GRU prediction forperformance degradation of rolling bearings based on optimalwavelet packet and Mahalanobis distance

PDF(1281 KB)

Journal of Vibration and Shock ›› 2020, Vol. 39 ›› Issue (17) : 39-46.

ZHENG Xiaoxia, QIAN Yiqun, WANG Shuai

Author information +

History +

Abstract

To effectively describe performance degradation trend of rolling bearings, a new method based on optimal wavelet packet and Mahalanobis distance (MD) for rolling bearing performance degradation gated recurrent unit (GRU) prediction was proposed. The wavelet packet method was used to extract energy features of rolling bearing data considering effects of decomposition layers and wavelet basis functions on effect of feature extraction, and establish the principle of optimal decomposition layer number. According to the variation rate of energy fluctuation, the optimal wavelet basis function was selected. Then, MDs among feature vectors were calculated, and these MDs were taken as bearing performance degradation indexes. Finally, these indexes were taken as input of GRU to construct a prediction model of degradation trend. The actually measured rolling bearing full life data were used to verify the prediction model. Results showed that the proposed method can gain correct prediction results; it is beneficial for equipment maintenance personnel to better grasp degradation trend of rolling bearings.

Key words

trend prediction / optimal wavelet packet / Mahalanobis distance (MD) / gated recurrent unit (GRU)

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

ZHENG Xiaoxia, QIAN Yiqun, WANG Shuai. GRU prediction forperformance degradation of rolling bearings based on optimalwavelet packet and Mahalanobis distance[J]. Journal of Vibration and Shock, 2020, 39(17): 39-46

