一种基于TC-CAE的轴承寿命预测方法

PDF(2391 KB)

振动与冲击 ›› 2022, Vol. 41 ›› Issue (14) : 105-113.

论文

一种基于TC-CAE的轴承寿命预测方法

李海浪，刘永志，邹益胜，刘彦涛，宋小欣

作者信息 +

Bearing life prediction based on the method of TC-CAE

LI Hailang，LIU Yongzhi，ZOU Yisheng，LIU Yantao，SONG Xiaoxin

Author information +

文章历史 +

摘要

在预测轴承的剩余使用寿命(Remaining useful life, RUL)时,能否有效提取退化特征是实现准确预测的关键之一。轴承的个体异质性和工况差异性导致退化特征曲线不同，同一特征的变化趋势在不同轴承上是具有差异的，从而导致训练轴承建立的RUL预测模型与测试轴承不匹配。在提取特征时应当考虑轴承的个体差异性，减少轴承特征的个体差异性有利于提升预测精度。为了促进同一特征在不同轴承上的趋势一致性，减少退化特征的轴承个体差异性，提出了一种基于趋势一致性约束卷积编码(Trend consistency convolutional auto-encoder, TC-CAE)的轴承寿命预测方法。通过构造趋势一致性约束，并和卷积自编码相结合，形成了TC-CAE特征提取模型。预测流程为先用TC-CAE模型在频域信号内提取特征，再用长短期记忆网络(Long Short-Term Memory, LSTM)预测。在一个轴承公开数据集上进行实验，实验结果表明：相比于普通卷积自编码方法的预测结果，该方法的综合平均误差降低了21.1%，相比于特征评价方法和卷积神经网络方法分别降低了35.6%和25.9%。
关键词：轴承；寿命预测；特征提取；趋势一致性

Abstract

When predicting the remaining useful life (RUL) of a bearing, the ability to effectively extract the degradation features is one of the keys to achieve accurate prediction. The individual heterogeneity of the bearing and the difference in operating conditions lead to different degradation characteristic curves, and the change trend of the same characteristic is different on different bearings, which leads to the mismatch between the RUL prediction model established by the training bearing and the test bearing. The individual variability of the bearing should be considered when extracting features. Reducing the individual variability of bearing features is beneficial to improve the prediction accuracy. In order to promote the trend consistency of the same feature on different bearings and reduce the individual bearing differences of degraded features, a bearing life prediction based on trend consistency convolutional auto-encoder (TC-CAE) is proposed. method. By constructing trend consistency constraints and combining with convolutional autoencoding, a TC-CAE feature extraction model is formed. The prediction process is to first use the TC-CAE model to extract features in the frequency domain signal, and then use Long Short-Term Memory (LSTM) to predict. Experiments on a bearing public data set, the experimental results show that: Compared with the prediction results of the ordinary convolutional autoencoding method, the comprehensive average error of the method is reduced by 21.1%, compared with the feature evaluation method and the convolutional neural network method. respectively reduced by 35.6% and 25.9%.
Key words: bearing; life prediction; feature extraction; trend consistency

导出引用

李海浪，刘永志，邹益胜，刘彦涛，宋小欣. 一种基于TC-CAE的轴承寿命预测方法[J]. 振动与冲击, 2022, 41(14): 105-113

LI Hailang，LIU Yongzhi，ZOU Yisheng，LIU Yantao，SONG Xiaoxin. Bearing life prediction based on the method of TC-CAE[J]. Journal of Vibration and Shock, 2022, 41(14): 105-113

