Bearing fault diagnosis based on improved sparse filter and deep network fusion

doi:10.3785/j.issn.1008-973X.2020.12.004

Journal of ZheJiang University (Engineering Science)

2020, Vol. 54

Issue (12): 2301-2309 DOI: 10.3785/j.issn.1008-973X.2020.12.004

Bearing fault diagnosis based on improved sparse filter and deep network fusion

Mei-ying QIAO(

),Xia-xia TANG(

),Shu-hao YAN,Jian-ke SHI

College of Electrical Engineering and Automation, Henan Polytechnic University, Jiaozuo 454000, China

Download:

HTML

PDF(2061KB) HTML
Export: BibTeX | EndNote (RIS)

Abstract

An improved model combining sparse filtering and deep dilated gate convolutional network was proposed in order to solve the problem that feature extraction relies on manual experience when rolling bearing faults occur, and that the fault category was difficult to automatically and accurately identify. Sliding window was used to sample bearing vibration signals with time series characteristics in order to avoid over fitting. Heteroscedasticity was eliminated and data features were extracted by improving the sparse filtering of the objective function in order to shorten the calculation time and improve the accuracy of classification. The fault classification model was established by combining the dilated gate convolution and the bidirectional LSTM network, and the data noise can be filtered out. Data experiments from Case Western Reserve University and laboratory power equipment were compared. Results show that the fault diagnosis accuracy rate of this model can reach 98%. Different load and different SNR experiments with bearing vibration signals show that the model has generalization and anti-noise performance.

Key words： feature extraction sparse filter dilated gate convolution bidirectional LSTM fault classification anti-noise performance

Received: 09 January 2020 Published: 31 December 2020

CLC:

TM 307

	Service
	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Mei-ying QIAO
	Xia-xia TANG
	Shu-hao YAN
	Jian-ke SHI

Cite this article:

Mei-ying QIAO,Xia-xia TANG,Shu-hao YAN,Jian-ke SHI. Bearing fault diagnosis based on improved sparse filter and deep network fusion. Journal of ZheJiang University (Engineering Science), 2020, 54(12): 2301-2309.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2020.12.004 OR http://www.zjujournals.com/eng/Y2020/V54/I12/2301

基于改进稀疏滤波与深度网络融合的轴承故障诊断

针对滚动轴承故障时特征提取依赖人工经验，以及故障类别难以自动准确识别的问题，提出了一种改进稀疏滤波和深层空洞门卷积网络相结合的故障诊断模型. 采用滑动窗对具有时序特征的轴承振动信号进行采样处理以避免过拟合；通过改进目标函数的稀疏滤波消除数据中的异方差并提取数据特征，达到缩短计算时间和提高分类准确率的效果；利用空洞门卷积和双向LSTM网络对噪声进行滤除，同时进行故障分类识别. 对比凯斯西储大学和动力系统装置的轴承实验数据，显示该模型故障诊断准确率可达98%. 不同负载和不同信噪比的轴承振动信号实验，表明该模型具有泛化性和抗噪性.

关键词： 特征提取, 稀疏滤波, 空洞门卷积, 双向LSTM, 故障分类, 抗噪性

Fig.1 Original data scatter diagram

Fig.2 Scatter graph based on traditional sparse filtering

Fig.3 Scatter graph based on improved sparse filtering

Fig.4 Dilated gate convolution network structure graph

Fig.5 Dilated convolution with different dilate rates

Fig.6 ISP-DGLSTM network flow chart

Fig.7 CWRU test-bed

Tab.1 Bearing fault data set of CWRU

Fig.8 Driveline diagnostic simulator test bench

Tab.2 Results comparison of four network models

Fig.9 CWRU bearing data signal figure

Tab.3 Parameter setting of each layer of ISP-DGLSTM network model

Fig.10 Four kinds of models accuracy compared by curve under CWRU data

Fig.11 Driveline bearing data signal figure

Fig.12 Four kinds of models accuracy compared by curve under driveline data

Fig.13 Four kinds of model accuracy figure under CWRU data

Fig.14 Four kinds of model accuracy figure under driveline data

Fig.15 Add signal diagram with signal-to-noise ratio to CWRU data

Fig.16 Accuracy comparison curve with signal-to-noise ratio of 1 added to Case Western Reserve University data

Fig.17 Accuracy comparison curve with signal-to-noise ratio of −4 added to Case Western Reserve University data


[1]	朱会杰, 王新晴, 芮挺, 等基于平移不变CNN的机械故障诊断研究[J]. 振动与冲击, 2019, 38 (5): 45- 52 ZHU Hui-jie, WANG Xin-qing, RUI Ting, et al Machinery fault diagnosis based on shift invariant CNN[J]. Journal of Vibration and Shock, 2019, 38 (5): 45- 52

[2]	杨宇, 罗鹏, 甘磊, 等. SADBN及其在滚动轴承故障分类识别中的应用[J]. 振动与冲击, 2019, 38(15): 11-16. YANG Yu, LUO Peng, GAN Lei, et al. SADBN and its application in rolling bearing fault identification and classification [J]. Journal of Vibration and Shock, 2019, 38(15): 11-16.

