Residual life prediction of rolling bearing based on attention GRU algorithm
YAO Dechen1,2, LI Boyang1,2, LIU Hengchang1,2, YAO Juanjuan3, PI Yannan3
1.School of Machine-Electricity and Automobile Engineering, Beijing University of Civil Engineering Architecture, Beijing 100044, China;
2.Beijing Municipal Key Lab of Performance Guarantee on Urban Rail Transit Vehicles, Beijing University of Civil Engineering Architecture,
Beijing 100044, China; 3.Beijing Mass Transit Railway Operation Co., Ltd., Beijing 100044, China
Abstract：Aiming at the problem of it being difficult to accurately predict the change trend of rolling bearing residual life with time variation in rotating machinery, making full use of the ability of recurrent neural network (RNN) to process time series data, the attention gated recurrent unit (AGRU) algorithm based on attention mechanism was proposed to predict rolling bearing residual life.Firstly, multiple time-domain features were extracted from original vibration signals to construct a data set, and the data set was normalized.Secondly, the attention mechanism was introduced into GRU model.Finally, the feature data set was divided into a training set and a testing set.The training set was used to train the model and determine the optimal model parameters.The testing set was used to evaluate the effect of the model.The experimental results showed that the improved GRU model can effectively predict the change trend of residual life of different types of rolling bearing with time variation; it can provide a new idea for predicting residual life of rolling bearing components.
姚德臣，李博阳，刘恒畅，姚娟娟，皮雁南. 基于注意力GRU算法的滚动轴承剩余寿命预测[J]. 振动与冲击, 2021, 40(17): 116-123.
YAO Dechen, LI Boyang, LIU Hengchang, YAO Juanjuan, PI Yannan. Residual life prediction of rolling bearing based on attention GRU algorithm. JOURNAL OF VIBRATION AND SHOCK, 2021, 40(17): 116-123.
［1］RAI A，UPADHYAY S H.A review on signal processing techniques utilized in the fault diagnosis of rolling element bearings［J］.Tribology International，2016，96： 289-306.
［2］徐继亚, 王艳, 严大虎, 等.融合KPCA与信息粒化的滚动轴承性能退化SVM预测［J］.系统仿真学报,2018,30(6):2345-2354.
XU Jiya, WANG Yan, YAN Dahu, et al.SVM prediction of performance degradation of rolling bearing integrating KPCA and information granulation ［J］.Journal of System Simulation, 2018,30(6): 2345-2354.
［3］刘波, 宁芊, 刘才学, 等.基于连续型HMM和PSO-SVM的滚动轴承剩余寿命预测［J］.计算机应用,2019,39(增刊1):31-35.
LIU Bo, NING Qian, LIU Caixue, et al.Prediction of residual life of rolling bearing based on continuous HMM and PSO-SVM ［J］.Computer Application, 2019,39(Suppl.1): 31-35.
［4］马海龙.基于主元特征融合和SVM的轴承剩余寿命预测［J］.工矿自动化,2019,45(8):74-78.
MA Hailong.Prediction of bearing residual life based on principal component feature fusion and SVM ［J］.Industrial and Mining Automation, 2019,45(8): 74-78.
［5］者娜, 杨剑锋, 刘文彬, 等.KPCA和改进SVM在滚动轴承剩余寿命预测中的应用研究［J］.机械设计与制造,2019(11):1-4.
ZHE Na, YANG Jianfeng, LIU Wenbin, et al.Application of KPCA and improved SVM in residual life prediction of rolling bearing ［J］.Mechanical Design and Manufacturing, 2019 (11): 1-4.
［6］赵小强, 张青青, 陈鹏.基于PSO-BFA和改进Alexnet的滚动轴承故障诊断方法［J］.振动与冲击,2020,39(7):21-28.
ZHAO Xiaoqiang, ZHANG Qingqing, CHEN Peng.Fault diagnosis method of rolling bearing based on PSO-BFA and improved Alexnet ［J］.Journal of Vibration and Shock, 2020,39(7): 21-28.
