Cross-domain Fault Diagnosis of Bearings Based on Joint Subdomain Contrast Alignment

doi:10.3969/j.issn.1004-132X.2025.05.018

Abstract

Abstract: The fault data of bearings exhibited significant distribution discrepancies under varying operating conditions， relatively low diagnostic accuracy was resulted in practical fault detection models. Additionally， most existing research on cross-domain bearing fault diagnosis primarily emphasized inter-domain alignment and intra-class comparison， while neglecting the influences of interactions between subdomains. Therefore， a cross-domain fault diagnosis method of bearings was proposed based on joint subdomain contrast alignment. In order to highlight the fault features， the bearing vibration signals were transformed into time-frequency graph by short-time Fourier transform， and the fault features were obtained by inputting them into the feature extraction module. Domain adaptation methods achieved cross-domain recognition by transferring knowledge learned from the source domain to the target domain. During the domain adaptation processes， a joint subdomain contrast alignment strategy was used to bring samples from the same subdomain closer together while separating samples from different subdomains， which aligned the subdomain distributions of the same class samples among the source and target domains， thereby enhancing the models generalization ability in the target domain. Resnet34 was used as the feature extraction network on the model architecture， and the maximum mean difference was used at the output of the network to align the global distribution of the source domain and the target domain. Compared with the classical domain adaptation methods， the experimental results on the bearing fault data set of Case Western Reserve University shows that the cross-domain fault diagnosis method of bearings based on joint subdomain contrast alignment has better feature transfer ability.

Key words: , fault diagnosis, rolling bearing, transfer learning, contrast alignment, subdomain adaptation

摘要： 变工况下轴承故障数据分布差异大，故障诊断模型实际识别精度较低，且目前大多数轴承跨域故障诊断研究侧重于域间对齐和类内对比而忽略了子域间的影响，为此，提出了一种基于联合子域对比对齐的轴承跨域故障诊断方法。为突出故障特征，对轴承振动信号进行短时傅里叶变换将其转化为时频图，输入特征提取模块，得到故障特征，并通过领域自适应方法将从源域学到的知识迁移到目标域，实现跨域识别。在域适应过程中采用联合子域对比对齐策略，将来自相同子域样本拉近的同时分开不同子域样本，对齐源域与目标域中同类别样本所属子域分布，从而提高模型在目标域上的泛化能力。在模型架构上使用Resnet34作为特征提取网络，在网络输出端使用最大均值差异，对齐源域与目标域的全局分布。基于凯斯西储大学轴承故障数据集进行实验验证，与经典域适应方法进行对比，实验结果表明基于联合子域对比对齐的轴承跨域故障诊断方法具有更好的特征迁移能力。

关键词: 故障诊断, 滚动轴承, 迁移学习, 对比对齐, 子域自适应

CLC Number:

TH17

YANG Kang1, CHEN Xuejun1, 2, ZHANG Lei3, LIU Feng3. Cross-domain Fault Diagnosis of Bearings Based on Joint Subdomain Contrast Alignment[J]. China Mechanical Engineering, 2025, 36(05): 1065-1073.

杨康1, 陈学军1, 2, 张磊3, 刘烽3. 基于联合子域对比对齐的轴承跨域故障诊断[J]. 中国机械工程, 2025, 36(05): 1065-1073.