[3]	JORDI B V, RUBEN P, JAVIER M R, et al Short-frequency Fourier transform for fault diagnosis of induction machines working in transient regime[J]. IEEE Transactions on Instrumentation and Measurement, 2017, 66 (3): 432- 440 doi: 10.1109/TIM.2016.2647458

[4]	程卫东, 赵德尊用于滚动轴承转频估计的EMD软阈值降噪算法[J]. 浙江大学学报: 工学版, 2016, 50 (3): 428- 435 CHENG Wei-dong, ZHAO De-zun EMD soft-thresholding denoising algorithm for rolling element bearing rotational frequency estimation[J]. Journal of Zhejiang University: Engineering Science, 2016, 50 (3): 428- 435

[5]	SHUAI J, LEE S K Bearing fault detection utilizing group delay and the Hilbert-Huang transform[J]. Journal of Mechanical Science and Technology, 2017, 31 (3): 1089- 1096 doi: 10.1007/s12206-017-0208-z

[6]	杨先勇, 周晓军, 张文斌, 等基于形态小波和S变换的滚动轴承故障特征提取[J]. 浙江大学学报: 工学版, 2010, 44 (11): 2088- 2092 YANG Xian-yong, ZHOU Xiao-jun, ZHANG Wen-bin, et al Rolling bearing fault feature extraction based on morphological wavelet and S-transform[J]. Journal of Zhejiang University: Engineering Science, 2010, 44 (11): 2088- 2092

[7]	周奇才, 沈鹤鸿, 赵炯, 等基于改进堆叠式循环神经网络的轴承故障诊断[J]. 同济大学学报: 自然科学版, 2019, 47 (10): 1500- 1507 ZHOU Qi-cai, SHEN He-hong, ZHAO Jiong, et al Bearing fault diagnosis based on improved stacked recurrent neural network[J]. Journal of Tongji University: Natural Science, 2019, 47 (10): 1500- 1507

[8]	ZHUANG Z, LV H, XU, et al A deep learning method for bearing fault diagnosis through stacked residual dilated convolutions[J]. Applied Sciences, 2019, 9 (9): 1823 doi: 10.3390/app9091823

[9]	田源. 基于深度学习的中文命名实体识别方法研究[D]. 长沙: 湖南大学, 2018: 42-56. TIAN Yuan. The research of Chinese named entity recognition based on deep learning [D]. Changsha: Hunan University, 2018: 42-56.

[10]	LEI Y, JIA F, LIN J, et al An intelligent fault diagnosis method using unsupervised feature learning towards mechanical big data[J]. IEEE Transactions on Industrial Electronics, 2016, 63 (5): 3137- 3147 doi: 10.1109/TIE.2016.2519325

[11]	于洋, 何明, 刘博, 等基于TL-LSTM的轴承故障声发射信号识别研究[J]. 仪器仪表学报, 2019, 40 (5): 51- 59 YU Yang, HE Ming, LIU Bo, et al Research on acoustic emission signal recognition of bearing fault based on TL-LSTM[J]. Chinese Journal of Scientific Instrument, 2019, 40 (5): 51- 59

[12]	NGIAM J, CHEN Z, BHASKAR S A, et al. Sparse filtering [C] // Proceedings of Advances in Neural Information Processing Systems. Cambridge: MIT Press, 2011: 1125-1133.

[13]	陈赟基于稀疏滤波的信号识别[J]. 电信快报, 2015, (8): 46- 48 CHEN Yun signal recognition based on sparse filtering[J]. Telecom Express, 2015, (8): 46- 48

[14]	邱立达, 刘天键, 傅平基于深度稀疏滤波的目标跟踪[J]. 计算机辅助设计与图形学学报, 2017, 29 (3): 459- 468 QIU Li-da, LIU Tian-jian, FU Ping Target tracking based on deep sparse filtering[J]. Journal of Computer-Aided Design and Computer Graphics, 2017, 29 (3): 459- 468 doi: 10.3969/j.issn.1003-9775.2017.03.008

