Bearing fault diagnosis based on the synchrosqueezed S transform and ensemble deep ridgelet auto-encoder

PDF(1920 KB)

Journal of Vibration and Shock ›› 2020, Vol. 39 ›› Issue (14) : 59-68.

DU Xiaolei1,2，CHEN Zhigang1,2，WANG Yanxue1,2

Author information +

History +

Abstract

Aiming at the problems of traditional fault diagnosis algorithms of rolling bearings having such shortcomings as largely dependent on expert prior knowledge and difficulty in fault feature extraction, a method based on synchrosqueezed S transform (SSST) and ensemble deep ridgelet auto-encoder (EDRAE) was proposed. Firstly, the vibration signals were transformed by SSST to get time-frequency images. Then time-frequency images were compressed by two-directional two dimensional principal components analysis (TD-2DPCA). Secondly, different ridgelet auto-encoders (RAEs) with different ridgelet functions were designed. Then, different deep ridgelet auto-encoders (DRAEs) with different RAEs were designed and a "cross-layer" connection was introduced to alleviate gradient disappearance of DRAEs. Finally, the compressed images were input into each DRAE for unsupervised pre-training and supervised fine-tuning, and the recognition result was given by the weighted averaging method. Experimental results show that the proposed method can effectively identify the bearing faults under multiple working conditions and multiple fault severities. The proposed method has better ability of feature extraction and recognition than artificial neural network, deep belief network, deep auto-encoder and so on.

Key words

synchrosqueezed S transform / ridgelet auto-encoder / rolling bearing / fault diagnosis

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

DU Xiaolei1,2，CHEN Zhigang1,2，WANG Yanxue1,2. Bearing fault diagnosis based on the synchrosqueezed S transform and ensemble deep ridgelet auto-encoder[J]. Journal of Vibration and Shock, 2020, 39(14): 59-68

