A Review of Industrial Fault Diagnosis Technologies Based on Machine Learning
Main Article Content
Keywords
machine learning, industrial fault diagnosis, feature extraction, deep learning, small-sample learning, intelligent diagnosis
Abstract
The stable operation of industrial equipment is a fundamental guarantee for intelligent manufacturing, and fault diagnosis technologies have evolved from experience-driven and model-driven approaches to data-driven paradigms based on machine learning. In response to challenges such as nonlinearity, strong coupling, and the scarcity of fault samples in industrial systems, machine learning has become a core approach for fault diagnosis due to its advantages in feature learning and pattern recognition. This paper systematically reviews the research progress of machine learning in industrial fault diagnosis. From the perspective of technical frameworks, it elaborates on the application principles and engineering performance of traditional machine learning, deep learning, and improved algorithms. Furthermore, it analyzes algorithm adaptability in typical industrial scenarios, identifies key challenges such as small-sample diagnosis and model interpretability, and finally discusses future development trends from the perspectives of algorithm integration and data augmentation. This study aims to provide references for technological optimization and engineering applications in this field.
References
- [1] Lei, Y. G., Jia, F., et al. (2018). Opportunities and challenges of intelligent fault diagnosis for machinery in the era of big data. Journal of Mechanical Engineering, 54(1), 94–104.
- [2] Guo, Y. F., & Tang, J. Y. (2019). A review of life prediction and fault diagnosis technologies based on machine learning algorithms. Computer Measurement & Control, 27(3), 1–5, 10.
- [3] Zhao, H. S., Zhang, J. P., & Li, G. (2012). Fault early warning method for wind turbine gearbox based on cluster analysis. Automation of Electric Power Systems, 36(15), 102–106.
- [4] Song, Y. Q., Zhu, Z. J., & Ji, Y. F. (2014). Intelligent fault diagnosis based on multi-sensor information fusion. Journal of Xi’an Polytechnic University, (5), 568–573.
- [5] Bian, W. B., Deng, A. D., Liu, D. C., et al. (2023). Fault diagnosis method for rolling bearings of wind turbines based on improved deep residual shrinkage network. Journal of Mechanical Engineering, 59(12), 160–171.
- [6] Xiao, Y., Xiang, J. W., Tang, H. S., et al. (2024). Fault diagnosis method for hydraulic waterproof valves based on multi-sensor information fusion and CNN-BiGRU-Attention model. Mechanical & Electrical Engineering, 41(9), 1531–1540.
- [7] Xu, K., Chen, Z. H., Zhang, C. B., et al. (2019). Rolling bearing fault diagnosis based on empirical mode decomposition and support vector machine. Control Theory & Applications, 36(6), 915–922.
- [8] Ruan, W. Y., Ma, Z. Q., & Li, Y. C. (2020). Rolling bearing fault diagnosis based on variational mode decomposition, permutation entropy, and PSO-SVM. Journal of University of Jinan (Natural Science Edition), 34(2), 131–137.
- [9] Zhang, Y. F., Yao, Z. Y., Tang, S. L., et al. (2016). Engine fault diagnosis method based on principal component analysis and support vector machine. China Mechanical Engineering, (24), 3307–3311.
- [10] Wen, J. T., & Zhou, X. N. (2018). Fault diagnosis of rotating machinery based on fuzzy granular unsupervised learning combined with random forest. Mechanical Science and Technology for Aerospace Engineering, 37(11), 1722–1730.
- [11] Wu, C. Z., Jiang, P. C., Feng, F. Z., et al. (2018). Gearbox fault diagnosis based on one-dimensional convolutional neural network. Journal of Vibration and Shock, (22), 51–56.
- [12] Yu, F., Fan, Q. C., & Xuan, M. (2022). Motor bearing fault prediction based on grasshopper optimization Bi-LSTM network. Electric Machines and Control, 26(6), 9–17.
- [13] Tao, D. Q., Bo, C. M., & Yi, H. (2015). Transformer fault detection method based on random forest. Electronic Components and Materials, 38(4), 840–844.
- [14] Wang, T. T., Wang, Y., & Ji, Z. C. (2018). Rolling bearing fault diagnosis based on improved extreme learning machine. Journal of System Simulation, 30(11), 4413–4420.
- [15] Gao, D. X., Wang, Y., Zheng, X. Y., et al. (2023). Charging state monitoring and safety early warning method for electric vehicles based on deep learning. Electric Machines and Control, 27(7), 122–132.
- [16] Li, Y. B., Yang, Y. T., Wang, X. Z., et al. (2018). Early fault diagnosis of rolling bearings based on hierarchical symbol dynamic entropy and binary tree support vector machine. Journal of Sound and Vibration, 428, 72–86.
- [17] Fu, Q., Jing, B., He, P. J., et al. (2018). Fault feature selection and diagnosis of rolling bearings based on EEMD and optimized Elman-AdaBoost algorithm. IEEE Sensors Journal, 18(12), 5024–5034.
- [18] Zhang, Z. J., Li, H., Huang, Y. S., et al. (2024). Application of deep residual shrinkage network in rolling bearing fault diagnosis. Journal of Northeastern University (Natural Science), 45(11), 1587–1594.
- [19] Chen, T., & Lei, Y. G. (2021). Cross-equipment fault diagnosis based on transfer component analysis. Journal of Mechanical Engineering, 57(12), 112–120.
- [20] Zhou, Q., Chen, W. G., et al. (2014). Transformer multiple fault diagnosis and short-term prediction method based on cloud theory. High Voltage Engineering, 40(5), 1450–1456.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.