A Lightweight and Small Sample Bearing Fault Diagnosis Algorithm Based on Probabilistic Decoupling Knowledge Distillation and Meta-Learning

被引:0
作者
Luo, Hao [1 ]
Ren, Tongli [1 ]
Zhang, Ying [1 ]
Zhang, Li [1 ]
机构
[1] Liaoning Univ, Coll Informat, Shenyang 110036, Peoples R China
基金
中国国家自然科学基金;
关键词
fault diagnosis; small sample; meta-learning; knowledge distillation; lightweight;
D O I
10.3390/s24248157
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
Rolling bearings play a crucial role in industrial equipment, and their failure is highly likely to cause a series of serious consequences. Traditional deep learning-based bearing fault diagnosis algorithms rely on large amounts of training data; training and inference processes consume significant computational resources. Thus, developing a lightweight and suitable fault diagnosis algorithm for small samples is particularly crucial. In this paper, we propose a bearing fault diagnosis algorithm based on probabilistic decoupling knowledge distillation and meta-learning (MIX-MPDKD). This algorithm is lightweight and deployable, performing well in small sample scenarios and effectively solving the deployment problem of large networks in resource-constrained environments. Firstly, our model utilizes the Model-Agnostic Meta-Learning algorithm to initialize the parameters of the teacher model and conduct efficient training. Subsequently, by employing the proposed probability-based decoupled knowledge distillation approach, the outstanding performance of the teacher model was imparted to the student model, enabling the student model to converge rapidly in the context of a small sample size. Finally, the Paderborn University dataset was used for meta-training, while the bearing dataset from Case Western Reserve University, along with our laboratory dataset, was used to validate the results. The experimental results demonstrate that the algorithm achieved satisfactory accuracy performance.
引用
收藏
页数:25
相关论文
共 44 条
  • [1] Kim J.H., Kim B.S., Im D., Park J.H., Moon D.J., Park Y.J., Optimization of high-speed angular contact ball bearing for aircraft gearbox utilizing an evolutionary multi-objective algorithm, NSGA-III, Adv. Mech. Eng, 16, pp. 1-14, (2024)
  • [2] Zhong H., He D.Q., Lao Z.P., Jin Z.Z., Shen G.Q., Chen Y.J., Improved metric-based meta learning with attention mechanism for few-shot cross-domain train bearing fault diagnosis, Meas. Sci. Technol, 35, (2024)
  • [3] Liu Z.Y., Wu X.C., Cai J., Yang Y., Liu C.Z., Co-excitation control of a four-degree-of-freedom magnetic bearing switched reluctance motor based on mathematical models, IET Electr. Power Appl, pp. 1-16, (2024)
  • [4] Li H.F., Zhao X.Y., Yang S.P., Wei J.L., Gu X.H., Liu Y.Q., Liu P.F., Duan S.Y., Wang X.D., Fatigue failure mechanism of high-speed train bearing steel after long-term service, Eng. Fail. Anal, 165, (2024)
  • [5] Bonnett A.H., Yung C., Increased Efficiency Versus Increased Reliability, IEEE Ind. Appl. Mag, 14, pp. 29-36, (2008)
  • [6] Gong T., Wang Z.Q., Ma Q., Yang J.H., Motor Bearing Fault Diagnosis in an Industrial Robot Under Complex Variable Speed Conditions, J. Comput. Nonlinear Dyn, 19, pp. 1-42, (2024)
  • [7] Han D.F., Qi H.Y., Wang S.X., Hou D.M., Wang C.P., Adaptive stepsize forward-backward pursuit and acoustic emission-based health state assessment of high-speed train bearings, Struct. Health Monit. Int. J, (2024)
  • [8] Kim J., Kim J.M., Bearing Fault Diagnosis Using Grad-CAM and Acoustic Emission Signals, Appl. Sci, 10, (2020)
  • [9] Hou D.M., Qi H.Y., Wang C.P., Han D.F., High-speed train wheel set bearing fault diagnosis and prognostics: Fingerprint feature recognition method based on acoustic emission, Mech. Syst. Signal Process, 171, (2022)
  • [10] Wang B., Qiu W.T., Hu X., Wang W., A rolling bearing fault diagnosis technique based on recurrence quantification analysis and Bayesian optimization SVM, Appl. Soft Comput, 156, (2024)