WANG Hengdi,CHEN Peng,WANG Haokui,et al. Life prediction method of rolling bearings based on DRSN-ADA[J]. Journal of Mechanical Transmission,2026,50(1):184-191.
WANG Hengdi,CHEN Peng,WANG Haokui,et al. Life prediction method of rolling bearings based on DRSN-ADA[J]. Journal of Mechanical Transmission,2026,50(1):184-191. DOI: 10.16578/j.issn.1004.2539.2026.01.022.
Life prediction method of rolling bearings based on DRSN-ADA
首先,构建了深度残差收缩网络和对抗式领域自适应健康状态评估模型,并利用DRSN可以规避振动信号中的噪声并自适应提取轴承退化特征的性能,构建了健康指标曲线;其次,利用ADA使测试集健康指标和训练集健康指标分布对齐;最后,将DRSN-ADA模型输出的健康指标输入到卷积长短时记忆(Convolutional Long Short-Term Memory
A health state assessment method combining deep residual shrinkage network (DRSN) and adversarial domain adaptation (ADA) was proposed to address the problems of vibration signal noise interference and inconsistent data distribution under different working conditions in the remaining useful life (RUL) prediction of rolling bearings
so as to improve the accuracy and generalization ability of RUL prediction.
Methods
2
Firstly
a health state assessment model combining deep residual shrinkage network and adversarial domain adaptation was constructed. The performance of DRSN in avoiding noise in vibration signals and adaptively extracting bearing degradation features was utilized to build the health indicator curve. Then
ADA was used to align the distribution of health indicators between the test set and the training set
so as to eliminate the difference in data distribution under different working conditions. Finally
the health indicators output by the DRSN-ADA model were input into the convolutional long short-term memory (ConvLSTM) network model
and the accurate RUL prediction of rolling bearings was realized.
Results
2
In the XJTU-SY dataset and engineering tests
the health indicators constructed by DRSN-ADA are superior to the comparison methods in monotonicity
robustness and correlation
with their mean values reaching 0.61
0.97 and 0.98 respectively. The mean values of mean squared error (MSE) and mean absolute error (MAE) of the RUL prediction results are 2.52% and 2.19% respectively
and the average score is 0.86
which is significantly better than the DRN
principal component analysis and root mean square (RMS) methods. These results verify the effectiveness of the proposed method in noise suppression and cross-working condition prediction.
ZHANG Zhengxin , HU Changhua , SI Xiaosheng , et al . Degradation modeling and remaining useful life prediction with bivariate time scale [J]. Acta Automatica Sinica , 2017 , 43 ( 10 ): 1789 - 1798 .
LI Gang , YU Changhai , LIU Yunpeng , et al . Challenges and prospects of fault prognostic and health management for power transformer [J]. Automation of Electric Power Systems , 2017 , 41 ( 23 ): 156 - 167 .
LI W , ZHU Z C , JIANG F , et al . Fault diagnosis of rotating machinery with a novel statistical feature extraction and evaluation method [J]. Mechanical Systems and Signal Processing , 2015 , 50 : 414 - 426 .
WEN Jinghui , WU Rongsen , LI Shuaiyong , et al . Bearing residual life prediction method based on DRSN and optimized BiLSTM [J]. Computer Integrated Manufacturing Systems , 2024 , 30 ( 5 ): 1877 - 1888 .
GEBRAEEL N , LAWLEY M , LIU R , et al . Residual life predictions from vibration-based degradation signals:a neural network approach [J]. IEEE Transactions on Industrial Electronics , 2004 , 51 ( 3 ): 694 - 700 .
MENG Wenjun , ZHANG Sicong , DAN Ziyan , et al . Method of dynamic life prediction of rolling bearing [J]. Journal of Vibration ,Measurement & Diagnosis, 2019 , 39 ( 3 ): 652 - 658 .
HUANG Darong , CHEN Changsha , SUN Guoxi , et al . Linear discriminant analysis and back propagation neural network cooperative diagnosis method for multiple faults of complex equipment bearings [J]. Acta Armamentarii , 2017 , 38 ( 8 ): 1649 - 1657 .
LIAO L X , JIN W J , PAVEL R . Enhanced restricted Boltzmann machine with prognosability regularization for prognostics and health assessment [J]. IEEE Transactions on Industrial Electronics , 2016 , 63 ( 11 ): 7076 - 7083 .
GUO L , LEI Y G , LI N P , et al . Machinery health indicator construction based on convolutional neural networks considering trend burr [J]. Neurocomputing , 2018 , 292 : 142 - 150 .
WU J , HU K , CHENG Y W , et al . Data-driven remaining useful life prediction via multiple sensor signals and deep long short-term memory neural network [J]. ISA Transactions , 2020 , 97 : 241 - 250 .
CHEN Weixing , CHANG Dongrun , LI Zongshuai . Aeroengine residual life prediction method based on improved generative adversarial network and ConvLSTM [J]. Journal of Electronic Measurement and Instrumentation , 2023 , 37 ( 3 ): 211 - 221 .
ZHANG A S , WANG H L , LI S B , et al . Transfer learning with deep recurrent neural networks for remaining useful life estimation [J]. Applied Sciences , 2018 , 8 ( 12 ): 2416 .
SUN Tong . Research on health status assessment and remaining life prediction method based on BILSTM-ADA [D]. Xi’an : Xidian University , 2022 : 15 - 17 .
XIAO Zongzhen , DU Haofei , WANG Yong , et al . Prediction of fan bearing life based on ConvLSTM [J]. Modular Machine Tool & Automatic Manufacturing Technique , 2024 ( 6 ): 161 - 165 .
LEI Yaguo , HAN Tianyu , WANG Biao , et al . XJTU-SY rolling element bearing accelerated life test datasets:a tutorial [J]. Journal of Mechanical Engineering , 2019 , 55 ( 16 ): 1 - 6 .
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 : 1 - 8 .