基于LMD和MSEE的滚动轴承复合故障特征提取方法
|
摘要
针对滚动轴承复合故障特征相近、不易区分的问题,提出了一种基于局域均值分解和多尺度熵能量的滚动轴承复合故障特征提取方法。首先,将信号进行LMD处理,得到一系列PF分量;然后,通过相关系数选择合适的PF分量计算能量并获得新的时间序列;最后,计算新时间序列的多尺度熵,与能量结合构建MSEE进行故障特征提取。机械故障模拟试验台的结果表明:该方法不仅降低了噪声干扰,而且提升了特征提取的精度,可以定量表征滚动轴承复合故障信号的特征,在滚动轴承复合故障信号中有良好的特征提取效果,与单独使用MSE和能量的特征提取方法相比,故障诊断率分别提升了8. 33%和11. 29%。
The compound fault features of rolling bearings are similar and not easy to distinguish,and a extraction method for compound fault features of rolling bearings is proposed based on local mean decomposition( LMD) and multi-scale entropy energy( MSEE). Firstly,the signal is decomposed by LMD,obtaining a series of PF components. Then,the appropriate PF components are selected by correlation coefficients to calculate energy,and the new time series are obtained. Finally,the multi-scale entropy of new time series is calculated and combined with energy to build MSEE for fault feature extraction. The results obtained by mechanical fault simulation test bench show that the method not only reduces noise interference,but also improves accuracy of feature extraction,which characterizes features of compound fault signal of rolling bearings quantitatively. The fault diagnosis rate of proposed method increases by 8. 3% and 11. 3%compared with feature extraction method using MSE and energy,which has favorable feature extraction effect in compound fault signal of rolling bearings.
The compound fault features of rolling bearings are similar and not easy to distinguish,and a extraction method for compound fault features of rolling bearings is proposed based on local mean decomposition( LMD) and multi-scale entropy energy( MSEE). Firstly,the signal is decomposed by LMD,obtaining a series of PF components. Then,the appropriate PF components are selected by correlation coefficients to calculate energy,and the new time series are obtained. Finally,the multi-scale entropy of new time series is calculated and combined with energy to build MSEE for fault feature extraction. The results obtained by mechanical fault simulation test bench show that the method not only reduces noise interference,but also improves accuracy of feature extraction,which characterizes features of compound fault signal of rolling bearings quantitatively. The fault diagnosis rate of proposed method increases by 8. 3% and 11. 3%compared with feature extraction method using MSE and energy,which has favorable feature extraction effect in compound fault signal of rolling bearings.
引文
[1]LI B,ZHANG P L,TIAN H,et al.A new feature extraction and selection scheme for hybrid fault diagnosis of gearbox[J].Expert Systems with Applications,2011,38(8):10000-10009.
[2]CHEN Y M,ZI Y Y,CAO H R,et al.A data-driven threshold for wavelet sliding window denoisingin mechanical fault detection[J].Science China-Technological Sciences,2014,57(3):589-597.
[3]石娟娟,王林,罗春艳,等.基于优化最小算法的齿轮箱复合故障特征稀疏表示[J].振动工程学报,2017,30(6):1045-1055.
[4]张晓涛,唐力伟,王平,等.最小周期相关熵解卷积结合窄带解调的轴承复合故障诊断研究[J].振动工程学报,2015,28(4):666-672.
[5]沈仁发,祁彦洁,康海英,等.基于阶次小波包与粗糙集的轴承复合故障诊断[J].轴承,2009(9):53-56.
[6]韩涛,袁建虎,唐建,等.基于MWT和CNN的滚动轴承智能复合故障诊断方法[J].机械传动,2016(12):139-143.
[7]马新娜,杨绍普.滚动轴承复合故障诊断的自适应方法研究[J].振动与冲击,2016,35(10):145-150.
[8]郑近德,程军圣,杨宇.基于多尺度熵的滚动轴承故障诊断方法[J].湖南大学学报(自然科学版),2012,39(5):38-41.
[9]石瑞敏,杨兆建.基于LMD能量特征的滚动轴承故障诊断方法[J].振动.测试与诊断,2015,35(5):832-836.
[10]SMITH J S.The local mean decomposition and its application to EEG perception data[J].Journal of the Royal Society Interface,2005,2(5):443-454.
[11]武哲,杨绍普,张建超.基于LMD自适应多尺度形态学和Teager能量算子方法在轴承故障诊断中的应用[J].振动与冲击,2016,35(3):7-13.
[12]MA J,WU J,YUAN X.The fault diagnosis of the rolling bearing based on the LMD and time-frequency analysis[J].International Journal of Control&Automation,2013,6(4)357-376.
[13]李沁雪,廖晓文,张清华,等.基于EWT和多尺度熵的轴承特征提取及分类[J].轴承,2016(1):48-52.
[14]郑近德,程军圣,胡思宇.多尺度熵在转子故障诊断中的应用[J].振动、测试与诊断,2013,33(2):294-297.
[15]胥永刚,孟志鹏,赵国亮.基于双树复小波变换的轴承复合故障诊断研究[J].仪器仪表学报,2014,35(2):447-452.
[2]CHEN Y M,ZI Y Y,CAO H R,et al.A data-driven threshold for wavelet sliding window denoisingin mechanical fault detection[J].Science China-Technological Sciences,2014,57(3):589-597.
[3]石娟娟,王林,罗春艳,等.基于优化最小算法的齿轮箱复合故障特征稀疏表示[J].振动工程学报,2017,30(6):1045-1055.
[4]张晓涛,唐力伟,王平,等.最小周期相关熵解卷积结合窄带解调的轴承复合故障诊断研究[J].振动工程学报,2015,28(4):666-672.
[5]沈仁发,祁彦洁,康海英,等.基于阶次小波包与粗糙集的轴承复合故障诊断[J].轴承,2009(9):53-56.
[6]韩涛,袁建虎,唐建,等.基于MWT和CNN的滚动轴承智能复合故障诊断方法[J].机械传动,2016(12):139-143.
[7]马新娜,杨绍普.滚动轴承复合故障诊断的自适应方法研究[J].振动与冲击,2016,35(10):145-150.
[8]郑近德,程军圣,杨宇.基于多尺度熵的滚动轴承故障诊断方法[J].湖南大学学报(自然科学版),2012,39(5):38-41.
[9]石瑞敏,杨兆建.基于LMD能量特征的滚动轴承故障诊断方法[J].振动.测试与诊断,2015,35(5):832-836.
[10]SMITH J S.The local mean decomposition and its application to EEG perception data[J].Journal of the Royal Society Interface,2005,2(5):443-454.
[11]武哲,杨绍普,张建超.基于LMD自适应多尺度形态学和Teager能量算子方法在轴承故障诊断中的应用[J].振动与冲击,2016,35(3):7-13.
[12]MA J,WU J,YUAN X.The fault diagnosis of the rolling bearing based on the LMD and time-frequency analysis[J].International Journal of Control&Automation,2013,6(4)357-376.
[13]李沁雪,廖晓文,张清华,等.基于EWT和多尺度熵的轴承特征提取及分类[J].轴承,2016(1):48-52.
[14]郑近德,程军圣,胡思宇.多尺度熵在转子故障诊断中的应用[J].振动、测试与诊断,2013,33(2):294-297.
[15]胥永刚,孟志鹏,赵国亮.基于双树复小波变换的轴承复合故障诊断研究[J].仪器仪表学报,2014,35(2):447-452.