基于VMD-HMM的滚动轴承磨损状态识别

PDF(2065 KB)

振动与冲击 ›› 2018, Vol. 37 ›› Issue (21) : 61-67.

论文

李奕江1，张金萍1，李允公2

作者信息 +

Wear state recognition of rolling bearings based on VMD-HMM

LI Yijiang1, ZHANG Jinping1, LI Yungong2

Author information +

文章历史 +

摘要

根据变分模态分解（VMD）在信号处理上的良好特性与隐马尔可夫模型（HMM）在时间序列上的分类能力，提出了一种基于VMD分解与HMM模型相结合的滚动轴承磨损状态识别方法。该方法首先利用VMD对轴承各个磨损时期信号进行分解，进而计算VMD分解后各IMF的能量熵，然后依次提取轴承振动信号的各层IMF能量熵构成特征向量序列，最后每种状态随机选取20组（共80组）输入HMM模型训练，剩余的特征向量序列进行测试，通过对比对数似然概率值来判别磨损状态。实验结果表明该方法能够准确分辨出轴承的磨损状态，然后与EMD-HMM、谐波小波样本熵HMM模型进行了对比，验证了该方法具有识别性高准确性强的优点。

Abstract

Based on good performance of the variational mode decomposition (VMD) in signal processing and classification ability of the hidden Markov model (HMM) to time series, a rolling bearing wear state recognition method based on VMD-HMM was proposed.Firstly, VMD was used to decompose vibration signals of a bearing in its various wear durations, and the energy entropy of each IMF after VMD was calculated.Then, various IMFs’ energy entropies of bearing vibration signals in various wear durations were extracted to form eigenvector sequences.Finally, the randomly selected 20 groups in total 80 groups of eigenvector sequences for each wear state were input into HMM model to be trained, and the rest eigenvector sequences were tested.Through comparing logarithmic likelihood probability values, the bearing wear state was determined.The test results showed that the proposed method can be used to accurately distinguish the wear state of the bearing; compared with EMD-HMM and the harmonic wavelet sample entropy HMM model, it has higher recognition and accuracy.

导出引用

李奕江1，张金萍1，李允公2. 基于VMD-HMM的滚动轴承磨损状态识别[J]. 振动与冲击, 2018, 37(21): 61-67

LI Yijiang1, ZHANG Jinping1, LI Yungong2. Wear state recognition of rolling bearings based on VMD-HMM[J]. Journal of Vibration and Shock, 2018, 37(21): 61-67

参考文献

[1] 钟秉林，黄仁.机械故障诊断学[M].北京：机械工业出版社，2007
[2] 林丽,李国禄,王海斗,康嘉杰.基于声发射技术监测疲劳磨损失效的研究进展[J].材料导报,2015,(03):109-114.
Lin Li, Li Guo-lu, Wang Hai-dou, Kang Jia-jie. Research Progress of Monitoring Fatigue Wear Failure Based on Acoustic Emission Technology[J] Materials Herald, 2015,(03): 109-114.
[3] 赵迎祥,魏宗平. 滚动轴承磨损寿命数据的灰色预测[J]. 机械制造,2010,(11):66-68.
Zhao Ying-xiang, Wei Zong-ping. Gray Prediction of Rolling Bearing Wear Life Data[J].Machinery, 2010,(11):66-68.
[4] 郭四洲,佘宝瑛,梅荣海,杨下沙. 圆柱滚子轴承不均匀磨损的检测方法研究与应用[J]. 科技创新与应用,2015,16:12-14.
Guo Si-zhou, She Bao-ying, Mei Rong-hai, Yang Xia-sha. Study and application of nonuniform wear detection for cylindrical roller bearing [J]. Science and Technology Innovation and Application, 2015, 16: 12-14.
[5] Zhou H, Chen J, Dong G, et al. Detection and diagnosis of bearing faults using shift-invariant dictionary learning and hidden Markov model[J]. Mechanical Systems and Signal Processing, 2016, 72: 65-79.
[6] Wang Y, Xu G, Liang L, et al. Detection of weak transient signals based on wavelet packet transform and manifold learning for rolling element bearing fault diagnosis[J]. Mechanical Systems and Signal Processing, 2015, 54: 259-276.
[7] Zhang Q, Zhao G, Shu L, et al. Research of dimensionless index for fault diagnosis positioning based on EMD[J]. Journal of Computers, 2016, 27(1): 62-73.
[8] Žvokelj M, Zupan S, Prebil I. EEMD-based multiscale ICA method for slewing bearing fault detection and diagnosis[J]. Journal of Sound and Vibration, 2016, 370: 394-423.
[9] Gupta K K, Raju K S. Bearing fault analysis using variational mode decomposition[C]//Industrial and Information Systems (ICIIS),2014 9th International Conference on. IEEE, 2014: 1-6.
[10] Dragomiretskiy K, Zosso D. Variational mode decomposition[J]. IEEE transactions on signal processing, 2014, 62(3): 531-544.
[11] 钱林,康敏,傅秀清,王兴盛,费秀国. 基于VMD的自适应形态学在轴承故障诊断中的应用[J].振动与冲击,2017,(03):227-233.
Qian Lin, Kang Min, Fu Xiu-qing et al.Application of adaptivemorphology in bearing fault diagnosis based on VMD[J]. Journal of Vibration and Shock, 2017,(03):227-233.
[12] 王永鑫,贾珈,张雨辰,蔡莲红. 基于HMM语音合成的语调控制[J].清华大学学报(自然科学版),2013,(06):781-786.
Wang Yong-xin, Jia Jia, Zhang Yu-chen et al. Contronl of intonation in HMM based text-to-speech systems[J]. Tsinghua Univ(Sci & Tech),2013,(06):781-786.
[13] 郭明威,倪世宏,朱家海. 基于EMD-HMM的BIT间歇故障识别[J].振动.测试与诊断,2012,(03):467-470+518.
Guo Ming-wei, Ni Shi-hong, Zhu Jia-hai. Contronl of intonation in HMM based text-to-speech systems[J]. Journal of Vibration,Measurment&Diagnosis,2012,(03):467-470+518.
[14] 李庆,LIANG Steven Y,杨建国. 谐波小波样本熵与HMM模型的轴承故障模式识别[J].上海交通大学学报,2016,(05):723-729+735.
Li Qing, LIANG Steven Y, Yang Jian-guo. Bearing Fault Pattern Recognition Using Harmonic Wavelet Sample Entropy and Hidden Markov Model[J]. Journal of Shanghai Jiao Tong University,2016,(05):723-729+735.