态监测[J].电工电能新技术,2017,36(1):73-80.HUANG Nantian,FANG Lihua,WANG Yuqiang,et al.Machinery condition monitoring of high voltage circuit breakers based on local mean decomposition and support vector data description[J].Advanced Technology of Electrical Engineering and Energy,2017,36(1):73-80.
[2]赵莉华,付荣荣,荣强,等.基于自适应神经模糊推理系统的高压断路器操作机构状态评估[J].高电压技术,2017,43(6):2007-2015.ZHAO Lihua,FU Rongrong,RONG Qiang,et al.State evaluation of high voltage circuit breaker operating mechanism based on adaptive neural fuzzy inference system[J].High Voltage Engineering,2017,43(6):2007-2015.
[3]LEE D S,LITHGOW B,MORRISON R E.New fault diagnosis of circuit breakers[J].IEEE Transactions on Power Delivery,2003,18(2):454-459.
[4]孙鹏,曹雨晨,刘洋,等.采用二进制蚁群模糊神经网络的配电网故障分类方法[J].高电压技术,2016,42(7):2063-2072.SUN Peng,CAO Yuchen,LIU Yang,et al.Fault classification technique for power distribution network using binary ant colony algorithm and fuzzy neural network[J].High Voltage Engineering,2016,42(7):2063-2072.
[5]姚亚夫,张星.基于瞬时能量熵和SVM的滚动轴承故障诊断[J].电子测量与仪器学报,2013,27(10):957-962.YAO Yafu,ZHANG Xing.Fault diagnosis approach for roller bearing based on EMD momentary energy entropy and SVM[J].Journal of Electronic Measurement and Instrument,2013,27(10):957-962.
[6]孙来军,胡晓光,纪延超.改进的小波包-特征熵在高压断路器故障诊断中的应用[J].中国电机工程学报,2007,27(12):103-108.SUN Laijun,HU Xiaoguang,JI Yanchao.Fault diagnosis for high voltage circuit breakers with improved characteristic entropy of wavelet packet[J].Proceedings of the CSEE,2007,27(12):103-108.
[7]LANDRY M,LéONARD F,LANDRY C,et al.An improved vibration analysis algorithm as a diagnostic tool for detecting mechanical anomalies on power circuit breakers[J].IEEE Transactions on Power Delivery,2008,23(4):1986-1994.
[8]马强,荣命哲,贾申利.基于振动信号小波包提取和短时能量分析的高压断路器合闸同期性的研究[J].中国电机工程学报,2005,25(13):149-154.MA Qiang,RONG Mingzhe,JIA Shenli.Study of switching synchronization of high voltage breakers based on the wavelet packets extraction algorithm and short time analysis method[J].Proceedings of the CSEE,2005,25(13):149-154.
[9]吴广宁,袁海满,宋臻杰,等.基于粗糙集与多类支持向量机的电力变压器故障诊断[J].高电压技术,2017,43(11):202-208.WU Guangning,YUAN Haiman,SONG Zhenjie,et al.Fault diagnosis for power transformer based on rough set and multi-class support vector machine[J].High Voltage Engineering,2017,43(11):202-208.
[10]龙嘉川,王先培,代荡荡,等.基于改进S变换的超高频局部放电信号特征提取及分类[J].高电压技术,2018,44(11):3649-3656.LONG Jiachuan,WANG Xianpei,DAI Dangdang,et al.Feature extraction and classification of UHF PD signals based on improved S-transform[J].High Voltage Engineering,2018,44(11):3649-3656.
[11]游子跃,王宁,李明明,等.基于EEMD和BP神经网络的风机齿轮箱故障诊断方法[J].东北电力大学学报,2015,35(1):64-72.YOU Ziyue,WANG Ning,LI Mingming,et al.Method of fault diagnosis of gearbox based on EEMD and BP neural network[J].Journal of Northeast Electric Power University,2015,35(1):64-72.
[12]LIU H H,HAN M H.A fault diagnosis method based on local mean decomposition and multi-scale entropy for roller bearings[J].Mechanism&Machine Theory,2014,75(5):67-78.
[13]缪希仁,林瑞聪.基于关联维数与极端学习机的高压输电线路雷击过电压故障识别[J].高电压技术,2016,42(5):1519-1526.MIAO Xiren,LIN Ruicong.Lightning over-voltage fault identification of high-voltage transmission line based on correlation dimension and extreme learning machine[J].High Voltage Engineering,2016,42(5):1519-1526.
[14]何星,王宏利,陆敬辉,等.基于优选小波包和ELM的模拟电路故障诊断[J].仪器仪表学报,2013,34(11):2614-2619.HE Xing,WANG Hongli,LU Jinghui,et al.Analog circuit fault diagnosis method based on preferred wavelet packet and ELM[J].Chinese Journal of Scientific Instrument,2013,34(11):2614-2619.
[15]黄南天,张书鑫,蔡国伟,等.采用EWT和OCSVM的高压断路器机械故障诊断[J].仪器仪表学报,2015,36(12):2774-2781.HUANG Nantian,ZHANG Shuxin,CAI Guowei,et al.Mechanical fault diagnosis of high voltage circuit breakers utilizing EWT and OCSVM[J].Chinese Journal of Scientific Instrument,2015,36(12):2774-2781.
[16]YOON J,HE D,BRANDON VAN H B.On the use of a single piezoelectric strain sensor for wind turbin planetary gearbox fault diagnosis[J].IEEE Transactions on Industrial Electronics,2015,62(10):6585-6592.
[17]THEODORIDIS S,KOUTROUMBAS K.Pttern recognition[M].7th edition.Singapore:Academic Press,2008.
[18]黄南天,方立华,王玉强,等.基于局域均值分解和支持向量数据描述的高压断路器机械状态监测[J].电工电能新技术,2017,36(1):73-80.HUANG Nantian,FANG Lihua,WANG Yuqiang,et al.Machinery condition monitoring of high voltage circuit breakers based on local mean decomposition and support vector data description[J].Advanced Technology of Electrical Engineering and Energy,2017,36(1):73-80.