References

[1]Rai A, Upadhyay S H. The use of MD-CUMSUM and NARX neural network for anticipating the remaining useful life of bearings [J]. Measurement, 2017, 111.
[2]Gu X, Yang S, Liu Y, et al. Rolling element bearing faults diagnosis based on kurtogram and frequency domain correlated kurtosis [J]. Measurement Science & Technology, 2016, 27(12):125019.
[3]陈昌, 汤宝平, 吕中亮. 基于威布尔分布及最小二乘支持向量机的滚动轴承退化趋势预测 [J]. 振动与冲击, 2014, 33(20):52-56.
Chen chang, Tang Baoping, Lv Zhongliang. Degradation trend prediction of rolling bearings based on Weibull distribution and least squares support vector machine [J]. Journal of Vibration and Shock, 2014, 33(20):52-56.
[4]Zhu K. Performance degradation assessment of rolling element bearings based on hierarchical entropy and general distance [J]. Journal of Vibration and Control, 2017, 24(14):3194-3205.
[5]许迪, 葛江华, 王亚萍, 等. 流形学习和M-KH-SVR的滚动轴承衰退预测 [J]. 振动工程学报, 2018, 31(5):170-179.
Xu di, Ge Jianghua, Wang Yaping, et al. Prediction for rolling bearing performance degradation based on manifold learning and M-KH-SVR [J]. Journal of Vibration Engineering, 2018, 31(5):170-179.
[6]Wu J, Wu C, Cao S, et al. Degradation data-driven time-to-failure prognostics approach for rolling element bearings in electrical machines [J]. IEEE Transactions on Industrial Electronics, 2019, 66(1):529-539.
[7]王恒, 马海波, 徐海黎, 等. 机械设备性能退化评估与预测研究综述 [J]. 机械强度, 2013(6):716-723.
Wang Heng, Ma Haibo, Xu Haili, et al. Review on machinery performance degradation assessment and prognostics [J]. Journal of Mechanical Strength, 2013(6):716-723.
[8]Ming Dong, David He, Prashant Banerjee, et al. Equipment health diagnosis and prognosis using hidden semi Markov models [J]. The International Journal of Advanced Manufacturing Technology, 2006, 30(7-8):738-749.
[9]Lin J, Chen Q. Fault diagnosis of rolling bearings based on multifractal detrended fluctuation analysis and Mahalanobis distance criterion [J]. Mechanical Systems & Signal Processing, 2013, 38(2):515-533.
[10]周小龙, 刘薇娜, 姜振海, 等. 基于改进HHT和马氏距离的齿轮故障诊断 [J]. 振动与冲击, 2017, 36(22):218-224.
Zhou Xiaolong, Liu Weina, Jiang Zhenhai, et al. Great fault diagnosis based on improved HHT and Mahalanobis distance [J]. Journal of Vibration and Shock, 2017, 36(22):218-224.
[11]Yu J. Health condition monitoring of machine based on hidden markov model and contribution analysis [J]. IEEE Transactions on Instrumentation and Measurement, 2012, 61(8):2200-2211.
[12]Yuan N, Yang W, Kang B, et al. Manifold learning-based fuzzy k-principal curve similarity evaluation for wind turbine condition monitoring [J]. Energy Science & Engineering, 2018, 6(6):727-738.
[13]赵乾坤, 万小金, 徐增丙, 等. 基于集成软竞争ART的滚动轴承性能退化趋势预测 [J]. 机械传动, 2018, 42(1):131-136.
Zhao Qiankun, Wan Xiaojin, Xu Zengbin, et al. Regression trend prediction of rolling bearing performance based on integrated soft competition ART [J]. Journal of Mechanical Transmission, 2018, 42(1):131-136.
[14]张焱, 汤宝平, 熊鹏. 多尺度变异粒子群优化MK-LSSVM的轴承寿命预测 [J]. 仪器仪表学报, 2016, 37(11):2489-2496.
Zhang Miao, Tang Baoping, Xiong Peng. Rolling element bearing life prediction based on multi-scale mutation particle swarm optimized multi-kernel least square support vector machine [J]. Chinese Journal of Scientific Instrument, 2016, 37(11):2489-2496.
[15]陈法法, 杨勇, 马婧华, 等. 信息熵与优化LS-SVM的轴承性能退化模糊粒化预测 [J]. 仪器仪表学报, 2016, 37(4):779-787.
Chen Fafa, Yang Yong, Ma Jinghua, et al. Fuzzy granulation prediction for bearing performance degradation based on information entropy and optimized LS-SVM [J]. Chinese Journal of Scientific Instrument, 2016, 37(4):779-787.
[16]李时. 应用统计学 [M].北京:清华大学出版社,北京交通大学出版社, 2005.
Li Shi. Applied Statistics [M]. Beijing: Tsinghua University Press, Beijing Jiaotong University Press, 2005.
[17]王鑫, 吴际, 刘超,等. 基于LSTM循环神经网络的故障时间序列预测 [J]. 北京航空航天大学学报, 2018, 44(4):772-784.
Wang Xin, Wu Ji, Liu Chao, et al. Exploring LSTM based recurrent neural network for failure time series prediction [J]. Journal of Beijing University of Aeronautics and Astronautics, 2018, 44(4):772-784.
[18]牛哲文, 余泽远, 李波, 等. 基于深度门控循环单元神经网络的短期风功率预测模型 [J]. 电力自动化设备, 2018, 38(5):36-42.
Niu Wenzhe, Yu Zeyuan, Libo, et al. Short-term wind power forecasting model based on deep gated recurrent unit neural network [J]. Electric Power Automation Equipment, 2018, 38(5):36-42.
[19]李雪莲, 段鸿, 许牧,等. 基于门循环单元神经网络的中文分词法[J]. 厦门大学学报(自然版), 2017(2):237-243.
Li Xuelian, Duan Hong, Xu Mu, et al. A gated recurrent unit neural network for Chinese word segmentation. Journal of Xiamen University (Natural Science), 2017(2):237-243.
[20]Jozefowicz R, Zaremba W, Sutskever I. An empirical exploration of recurrent network architectures [C]. International Conference on International Conference on Machine Learning. JMLR.org, 2015:2342-2350.
[21]Maesschalck R D, Jouan-Rimbaud D, Massart D L. The Mahalanobis distance [J]. Chemometrics & Intelligent Laboratory Systems, 2000, 50(1):1-18.
[22]Campora U, Cravero C, Zaccone R. Marine gas turbine monitoring and diagnostics by simulation and pattern recognition [J]. International Journal of Naval Architecture and Ocean Engineering, 2017, 10(5):617-628.
[23]Cho K, Merrienboer B V, Gulcehre C, et al. Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation [J]. Computer Science, 2014.
[24]刘明宇, 吴建平, 王钰博, 等. 基于深度学习的交通流量预测 [J]. 系统仿真学报, 2018, 30(11):77-82+91.
Liu Mingyu, Wu Jianping, Wang Yubo, et al. Traffic flow prediction based on deep learning [J]. Journal of System Simulation, 2018, 30(11):77-82+91.
[25]周莽, 高僮, 李晨光, 等. GRU神经网络短期电力负荷预测研究 [J]. 科技创新与应用, 2018, 33:52-53+57.
Zhou Mang, Gao Tong, Li Chenguang, et al. Research on short-term power load forecasting method based on GRU neural network [J]. Technology Innovation and Application, 2018, 33:52-53+57.
[26]Lee J, Qiu H, Yu G, et al. Rexnord technical services, ‘bearing data set’ [R]. IMS, University of Cincinnati, NASA Ames Prognostics Data Repository, <http://ti.arc.nasa.gov/project/prognostics-data-repository>, NASA Ames, Moffett Field, CA, 2007.
[27]Pincus S M. Approximate entropy as a measure of system complexity[J]. Proceedings of the National Academy of Sciences of the United States of America, 1991, 88(6):2297-2301.