References

［1］SHAO S Y， MCAIEER S， YAN R Q， et al. Highly Accurate Machine Fault Diagnosis Using Deep Transfer Learning［J］. IEEE Transactions on Industrial Informatics， 2019， 15（4）：2446-2455.
［2］安文杰，陈长征，田淼，等. 基于迁移学习的风电机组轴承故障诊断研究［J］. 太阳能学报， 2023， 44（6）：367-373.
AN Wenjie， CHEN Changzheng， TIAN Miao， et al. Resarch on Bearing Fault Diagnosis of Wind Turbines Based on Transfer Learning［J］. Acta Energiae Solaris Sinica， 2023， 44（6）：367-373.
［3］雷亚国，贾峰，孔德同，等. 大数据下机械智能故障诊断的机遇与挑战［J］. 机械工程学报， 2018， 54（5）：94-104.
LEI Yaguo， JIA Feng， KONG Detong， et al. Opportunities and Challenges of Machinery Intelligent Fault Diagnosis in Big Data Era［J］. Journal of Mechanical Engineering， 2018， 54（5）：94-104.
［4］JIA F， LEI Y G， LU N， et al. Deep Normalized Convolutional Neural Network for Imbalanced Fault Classification of Machinery and Its Understanding via Visualization［J］. Mechanical Systems and Signal Processing， 2018， 110：349-367.
［5］KHAN S， YAIRI T. A Review on the Application of Deep Learning in System Health Management［J］. Mechanical Systems and Signal Processing， 2018， 107：241-265.
［6］邢自扬，赵荣珍，吴耀春，等. 多尺度卷积神经网络小样本轴承故障辨识方法［J］. 振动·测试与诊断， 2023， 43（5）：915-922.
XING Ziyang， ZHAO Rongzhen， WU Yaochun， et al. Small Sample Bearing Fault Identification Method Using Novel Multi-scale Convolutional Neural Network［J］. Journal of Vibration， Measurement & Diagnosis， 2023， 43（5）：915-922.
［7］叶楠，常佩泽，张露予，等. 基于改进后半监督深度信念网络的多工况轴承故障诊断研究［J］. 机械工程学报， 2021， 57（15）：80-90.
YE Nan， CHANG Peize， ZHANG Luyu， et al. Research on Multi-condition Bearing Fault Diagnosis Based on Improved Semi-supervised Deep Belief Network［J］. Journal of Mechanical Engineering， 2021， 57（15）：80-90.
［8］CHEN M Z， SHAO H D， DOU H X， et al. Data Augmentation and Intelligent Fault Diagnosis of Planetary Gearbox Using ILoFGAN under Extremely Limited Samples［J］. IEEE Transactions on Reliability， 2023， 72（3）：1029-1037.
［9］ZHAO P， WU T， ZHAO S Y， et al. Robust Transfer Learning Based on Geometric Mean Metric Learning［J］. Knowledge-Based Systems， 2021， 227：107227.
［10］LI H， HUANG C G， SOARES C G. A Real-time Inspection and Opportunistic Maintenance Strategies for Floating Offshore Wind Turbines［J］. Ocean Engineering， 2022， 256：111433.
［11］SU Z Q， JIANG W L， CHEN K， et al. Multi-adversarial Deep Transfer Network for Multi-source Open-set Fault Diagnosis of Rotating Machinery with Category Shift［J］. Knowledge-Based Systems， 2023， 282：111106.
［12］LI C， ZHANG S H， QIN Y， et al. A Systematic Review of Deep Transfer Learning for Machinery Fault Diagnosis［J］. Neurocomputing， 2020， 407：121-135.
［13］HAN T， LIU C， WU R， et al. Deep Transfer Learning with Limited Data for Machinery Fault Diagnosis［J］. Applied Soft Computing， 2021， 103：107150.
［14］HE Z Y， SHAO H D， WANG P， et al. Deep Transfer Multi-wavelet Auto-encoder for Intelligent Fault Diagnosis of Gearbox with Few Target Training Samples［J］. Knowledge-Based Systems， 2020， 191：105313.
［15］BORGWARDT K M， GRETTON A， RASCH M J， et al. Integrating Structured Biological Data by Kernel Maximum Mean Discrepancy［J］. Bioinformatics， 2006， 22（14）：e49-e57.
［16］董绍江，周存芳，陈里里，等. 基于判别性特征提取和双重域对齐的轴承跨域故障诊断［J］. 中国机械工程， 2023， 34（15）：1856-1863.
DONG Shaojiang， ZHOU Cunfang， CHEN Lili， et al. Cross Domain Fault Diagnosis of Bearings Based on Discriminant Feature Extraction and Dual-domain Alignment［J］. China Mechanical Engineering， 2023， 34（15）：1856-1863.
［17］王鹏，李丹青，王恒. 基于改进交替迁移学习的滚动轴承故障诊断算法［J］. 振动与冲击， 2024， 43（5）：239-249.
WANG Peng， LI Danqing， WANG Heng. Rolling Bearing Fault Diagnosis Algorithm Based on Improved Alternating Transfer Learning［J］. Journal of Vibration and Shock， 2024， 43（5）：239-249.
［18］ZHU Y C， ZHUANG F Z， WANG J D， et al. Deep Subdomain Adaptation Network for Image Classification［J］. IEEE Transactions on Neural Networks and Learning Systems， 2021， 32（4）：1713-1722.
［19］陈忠，唐鑫，张大明，等. S变换子域适应的扶梯电机轴承迁移诊断［J］. 华南理工大学学报（自然科学版）， 2023， 51（12）：34-41.
CHEN Zhong， TANG Xin， ZHANG Daming， et al. Stockwell Transform Combined with Subdomain Adaptation for Escalator Motor Bearing Transfer Diagnosis［J］. Journal of South China University of Technology（Natural Science Edition）， 2023， 51（12）：34-41.
［20］GUO Y， ZHANG J D， SUN B， et al. Adversarial Deep Transfer Learning in Fault Diagnosis：Progress， Challenges， and Future Prospects［J］. Sensors， 2023， 23（16）：7263.
［21］CARON M， MISRA I， MAIRAL J， et al. Unsupervised Learning of Visual Features by Contrasting Cluster Assignments［J］. Advances in Neural Information Processing Systems， 2020， 33：9912-9924.

