Mechanical fault diagnosis method based on probability amplitude demodulation

PDF(2474 KB)

Journal of Vibration and Shock ›› 2022, Vol. 41 ›› Issue (14) : 218-225.

LI Zhinong1， LIU Xiaoyu1， XU Qiaoqiao1， GU Shipeng2, MA Yaping2

Author information +

History +

Abstract

Based on a large number of AM and FM signals in mechanical equipment when mechanical faults occur, a mechanical fault diagnosis method based on probability amplitude demodulation is proposed. In the proposed method, probability theory is used to transform demodulation into a reasoning process. Then some prior knowledge is used to infer the uncertain problems to achieve the purpose of inference and demodulation. The simulation results show that the proposed method is significantly better than the traditional Hilbert envelope demodulation method. and the demodulated envelope obtained by the proposed method has a high degree of fitting with the original signal. Finally, the experiment of rolling bearing’s fault diagnosis further verifies the effectiveness of the proposed method, and the proposed method can effectively extract the fault features of rolling bearing.
Key words: Probability amplitude demodulation; Bayesian inference; fault diagnosis; rolling bearing.

Key words

Probability amplitude demodulation / Bayesian inference / fault diagnosis / rolling bearing.

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

LI Zhinong1， LIU Xiaoyu1， XU Qiaoqiao1， GU Shipeng2, MA Yaping2. Mechanical fault diagnosis method based on probability amplitude demodulation[J]. Journal of Vibration and Shock, 2022, 41(14): 218-225

References

[1] 于德介, 程军圣, 杨宇.机械故障诊断的HILBERT-HUANG变换方法[M]. 科学出版社, 2006.
YU De-jie,CHENG Jun-sheng,YANG Yu.The Manchine Fault Diagnosis Approach Based on theHilbert-HuangTransform[M].Beijing:Science Press,2006.75-76
[2] C. Chatfield. The Analysis of Time Series: An Introduction. Chapman & Hall/CRC,sixth edition, July 2003. ISBN 1584883170.
[3] 王惠中, 张岳.基于改进希尔伯特黄变换的电机故障特征提取方法研究[J].自动化与仪器仪表, 2014(07): 42-48.
Wang huizhiong, Zhang yue.Based on the improved Hilbert huang transform motor fault feature extraction method of research[J].Automation and instrumentation, 2014(07): 42-48.
[4] Turner R E. Statistical models for natural sounds[D]. University College London, 2010.
[5] Turner R E, Sahani M. Probabilistic amplitude demodulation[C]//International Conference on Independent Component Analysis and Signal Separation. Springer, Berlin, Heidelberg, 2007: 544-551
[6] Turner R E, Sahani M. Probabilistic amplitude and frequency demodulation[C]// Proceedings of the 24th International Conference on Neural Information Processing Systems.[S.l.] :NIPS,2011.
[7] Turner R E, Sahani M. Demodulation as Probabilistic Inference[J]. Audio Speech & Language Processing IEEE Transactions on, 2011, 19(8):2398-2411.
[8] Turner R E, Sahani M. Decomposing signals into a sum of amplitude and frequency modulated sinusoids using probabilistic inference[C]// 2012 IEEE International Conference on Acoustics, Speech and Signal Processing. Kyoto: IEEE, 2012.
[9] Turner R E, Sahani M. Statistical inference for single- and multi-band Probabilistic Amplitude Demodulation[C]// IEEE International Conference on Acoustics Speech & Signal Processing. IEEE, 2010:5466-5469.
[10] Vamoş C, Crăciun M. Numerical demodulation of a Gaussian white noise modulated in amplitude by a deterministic volatility[J]. European Physical Journal B, 2013, 86(4):1-8.
[11] C. E. Rasmussen & C. K. I. Williams, Gaussian Processes for Machine Learning[M]. Cambridge:MIT Press, 2006.
[12]万广通. 基于 LMD 和带通滤波解调变工下滚动轴承故障特征提取方法的研究[D]. 北京交通大学, 2018.
Wan guangtong.Based on LMD and band-pass filtering demodulation conditions of rolling bearing fault feature extraction method is studied[D].Beijing jiaotong university,2018.
[13]王平, 廖明夫. 滚动轴承故障诊断的自适应共振解调技术[J]. 航空动力学报, 2005, 20(4):606-612.
Wangping,Liao mingfu.,Adaptive Demodulated Resonance Technique for the Rolling Bearing Fault Diagnosis[J].Journal of Aerospace Power, 2005, 20(4):606-612.
[14]Hariprasad J, Roy U M, Ramasangu H. Design and Implementation of Universal Digital Demodulator Using Neural Network Approaches[M]//Emerging Trends in Photonics, Signal Processing and Communication Engineering. Springer, Singapore, 2020: 133-139.