基于小波奇异熵与SOFM神经网络的电机轴承故障识别

PDF(1317 KB)

振动与冲击 ›› 2017, Vol. 36 ›› Issue (10) : 217-223.

论文

贺岩松1,2，黄毅2 ，徐中明1,2 ，张志飞2

作者信息 +

Motor BearingFault Identification Based on Wavelet Singular Entropy and SOFM Neural Network

HE Yan-song1,2 HUANG Yi2 XU Zhong-ming1,2ZHANG Zhi-fei2

Author information +

文章历史 +

摘要

提出一种用小波奇异熵(WSE)和自组织特征映射(SOFM)神经网络进行电机轴承故障识别的建模方法。首先通过对电机驱动端和风扇端采集的故障振动信号的小波奇异熵的计算和比较来识别故障轴承的端位；在此基础上以故障端信号的小波包分解底层各结点能量为特征向量输入建立自组织特征映射神经网络模型来识别故障轴承内部的具体点蚀破坏位置。小波奇异熵和SOFM神经网络的结合实现了故障轴承端位及其内部点蚀位置的联合识别。分别对含有内外圈和滚动体点蚀故障的轴承进行建模和识别试验，结果表明：该模型可以有效地识别电机故障轴承的端位及其内部点蚀破坏位置；与传统支持向量机和BP神经网络识别模型相比,该模型故障识别准确率更高，识别稳定性更好，更适宜于故障识别这样的多分类问题。

Abstract

A new modeling method combining wavelet singular entropy andSelf Organizing Feature Map(SOFM) neural network is proposed to identify the motor bearing faults.The end position of the faulty bearing can be identifiedby computing and comparing the wavelet singular entropies of the fault vibration signals collected at the drive and fan ends of the motor firstly.Then the SOFM neural networkmodel using the bottom node energies of the fault signals decomposed by wavelet packet as input feature vectorsis built to identify the pitting corrosiondamage locationin the faulty bearing.The faulty bearing end positionand its internal pitting corrosion location can be identified jointly by the combination of wavelet singular entropy and SOFM neural network.Through the modeling and identification to the bearings damaged by pitting corrosionatinner,outer raceway and rolling element respectively,the results show that this model can identify the faulty bearing end position and its internal pitting corrosiondamage locationeffectively;this model has a more accurate identification ability and is more robust than that of the traditional support vector machine and the BP neural networkidentification model,and is more suitable for fault identification of such a multi classifica-tion problem.

导出引用

贺岩松1,2，黄毅2,徐中明1,2,张志飞2. 基于小波奇异熵与SOFM神经网络的电机轴承故障识别[J]. 振动与冲击, 2017, 36(10): 217-223

HE Yan-song1,2 HUANG Yi2 XU Zhong-ming1,2ZHANG Zhi-fei2. Motor BearingFault Identification Based on Wavelet Singular Entropy and SOFM Neural Network[J]. Journal of Vibration and Shock, 2017, 36(10): 217-223