[15]	ALEXIS B, PRASAD S, FRéDéRIC B, et al Well-posedness of the permutation problem in sparse filter estimation with ?p minimization [J]. Applied and Computational Harmonic Analysis, 2013, 35 (3): 394- 406 doi: 10.1016/j.acha.2012.11.001

[16]	何金洋基于稀疏滤波和神经网络的人脸识别算法[J]. 理论探讨, 2016, 7 (5): 28- 31 HE Jin-yang The facial recognition algorithm based on sparse filtering-neural networks[J]. Theoretical Investigation, 2016, 7 (5): 28- 31

[17]	李永国. 变量预测模型模式识别方法及其在滚动轴承故障诊断中的应用[D]. 长沙: 湖南大学, 2015: 6-12. LI Yong-guo. Variable predictive model based class discriminate and its application in fault diagnosis of roller bearing [D]. Changsha: Hunan University, 2015: 6-12.

[18]	YANG H, LV J, GUO C Weighted composite quantile regression estimation and variable selection for varying coefficient models with heteroscedasticity[J]. Journal of the Korean Statistical Society, 2015, 44 (1): 77- 94 doi: 10.1016/j.jkss.2014.05.005

[19]	HILL J B Robust generalized empirical likelihood for heavy tailed autoregressions with conditionally heteroscedastic errors[J]. Journal of Multivariate Analysis, 2015, 135: 131- 152 doi: 10.1016/j.jmva.2014.12.008

[20]	杨荣芳, 周惠靶场光学测量数据异方差性检验及修正[J]. 指挥控制与仿真, 2008, 30 (4): 114- 116 YANG Rong-fang, ZHOU Hui Test and adjust heteroscedasticity of optical measure data in proof range[J]. Command Control and Simulation, 2008, 30 (4): 114- 116 doi: 10.3969/j.issn.1673-3819.2008.04.031

[21]	HALUNGA A, ORME C D, YAMAGATA T A heteroskedasticity robust Breusch-Pagan test for contemporaneous correlation in dynamic panel data models[J]. Journal of Econometrics, 2017, 198 (2): 209- 230 doi: 10.1016/j.jeconom.2016.12.005

[22]	万书亭, 佟海侠, 董炳辉基于最小二乘支持向量机的滚动轴承故障诊断[J]. 振动. 测试与诊断, 2010, 30 (2): 149- 152 WAN Shu-ting, TONG Hai-xia, DONG Bing-hui Bearing fault diagnosis using wavelet packet transform and least square support vector machines[J]. Journal of Vibration, Measurement and Diagnosis, 2010, 30 (2): 149- 152

[23]	GE F, JU Y, QI Z, et al Parameter estimation of a Gaussian mixture model for wind power forecast error by Riemann L-BFGS optimization[J]. IEEE Access, 2018, (6): 38892- 38899

[24]	李旭芳, 段春林, 张冬波, 等遥测数据时间序列滑动窗口动态分割技术[J]. 飞行器测控学报, 2015, 34 (4): 345- 349 LI Xu-fang, DUAN Chun-lin, ZHANG Dong-bo, et al A dynamic slide windows segmentation technology for telemetry time series[J]. Journal of Spacecraft TT and C Technology, 2015, 34 (4): 345- 349

[25]	YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions [C] // International Conference on Learning Representations. San Juan, Puerto Rico: ICLR, 2016.

[26]	LORENZO J, BARRA R, YAMAGISHI J, et al Improving Spanish speech synthesis intelligibility under noisy environments[J]. Journal of the Acoustical Society of America, 2016, 140 (4): 2961- 2962

[27]	LONG J, SHELHAMER E, DARRELL T Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 39 (4): 640- 651

[28]	NAIM I, RILEY P, GILDEA D Feature-based decipherment for machine translation[J]. Computational Linguistics, 2018, 44 (2): 1- 33

[29]	张利刚. 基于全空洞卷积神经网络的图像语义分割[D]. 长春: 东北师范大学, 2018: 9-15. ZHANG Li-gang. Image semantic segmentation based on full cavitation convolutional neural network [D]. Changchun: Northeast Normal University, 2018: 9-15.