[1]	LUO Ji-Sai, XU De-Ge, SHI Mei-Li. Gearbox Fault Diagnosis with Rotating Speed Fluctuation Based on SVD and Chirplet Path Pursuit Algorithm [J]. J4, 201016, 21(16): 1947-1951.
[2]	HU Chao, LIU Zuo-Min. Research on Rolling Bearing’s Heathcote’s Stick-slip Model Based on Materials Compatibility [J]. J4, 201016, 21(16): 1969-1973.
[3]	FENG Yukang, ZHU Shuo, JIANG Zhigang, YAN Wei, ZHANG Hua. Reliability Dynamic Prediction Method for Remanufactured Products Based on Data-model Integration and Transfer [J]. China Mechanical Engineering, 2026, 37(4): 1007-1015.
[4]	LI Ke, WANG Mengjun, YUAN Maojun, ZHANG Hongshuo, YUAN Keyan, LU Guoliang. Early Fault Detection for Rolling Bearings Based on One-dimensional Structure Graph Entropy [J]. China Mechanical Engineering, 2026, 37(3): 645-655.
[5]	HU Kun, CHEN Zhuo, HAN Xin, JIANG Hao, NIU Jie. Fault Diagnosis Method of Belt Conveyor Roller Bearings Based on NSST4-SVD-DBN [J]. China Mechanical Engineering, 2026, 37(3): 656-667.
[6]	GAO Xuejin, WANG Xuan, JIANG Kexin, GAO Huihui, QI Yongsheng. Fault Diagnosis of Chillers Based on Multi-scale Domain Generative Networks [J]. China Mechanical Engineering, 2026, 37(3): 688-696.
[7]	TONG Zhaojing, WANG Pengchao, FAN Yongkui, HAN Guangyang, WANG Ziqi. Fault Diagnosis Method of Rolling Bearings Based on Improved Refined Composite Multiscale Sample Entropy and Bayesian Network [J]. China Mechanical Engineering, 2025, 36(12): 2952-2959.
[8]	ZHANG Xiong, DONG Lecong, WANG Wenqiang, QU Weiying, WAN Shuting. A Multi-fault Diagnosis Method for Rolling Bearings Integrating Two-dimensional Convolutional and GRU [J]. China Mechanical Engineering, 2025, 36(12): 2978-2985.
[9]	Huiming YANG, Xia YANG, Yuqi HUANG. Full-cycle Fatigue Life Prediction of Rolling Bearings Based on Paris Modeling [J]. China Mechanical Engineering, 2025, 36(09): 2003-2010.
[10]	Siqian FENG, Jiaxu WANG, Xin ZHANG, Xinyue HUANG. Rolling Bearing Fault Diagnosis Using a New Resonance Demodulation Method [J]. China Mechanical Engineering, 2025, 36(09): 2022-2031.
[11]	Jian CHEN, Minghui YAN, Pin CHEN. Acoustic Signal Fault Diagnosis Method of Centrifugal Pumps Based on Bayesian Optimization Multiscale DenseNet [J]. China Mechanical Engineering, 2025, 36(09): 2032-2038.
[12]	ZHENG Chao, LIU Jianchao, WU JunXUE Xin. Vibration Reduction Analysis of All-metal Multi-directional Isolation Devices [J]. China Mechanical Engineering, 2025, 36(07): 1463-1470.
[13]	ZHAO Tingqi1, CHENG Hongyu2, SUN Maokai1, WANG Shenghai1, WANG Haoran1, CHEN Haiquan1. Nonlinear Stability Control of Multi-cable Anti-sway Systems for Marine Cranes [J]. China Mechanical Engineering, 2025, 36(07): 1487-1496.
[14]	ZHANG Jianyu, WANG Liuzhen, XIAO Yong, MA Yanan. Fusion of Degradation Feature Information and Remaining Life Prediction for Rolling Bearings [J]. China Mechanical Engineering, 2025, 36(07): 1553-1561.
[15]	SONG Lijun, LIU Songlin, XIN Yu, MA Jinghua, XIE Zhengqiu. Residual Life Prediction for Bearings Based on Bearing Degradation State Assessment and IGAT-BiGRU Network [J]. China Mechanical Engineering, 2025, 36(07): 1562-1572.