［7］杨宇, 张娜, 程军圣.全参数动态学习深度信念网络在滚动轴承寿命预测中的应用［J］.振动与冲击,2019,38(10):199-205.
YANG Yu, ZHANG Na, CHENG Junsheng.Application of full parameter dynamic learning depth belief network in life prediction of rolling bearing ［J］. Journal of Vibration and shock, 2019,38(10): 199-205.
［8］张继冬, 邹益胜, 邓佳林, 等.基于全卷积层神经网络的轴承剩余寿命预测［J］.中国机械工程,2019,30(18):2231-2235.
ZHANG Jidong, ZOU Yisheng, DENG Jialin, et al.Prediction of bearing residual life based on full volume cumulation neural network ［J］.China Mechanical Engineering, 2019,30(18): 2231-2235.
［9］葛阳, 郭兰中, 牛曙光, 等.基于t-SNE和LSTM的旋转机械剩余寿命预测［J］.振动与冲击,2020,39(7):223-231.
GE Yang, GUO Lanzhong, NIU Shuguang, et al.Prediction of residual life of rotating machinery based on t-SNE and LSTM ［J］.Journal of Vibration and Shock, 2020,39(7): 223-231.
［10］HOCHREITER S, SCHMIDHUBER J.Long short-term memory［J］.Neural Computation, 1997, 9(8):1735-1780.
［11］CHO K, VAN MERRIENBOER B, GULCEHRE C, et al.Learning phrase representations using RNN encoder-decoder for statistical machine translation［J］.Computer Science, 2014(9): 1-15.
［12］李锋, 向往, 王家序, 等.基于量子加权门限重复单元神经网络的性态退化趋势预测［J］.振动与冲击,2019,38(1):123-129.
LI Feng, XIANG Wang, WANG Jiaxu, et al.Prediction of behavior degradation trend based on quantum weighted threshold repetitive unit neural network ［J］.Journal of Vibration and Shock, 2019,38(1): 123-129.
［13］杨平, 苏燕辰.基于卷积门控循环网络的滚动轴承故障诊断［J］.航空动力学报,2019,34(11):2432-2439.
YANG Ping, SU Yanchen.Fault diagnosis of rolling bearing based on convolutional gated circulation network ［J］.Journal of Aeronautical Power, 2019,34(11): 2432-2439.
［14］吴静然, 丁恩杰, 崔冉, 等.采用多尺度注意力机制的旋转机械故障诊断方法［J］.西安交通大学学报,2020,54(2):51-58.
WU Jingran, DING Enjie, CUI Ran, et al.Fault diagnosis method of rotating machinery using multi-scale attention mechanism ［J］.Journal of Xi’an Jiaotong University, 2020,54(2): 51-58.
［15］雷亚国, 韩天宇, 王彪, 等.XJTU-SY滚动轴承加速寿命试验数据集解读［J］.机械工程学报,2019,55(16):1-6.
LEI Yaguo, HAN Tianyu, WANG Biao, et al.Interpretation of XJYU-SY rolling bearing accelerated life test data set ［J］.Journal of Mechanical Engineering, 2019,55(16): 1-6.
［16］文娟, 高宏力.一种基于UPF的轴承剩余寿命预测方法［J］.振动与冲击,2018,37(24):208-213.
WEN Juan, GAO Hongli.A prediction method of bearing residual life based on UPF ［J］.
Journal of Vibration and Shock, 2018,37(24): 208-213.
［17］WANG W J, LU Y M.Analysis of the mean absolute error (MAE) and the root mean square error (RMSE) in assessing rounding model［J］.IOP Conference Series: Materials Science and Engineering, 2018, 324:012049.
［18］QIU H, LEE J, LIN J, et al.Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics［J］.Journal of Sound & Vibration, 2006, 289(4/5):1066-1090.