参考文献

[1] 王小玲，陈进，从飞云.基于时频的频带熵方法在滚动轴承故障识别中的应用[J].振动与冲击，2012,31(18)：29-33.
WANG Xiao-lin,CHEN Jin,CONG Fei-yun.Application of spectral band entropy (SBE) method in rolling bearing fault diagosis based on time-frequency analysis[J].Journal of vibration and shock,2012,31(18):29-33.
[2] 张云强，张培林，吴定海等.基于最优广义S变换和脉冲耦合神经网络的轴承故障诊断[J].振动与冲击，2015,34(9)：26-31.
ZHANG Yun-qiang,ZHANG Pei-lin,WU Ding-hai,et al. Bearing fault diagnosis based on optimal generalized S transform and pulse coupled neural network[J].Journal of vibration and shock,2015,34(9):26-31.
[3] 向玲，唐贵基，胡爱军.旋转机械非平稳振动信号的时频分析比较[J].振动与冲击，2010,29(2)：42-45.
XIANG Lin,TANG Gui-ji,HU Ai-jun.Vibration signal’s time-frequency analysis and comparison for a rotating machinery[J].Journal of vibration and shock,2010,29(2):
42-45.
[4] Zheng H,Li Z,Chen X.Gear fault diagnosis based on con-tinuous wavelet transform[J].Mechanical Systems and Si-gnal Processing,2002,16(2):447-457.
[5] Ales B,Joze F.Wavelet analysis for gear crack identifica-tion[J].Mechanical Systemm and Signal Processing,200-9,23:1554-1572.
[6] 张玲玲，赵懿冠，肖云魁等.基于小波包-AR谱的变速器轴承故障特征提取[J].振动、测试与诊断，2011,31(4)：492-495.
ZHANG Ling-ling,ZHAO Yi-guan,XIAO Yun-kui,et al.
Automobile transmission bearing fault feature extraction based on wavelet packet-auto regressive model spectrum-[J].Journal of Vibration,Measurement & Diagnosis,2011,
31(4):492-495.
[7] Jena D P,Panigrahi S N,Kumar R.Gear fault identification and localization using analytic wavelet transform of vibra-tion signal[J].Measurement,2013,46:1115-1124.
[8] 赵元喜，高立新，崔玲丽等.基于谐波小波包和BP神经网络的滚动轴承声发射故障模式识别技术[J].振动与冲击，2010,29(10)：162-165.
ZHAO Yuan-xi,GAO Li-xin,CUI Ling-li,et al.Fault patte-rn recognition technique for roller bearing acoustic emissi-on based on harmonic wavelet packet and BP neural net-work[J].Journal of vibration and shock,2010,29(10):162-165.
[9] 窦东阳，赵英凯.基于分类器组的轴承故障识别方法研究[J].振动与冲击，2010,29(10):221-224.
DOU Dong-yang,ZHAO Yin-kai.Fault recognition meth-od of rolling bearings based on classifier ensembles[J].Jo-urnal of vibration and shock,2010,29(10):221-224.
[10] 胥永刚，孟志鹏，陆明.基于双树复小波包变换和SVM的滚动轴承故障诊断方法[J].航空动力学报，2014,29(1)：67-72.
XU Yong-gang,MENG Zhi-peng,LU Ming.Fault diagno-sis method of rolling bearing based on dual-tree complex wavelet packet transform and SVM[J].Journal of Aero-space Power,2014,29(1):67-72.
[11] 杨雪.支持向量机多分类分类方法的研究[D].哈尔滨：哈尔滨工程大学，2006.
YANG Xue.Research of multi-class classification support vector machine[D].Harbin:Harbin Engineering Universi-ty,2006.
[12] 秦政博.基于小波包和支持向量机的滚动轴承故障识别方法[D].太原：太原理工大学，2010.
QIN Zhen-bo.Rolling bearing fault identification method based on wavelet package and support vector machine[D].
Taiyuan:Taiyuan University of Technology,2010.
[13] 雷璐宁，石为人，范敏.基于改进的SOM神经网络在水质评价分析中的应用[J].仪器仪表学报，2009，30(11)：2379-2382.
LEI Lu-ning,SHI Wei-ren,FAN Min.Water quality evalua-tion analysis based on improved SOM neural network[J].
Chinese Jorunal of Scientific Instrument,2009,30(11):23-97-2382.
[14] 张莹，潘保芝.基于主成分分析的SOM神经网络在火山岩性别识别中的应用[J].测井技术，2009，33(6)：550-554.
ZHANG Ying,PAN Bao-zhi.Application of SOM neural network method to volcanic lithology recognition based on principal components analysis[J].Well Logging Tech-nology,2009,33(6):550-554.
[15] Rosson O A,Blanco S,Yordanova J,et al.Wavelet entropy:a new tool for analysis of short duration brain electrical signals[J].J Neurosci Meth,2001,105(1):65-75.
[16] He Z Y,Wu X,Qian Q Q.Automatic fault detection for po-wer system using wavelet singular entropy[C]．Proceeding of ICIMA IEEE compress,Cheng du,2004.
[17] 谢平，杜义浩，黄双峰.基于信息熵的复杂机械系统非线性特征提取方法研究[J].振动与冲击，2008,27(7)：135-137.
XIE Ping,DU Yi-hao,HUANG Shuang-feng.Research on non-linear feature extracting method of complex mechani-cal system based on information entropy[J].Journal of vibration and shock,2008,27(7):135-137.
[18] The Case Western Reserve University Bearing Data Cen-ter Website[EB/OL].http://csegroup.edu/bearingdatacen-ter/home.
[19] 王冰，刁鸣，宋凯.基于小波奇异熵和相关向量机的氢气传感器故障诊断[J].电机与控制学报，2015,19(1):96-101.
WANG Bing,DIAO Ming,SONG Kai.Fault diagnosis of hydrogen sensor based on wavelet singular entropy and relevance vector machine[J].Electric Mchine and Contr-ol,2015,19(1):96-101.
[20] 何正友，符玲，钱清泉等.小波奇异熵及其在高压输电线路故障选相中的应用[J].中国电机工程学报，2007,27(1):31-36.
HE Zheng-you,FU Ling,QIAN Qing-quan et al.Study on wavelet singular entropy and its application to faulty phase selection in HV transmission lines[J].Proceedings of the CSEE,2007,27(1):31-36.
[21] 曾宪伟，赵卫民，盛菊琴.小波包分解树结点与信号子空间频带的对应关系及其应用[J].地震学报，2008,30(1)：90-96.
ZENG Xian-wei,ZHAO Wei-ming,SHENG Ju-qin.Corr-esponding relationships between nodes of decomposition tree of wavelet packet and frequency bands of signal subs-pace[J].Acta Seimologica Sinica,2008,30(1):90-96.