References

[1] 姚德臣,杨建伟,程晓卿,等.基于多尺度本征模态排列熵和SA-SVM的轴承故障诊断研究[J].机械工程学报,2018,54(09):168-176.
YAO Dechen, YANG Jianwei, CHENG Xiaoqing, et al. Railway rolling bearing fault diagnosis based on muti-scale IMF permutation entropy and SA-SVM classifier[J]. Journal of Mechanical Engineering, 2018, 54(09): 168-176.
[2] 赵明. 变转速下机械动态信息的自适应提取与状态评估[D].西安: 西安交通大学,2017:30-34.
ZHAO Ming. Adaptive dynamic information extraction and state evaluation for machinery under varying speeds[D].Xian: Xi'an Jiaotong University, 2017:30-34.
[3] 程军圣,王健,桂林.改进的EEMD方法及其在滚动轴承故障诊断中的应用[J].振动与冲击,2018,37(16):51-56.
CHENG Junsheng, WANG Jian, GUI Lin. The improved EEMD analysis method and its application to fault diagnosis[J]. Journal of Vibration and Shock, 2018,37(16):51-56.
[4] 张颖,张超,王天正,等.基于改进固有时间尺度分解和谱峭度的轴承故障诊断方法[J].太阳能学报,2017,38(03):699-706.
ZHANG Ying, ZHANG Chao, WANG Tianzheng, et al. Fault diagnosis method for bearing based on improved intrinsic time-scale decomposition and spectrum kurtosis[J]. Acta Energiae Solaris Sinica, 2017, 38(03):699-706.
[5] 王志坚,吴文轩,马维金,等.基于LMD-MS的滚动轴承微弱故障提取方法[J].振动.测试与诊断,2018, 38(05): 1014-1020+1085.
WANG Zhijian, WU Wenxuan, MA Weijin, et al. Fault signal extraction method of rolling bearing weak fault based on LMD-MS[J]. Journal of Vibration,Measurement & Diagnosis, 2018, 38(05): 1014-1020+1085.
[6] 刘景良,郑锦仰,林友勤,等.变分模态分解和同步挤压小波变换识别时变结构瞬时频率[J].振动与冲击,2018,37(20):24-31.
LIU Jingliang, ZHENG Jinyang, LIN Youqin, et al.   Instantaneous frequency identification of time-varying structures using variational mode decomposition and synchrosqueezing wavelet transform[J]. Journal of Vibration and Shock, 2018,37(20):24-31.
[7] 颜丙生,聂士杰,汤宝平,等.基于阶次分析和EWT的轴承故障诊断研究[J].组合机床与自动化加工技术,2018(07):51-54.
YAN Bingsheng, NIE Shijie, TANG Baoping, et al. Research on bearing fault diagnosis based on order analysis and EWT[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2018(07):51-54.
[8] 陈向民,张亢,晋风华,等.基于CPP与S变换的自适应时频滤波及其在滚动轴承故障诊断中的应用[J].航空动力学报,2018,33(01):147-155.
CHEN Xiangmin, ZHANG Kang, JIN Fenghua, et al. Adaptive time-frequency filtering method based on CPP and S transform and its application in fault diagnosis of rolling bearing[J]. Journal of Aerospace Power, 2018,33(01):147-155.
[9] 沈微,陶新民,高珊,等.基于同步挤压小波变换的振动信号自适应降噪方法[J].振动与冲击, 2018, 37(14): 239-247.
SHEN Wei, TAO Xinmin, GAO Shan, et al. Self-adaptive de-noising algorithm for vibration signals based on synchrosqueezed wavelet transforms[J]. Journal of   Vibration and Shock, 2018, 37(14): 239-247.
[10] 黄忠来,张建中.同步挤压S变换[J].中国科学:信息科学,2016,46(05):643-650.
HUANG Zhonglai, ZHANG Jianzhong. Synchrosqueezing S-transform [J]. Scientia Sinica   (Informationis), 2016, 46(05): 643-650.
[11] SHAO H D, JIANG H K, HUI F W, et al. An enhancement deep feature fusion method for rotating machinery fault diagnosis [J]. Knowledge-Based Systems, 2017, 119: 200-220.
[12] JURGEN S. Deep learning in neural networks: An   overview [J]. Neural Networks, 2015, 61(1): 85-117.
[13] 侯文擎,叶鸣,李巍华.基于改进堆叠降噪自编码的滚动轴承故障分类[J].机械工程学报,2018,54(07):87-96.
HOU Wenqing, YE Ming, LI Weihua. Rolling element bearing fault classification using improved stacked de-noising auto-encoders[J]. Journal of Mechanical Engineering, 2018,54(07):87-96.
[14] 雷亚国,贾峰,周昕,等.基于深度学习理论的机械装备大数据健康监测方法[J].机械工程学报, 2015, 51(21): 49-56.
LEI Yaguo, JIA Feng, ZHOU Xin, et al. A deep learning-based method for machinery health monitoring with big data[J]. Journal of Mechanical Engineering, 2015, 51(21): 49-56.
[15] 李恒,张氢,秦仙蓉,等.基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法[J].振动与冲击, 2018, 37(19): 124-131.
LI Heng, ZHANG Qing, QIN Xianrong, et al. Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network[J]. Journal of Vibration and Shock, 2018, 37(19): 124-131.
[16] 王丽华,谢阳阳,周子贤,等.基于卷积神经网络的异步电机故障诊断[J].振动.测试与诊断, 2017, 37(06): 1208- 1215+1283.
WANG Lihua, XIE Yangyang, ZHOU Zixian, et al. Motor fault diagnosis based on convolutional neural networks[J]. Journal of Vibration,Measurement & Diagnosis, 2017, 37(06): 1208- 1215+1283.
[17] SHAO H D, JIANG H K, ZHAO K W, et al. A novel tracking deep wavelet auto-encoder method for intelligent fault diagnosis of electric locomotive bearings[J]. Mechanical Systems and Signal Processing, 2018, 110(9): 193-209.
[18] SHAO H D, JIANG H K, ZHAO H W, et al. A novel deep autoencoder feature learning method for rotating machinery fault diagnosis [J]. Mechanical Systems and Signal Processing, 2017, 95: 187-204.
[19] SHAO H D, JIANG H K, LIN Y, et al. A novel method for intelligent fault diagnosis of rolling bearings using ensemble deep auto-encoders[J]. Mechanical Systems and Signal Processing, 2018,102(3):278-297.
[20] CANDES E J. Ridgelet: Theory and Applications[D]. California: Stanford University, 1988:87-99.
[21] 石程. 基于视觉稀疏表示和深度脊波网络的遥感图像融合及分类[D].西安:西安电子科技大学,2016:47-54.
SHI Cheng. Visual sparse representation and deep ridgelet networks based remote sensing image fusion and classification[D]. Xian: Xidian University,2016:47-54.
[22] 马苗,许西丹,武杰.一种耦合深度信念网络的图像识别方法[J].西安电子科技大学学报,2018,45(05):102-107.
MA Miao, XU Xidan, WU Jie. Coupled-deep belief network based method for image recognition[J]. Journal of Xidian University, 2018,45(05):102-107.
[23] 韩霜,吴奇,孙礼兵,等.基于深度收缩自编码网络的飞行员疲劳状态识别[J].生物医学工程学杂志, 2018, 35(03): 443-451.
HAN Shuang, WU Qi, SUN Libing, et al. Recognition of fatigue status of pilots based on deep contractive sparse auto-encoding network[J]. Journal of Biomedical Engineering, 2018, 35(03): 443-451.
[24] GOODFELLOW I, BENGIO Y, COURVILLE A. Deep Learning[M]. Cambridge: MIT Press, 2016: 93-94.
[25] 张世雄,蔡艳平,石林锁,等.基于离散广义S变换与双向二维主成分分析的内燃机故障诊断[J].中国机械工程,2018,29(08):899-905.
ZHANG Shixiong, CAI Yanping, SHI Linsuo, et al. I.C.Engine fault diagnosis based on discrete generalized S-transformation and TD-2DPCA[J]. China Mechanical Engineering, 2018, 29(08) : 899-905.
[26] JIA F, LEI Y G, LIN J, et al. Deep neural networks: A promising tool for fault characteristic mining and   intelligent diagnosis of rotating machinery with massive data[J] . Mechanical Systems and Signal Processing, 2016, 72-73: 303-315.
[27] QU J X, ZHANG Z S, GONG T. A novel intelligent   method for mechanical fault diagnosis based on dual-tree complex wavelet packet transform and multiple classifier fusion [J]. Neurocomputing, 2016, 171: 837-853.