[30]	张新峰, 闫昆鹏, 赵珣基于双向LSTM的手写文字识别技术研究[J]. 南京师大学报, 2019, 42 (3): 58- 64 ZHANG Xin-feng, YAN Kun-peng, ZHAO Xun Handwriting Chinese text recognition using BiLSTM network[J]. Journal of Nanjing Normal University, 2019, 42 (3): 58- 64

[31]	YANG Y, FU P, HE Y Bearing fault automatic classification based on deep learning[J]. IEEE Access, 2018, (6): 71540- 71554

[32]	乔美英, 刘宇翔, 兰建义基于VMD和马氏距离SVM的滚动轴承故障诊断[J]. 中山大学学报: 自然科学版, 2019, 58 (5): 8- 16 QIAO Mei-ying, LIU Yu-xiang, LAN Jian-yi Fault diagnosis method of rolling bearings based on VMD and mahalanobis distance SVM[J]. Journal of Sun Yat-sen University: Medical Sciences, 2019, 58 (5): 8- 16

[1]	Fang LIU,Zhen WANG,Rui-di LIU,Kai WANG. Short-term forecasting method of wind power generation based on BP neural network with combined loss function[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(3): 594-600.

[2]	Yi-xiong FENG,Kang-jie LI,Yi-cong GAO,Hao ZHEN,Jian-rong TAN. Shaft surface defect detection method based on feature and morphology reconstruction[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(3): 427-434.

[3]	Xiao-hu ZHAO,Liang-fei YIN,Cheng-long ZHAO. Image captioning based on global-local feature and adaptive-attention[J]. Journal of ZheJiang University (Engineering Science), 2020, 54(1): 126-134.

[4]	Gui-ran HE,Qi LI,Hua-jun FENG,Zhi-hai XU,Yue-ting CHEN. Dual-focal camera continuous digital zoom based onCNN and feature extraction[J]. Journal of ZheJiang University (Engineering Science), 2019, 53(6): 1182-1189.

[5]	Yue DONG,Hua-jun FENG,Zhi-hai XU,Yue-ting CHEN,Qi LI. Attention Res-Unet: an efficient shadow detection algorithm[J]. Journal of ZheJiang University (Engineering Science), 2019, 53(2): 373-381.

[6]	Xiang-long ZHAO,Jie CHEN,Rong-jing HONG,Hua WANG,Yuan-yuan LI. Adaptive feature extraction method for slewing bearing based on Wavelet leader and optimized isometric mapping method[J]. Journal of ZheJiang University (Engineering Science), 2019, 53(11): 2092-2101.

[7]	TONG Shui-guang, ZHANG Yi-dong, XU Jian, CONG Fei-yun. Spectral band refined composite multiscale fuzzy entropy and its application in fault diagnosis of rolling bearings[J]. Journal of ZheJiang University (Engineering Science), 2018, 52(8): 1509-1516.

[8]	WANG Ri-jun, ZHAO Chang-jun, BAI Yue, ZENG Zhi-qiang, DU Wen-hua, DUAN Neng-quan. Faults detection and self-reconfiguration for execution units of Hex-Rotor unmanned aerial vehicle based on multiple fault classification[J]. Journal of ZheJiang University (Engineering Science), 2018, 52(7): 1406-1414.

[9]	YU Yong, ZHOU Yang, CAO Peng, ZHAO Gang. In-process model construction method based on model-based definition model[J]. Journal of ZheJiang University (Engineering Science), 2018, 52(6): 1025-1034.

[10]	ZHU Dong-yang, SHEN Jing-yi, HUANG Wei-ping, LIANG Jun. Fault classification based on modified active learning and weighted SVM[J]. Journal of ZheJiang University (Engineering Science), 2017, 51(4): 697-705.

[11]	TAN Hailong, LIU Kangling, JIN Xin, SHI Xiang rong, LIANG Jun. Multivariate time series classification based on μσ-DWC feature and tree-structured M-SVM[J]. Journal of ZheJiang University (Engineering Science), 2015, 49(6): 1061-1069.

[12]	BAI Fan, ZHENG Hui feng, SHEN Ping ping, WANG Cheng, YU Sang sang. Plant species identification method based on flower feature coding classification[J]. Journal of ZheJiang University (Engineering Science), 2015, 49(10): 1902-1908.

[13]	WANG Meng, LIN Lan-fen, WANG Feng. Short text expansion and classification based on pseudo-relevance feedback [J]. Journal of ZheJiang University (Engineering Science), 2014, 48(5): 2-.

[14]	WANG Meng, LIN Lan-fen, WANG Feng. Short text expansion and classification based on pseudo-relevance feedback [J]. Journal of ZheJiang University (Engineering Science), 2014, 48(10): 1835-1842.

[15]	SHI Jin-he, SHENG Ji-zhong, WANG Pan. Feature extraction and classification of four-class motor imagery EEG data[J]. Journal of ZheJiang University (Engineering Science), 2012, 46(2): 338-344.

Viewed

Full text

Abstract

Cited

Shared

Discussed