参考文献

[1] MA, Meng; MAO, Zhu. Deep-Convolution-Based LSTM Network for Remaining Useful Life Prediction[J]. IEEE Transactions on Industrial Informatics, 2020, 17.3: 1658-1667.
[2] Yaguo Lei,Naipeng Li,Liang Guo,Ningbo Li,Tao Yan,Jing Lin. Machinery health prognostics: A systematic review from data acquisition to RUL prediction[J]. Mechanical Systems and Signal Processing,2018,104.
[3] Wang B , Lei Y , Li N , et al. A Hybrid Prognostics Approach for Estimating Remaining Useful Life of Rolling Element Bearings[J]. IEEE Transactions on Reliability, 2018:1-12.
[4] 王久健,杨绍普,刘永强,文桂林.一种基于空间卷积长短时记忆神经网络的轴承剩余寿命预测方法[J/OL].机械工程学报:1-8[2021-05-10].http://kns.cnki.net/kcms/detail/11.2187.TH.20210420.1502.164.html.
Wang Jiujian, Yang Shaopu, Liu Yongqiang, Wen Guilin. A bearing remaining life prediction method based on spatial convolution long and short-term memory neural network [J/OL]. Chinese Journal of Mechanical Engineering: 1-8 [2021-05-10]. http://kns.cnki.net/kcms/detail/11.2187.TH.20210420.1502.164.html.
[5] Zhu J , Nan C , Peng W . Estimation of Bearing Remaining Useful Life based on Multiscale Convolutional Neural Network[J]. IEEE Transactions on Industrial Electronics, 2018, PP:1-1.
[6] Meng M , Zhu M . Deep Convolution-based LSTM Network for Remaining Useful Life Prediction[J]. IEEE Transactions on Industrial Informatics, 2020, PP(99):1-1.
[7] Hu C H , Pei H , Si X S , et al. A Prognostic Model Based on DBN and Diffusion Process for Degrading Bearing[J]. IEEE Transactions on Industrial Electronics, 2019, PP(99):1-1.
[8] 陈仁祥,黄鑫,杨黎霞,汤宝平,陈思杨,杨星.加噪样本扩展深度稀疏自编码神经网络的滚动轴承寿命阶段识别[J].振动工程学报,2017,30(05):874-882.
CHEN Renxiang，HUANG Xin，YANG Lixia，et al. Bearing life state recognition using deep sparse auto-encoder neural network with noise adding sample expansion[J]. Journal of Vibration Engineering,2017,30(5):874-882.
[9] 李华新,王衍学.基于分层稀疏编码的轴承剩余寿命预测方法[J].现代制造工程,2019(05):7-12.
LI Huaxin,WANG Yanxue. A RUL prediction approach for rolling bearing based on hierarchical sparse coding[J]. Modern Manufacturing Engineering, 2019(5):7-12.
[10] 张继冬,邹益胜,蒋雨良,曾大毅.基于全卷积变分自编码网络FCVAE的轴承剩余寿命预测方法[J].振动与冲击,2020,39(19):13-18+25.
ZHANG Jidong,ZOU Yisheng,JIANG Yuliang,et al. Prediction method for bearing residual life based on a FCVAE network[J].
Journal of Vibration and Shock, 2020,39(19):13-18.
[11] Kaji M , Parvizian J , Venn H . Constructing a Reliable Health Indicator for Bearings Using Convolutional Autoencoder and Continuous Wavelet Transform. Applied Sciences, 2020.
[12] 王奉涛,陈旭涛,柳晨曦,李宏坤,韩清凯,朱泓.基于KPCA和WPHM的滚动轴承可靠性评估与寿命预测[J].振动.测试与诊断,2017,37(03):476-483+626.
Wang Fengtao, Chen Xutao, Liu Chenxi, Li Hongkun, Han Qingkai, Zhu Hong. Reliability Evaluation and Life Prediction of Rolling Bearing Based on KPCA and WPHM [J]. Journal of Vibration and Vibration,2017,37(03):476-483+626.
[13] 申中杰,陈雪峰,何正嘉,孙闯,张小丽,刘治汶.基于相对特征和多变量支持向量机的滚动轴承剩余寿命预测[J].机械工程学报,2013,49(02):183-189.
Shen Zhongjie, Chen Xuefeng, He Zhengjia, Sun Chuang, Zhang Xiaoli, Liu Zhiwen. Remaining Life Prediction of Rolling Bearings Based on Relative Feature and Multivariable Support Vector Machine [J]. Journa al of Mechanical Engineering,2013,49(02):183-189.
[14] 康守强,叶立强,王玉静,谢金宝,Mikulovich V I.基于MCEA-KPCA和组合SVR的滚动轴承剩余使用寿命预测[J].电子测量与仪器学报,2017,31(09):1365-1371.
Kang Shouqiang, Ye Liqiang, Wang Yujing, Xie Jinbao, Mikulovich VI. Remaining service life prediction of rolling bearing based on MCEA-KPCA and combined SVR[J]. Journal of Electronic Measurement and Instrument, 2017, 31(09): 1365-1371
[15] 刘弹,李晓婉,梁霖,吴杰,徐光华,沈强.采用时间序列突变点检测的滚动轴承性能退化评价方法[J].西安交通大学学报,2019,53(12):10-16.
Liu Dan, Li Xiaowan, Liang Lin, Wu Jie, Xu Guanghua, Shen Qiang. Rolling bearing performance degradation evaluation method using time series mutation point detection[J]. Journal of Xi'an Jiaotong University, 2019, 53(12): 10-16 .
[16] Zhao H , Liu H , Jin Y , et al. Feature Extraction for Data-driven Remaining Useful Life Prediction of Rolling Bearings[J]. IEEE Transactions on Instrumentation and Measurement, 2021, PP(99):1-1.
[17] Ding N ， Li H ， Yin Z ， et al. Journal bearing seizure degradation assessment and remaining useful life prediction based on long short-term memory neural network[J]. Measurement， 2020， 166:108215.
[18] Nectoux P ， Gouriveau R ， Medjaher K ， et al. PRONOSTIA: An experimental platform for bearings accelerated degradation tests[C]// IEEE International Conference on Prognostics and Health Management. IEEE， 2012.
[19] 王奉涛,刘晓飞,邓刚,李宏坤,于晓光.基于长短期记忆网络的滚动轴承寿命预测方法[J].振动.测试与诊断,2020,40(02):303-309+419.
Wang Fengtao, Liu Xiaofei, Deng Gang, Li Hongkun, Yu Xiaoguang. Long short-term memory network based rolling bearing life prediction method[J]. Journal of Vibration，Measurement & Diagnosis, 2020, 40(02): 303-309+419.
[20] Cheng C ， Ma G ， Zhang Y ， et al. A Deep Learning-Based Remaining Useful Life Prediction Approach for Bearings[J]. IEEE/ASME Transactions on Mechatronics， 2020， 25(3):1243-1254.