Author(s)

Zhao Jiang, Yu Wei^*

Affiliation(s)

School of International Business, Zhejiang Yuexiu University, Shaoxing 312000, Zhejiang, China.

Corresponding Author

Yu Wei

ABSTRACT

Under the condition of data category imbalance, this paper proposes a TrAdaBoost-Least Squares Support Vector Machine (TrAdaBoost-LSSVM) algorithm based on instance transfer to solve the problem of low diagnostic accuracy of traditional machine learning methods. Firstly, use the K-means algorithm to screen the source domain data in order to eliminate those data with low similarity to the target domain, and then increase the inter-domain similarity; then, optimize the evaluation index of the base classifier to improve the model generalization ability. The simulation test results show that the method proposed in this paper exhibits the advantage of high fault recognition accuracy compared with the traditional machine learning method.

KEYWORDS

Data category imbalance; Instance transfer; Machine learning; Least squares support vector machine

CITE THIS PAPER

Zhao Jiang, Yu Wei. Early bearing fault detection and recognition method based on instance transfer. Multidisciplinary Journal of Engineering and Technology. 2025, 2(1): 1-8. DOI: https://doi.org/10.61784/mjet3019.

REFERENCES

[1] WANG Taiyong, GONG Liming, WANG Peng, et al. Fault diagnosis model of rotating machinery based on KD -DenseNet. Vibration and Shock, 2020, 39(16): 39-45.

[2] Xingxing Zhang, Shaobo Li, Longxuan Tsu, et al. Research on rolling bearing fault diagnosis based on machine learning algorithm. Combined machine tools and automated machining technology, 2020, 10(7): 36 -39.

[3] CHEN Renxiang, HUANG Xin, YANG Lixia, et al. Rolling bearing fault diagnosis based on convolutional neural network and discrete wavelet transform. Journal of Vibration Engineering, 2018, 31(5): 883-891.

[4] Mao WT, Tian SY, Dou Z, et al. A deep transfer learning-based online detection method for early faults in rolling bearings. Journal of Automation, 2022, 48(1): 302-314.

[5] Liu JC, Quan H, Yu X, et al. Rolling bearing fault diagnosis based on parameter optimization VMD and sample entropy. Journal of Automation, 2022, 48(3): 808-819.

[6] Shakya P, Kulkarni M S, Darpe A K. A novel methodology for online detection of bearing health status for naturally progressing defect. Journal of Sound & Vibration, 2014, 333(21): 5614-5629.

[7] Taylor M E, Stone P H. Transfer learning for reinforcement learning domains: a survey. The Journal of Machine Learning Research, 2009, 10(10):1633-1685.

[8] Yang S, Liu Z, Lu G. Early change detection in dynamical bearing degradation process based on hierarchical graph model and adaptive inputs weighting fusion. IEEE Transactions on Industrial Informatics, 2021, 17(5): 3186-3196.

[9] XU Le, YU Ruxin, XING Bangsheng, et al. Rolling bearing fault feature extraction based on LMD arrangement entropy. Mechanical Transmission, 2019, 43(1): 136-139.

[10] Chawla N V, Bowyer K W, Hall L O, et al. SMOTE: Synthetic minority over-sampling technique . Journal of Artificial Intelligence Research, 2011, 16(1): 321-357.

[11] Yen S J, Lee Y S. Cluster-based Under-sampling Approaches for Imbalanced Data Distributions. Expert Systems with Applications, 2006, 36(3): 5718-5727.

[12] Wang W, Wang C, Wang Z, et al. Abnormal detection technology of industrial control system based on transfer learning. Applied Mathematics and Computation, 2022, 412(1): 126-139.

[13] Freund Y. A decision-theoretic generalization of on-line learning and an application to boosting // Proceedings of the Second European Conference on Computational Learning Theory. Springer-Verlag, 1995.

[14] Qian Z, Hai G L, Yong Z, et al. Instance transfer learning with multisource dynamic trAdaBoost. The Scientific World Journal, 2014, 20(14): 1-8.

[15] Wang C, Yang A, Yuan A, et al. Diagnosis of wind turbine faults with transfer learning algorithms. Renewable Energy, 2021, 16(3): 2053-2067.

[16] Xia R, Zong C, Hu X, et al. Feature ensemble plus sample selection: domain adaptation for Sentiment Classification . IEEE Intelligent Systems, 2013, 28(3): 10-18.

[17] Wilson C, Check J H, Amui J, et al. Effect of multiple sources vs. single source of donor embryos on pregnancy and Implantation Rates per Transfer . Clinical and experimental obstetrics & gynecology, 2011, 38(4):324-325.

[18] Hao S. Kernel methods for transfer learning to avoid negative transfer . International Journal of Computing Science and Mathematics, 2016, 7(2):175-190.

[19] Wang T, Liu Z, Lu G, et al. Temporal-spatio graph based spectrum analysis for bearing fault detection and diagnosis. IEEE Transactions on Industrial Electronics, 2021, 68(3): 2598 -2607.

[20] Li Y, Xu M, Wang R, et al. A fault diagnosis scheme for rolling bearing based on local mean decomposition and improved multiscale fuzzy entropy.Journal of Sound and Vibration, 2016, 3(6): 277-299.

[21] Lei Y, He Z, Zi Y. EEMD method and WNN for fault diagnosis of locomotive roller bearings. Expert Systems with Applications, 2011, 38(6): 7334-7341.

[22] Fu C, Wu Z, Xue M, et al. Cross-domain decision making based on TrAdaBoost for diagnosis of breast lesions. Artificial Intelligence Review, 2023, 56(5): 3987-4017.

[23] Zheng L, Liu G, Yan C, et al. Improved TrAdaBoost and its application to transaction fraud detection. IEEE Transactions on Computational Social Systems, 2020, 7(5): 1304-1316.

[24] He H, Khoshelham K, Fraser C. A multiclass TrAdaBoost transfer learning algorithm for the classification of mobile lidar data. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 16(6): 118-127.

[25] Yong Z, Chengbin W. An indoor positioning system using Channel State Information based on TrAdaBoost Tranfer Learning //2021 4th International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE). IEEE, 2021: 1286-1293.

[26] Wang W, Wang C, Wang Z, et al. Abnormal detection technology of industrial control system based on transfer learning. Applied Mathematics and Computation, 2022, 4(12): 126-139.

[27] GUO Xiaoping, LIU Shiyang, LI Yuan. Research on multi-operating condition process fault detection method based on sparse residual distance. Journal of Automation, 2019, 45(3): 618-625.

[28] ZHAO Zhihong, YANG Shaopu. A sample entropy-based bearing fault diagnosis method. Vibration and Impact, 2012, 31(6): 136-140.

[29] Su W-S, Wang F-T, Zhang Z-X, et al. Application of EM noise reduction and spectral cliff method in early fault diagnosis of rolling bearings. Vibration and Shock, 2010, 29(3): 18-21.

[30] Lei Yaguo, Han Tianyu, Wang Biao, et al. Interpretation of XJTU-SY rolling bearing accelerated life test dataset. Journal of Mechanical Engineering, 2019, 55(16): 1-6.