Rolling Bearing Health Indicator: From Design to Modeling and Evaluation

Abstract

As a key component of industrial machinery, accurate prediction of the degradation trend of rolling bearings is crucial for equipment safety. However, traditional health indicator (HI) extraction methods often suffer from feature redundancy, and prediction models lack the ability to capture spatial dimension features, leading to significant prediction errors. To address these issues, 16 time-frequency domain features were first extracted, and a new HI was constructed by combining the Gaussian Process latent variable model (GPLVM) for non-linear feature fusion and exponentially weighted moving average (EWMA) for smoothing. Additionally, a spatial-temporal convolutional long short-term memory network (ST-CNet) was proposed, which integrates a 3-layer CLSTM, fully connected layers, and batch normalization to effectively capture local and long-term spatiotemporal dependencies. Case studies on IMS bearing datasets show that the constructed HI accurately describes the degradation process, and ST-CNet achieves superior performance with lower MAE and RMSE compared to existing methods.

Keywords

Prediction of degradation trend / Multi-feature fusion / ST-CNet / Rolling bearing / Health indicator

Statement of the Use of Generative AI and AI-Assisted Technologies in the Writing Process

During the preparation of this work, the authors used Deepseek to improve readability and language. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Author Contributions

G.H. and N.W., conceptualization, methodology, writing—original draft; S.W. and Q.W., formal analysis, visualization; W.W., L.L. and Y.F., supervision and writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Ethics Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not available.

Funding

This research was funded by the Fundamental Research Funds For the Central Universities (No. 25CAFUC04016), Sichuan Civil Aviation Flight Technology and Flight Safety Engineering Technology Research Center Project, (No. GY2024-41E) and the National Natural Science Foundation of China (No. 12304258).

Declaration of Competing Interest

The authors declare that there are no know conflicts of financial interests and it has not been published in other journals.

References

[1] Lan X, Li Y, Su Y, Meng L, Kong X, Xu T. Performance degradation prediction model of rolling bearing based on self-checking long short-term memory network. Meas. Sci. Technol. 2023, 34, 015016. doi:10.1088/1361-6501/ac90dc.

[2] Liu H, Yuan R, Lv Y, Yang X, Li H, Song G. Degradation tracking of rolling bearings based on local polynomial phase space warping. IEEE Trans. Reliab. 2023, 73, 1380-1392. doi:10.1109/TR.2023.3335899.

[3] Zhang Y, Peng Y, Liu L. Degradation estimation of electromechanical actuator with multiple failure modes using integrated health indicators. IEEE Sens. J. 2020, 20, 7216-7225. doi:10.1109/JSEN.2020.2978140.

[4] Yan T, Wang D, Sun S, Shen C, Peng Z. Novel sparse representation degradation modeling for locating informative frequency bands for machine performance degradation assessment. Mech. Syst. Signal Process. 2022, 179, 109372. doi:10.1016/j.ymssp.2022.109372.

[5] Chen B, Song D, Cheng Y, Zhang W, Huang B, Muhamedsalish Y. IGIgram: An improved Gini index-based envelope analysis for rolling bearing fault diagnosis. J. Dyn. Monit. Diagn. 2022, 1, 111-124. doi:10.37965/jdmd.2022.65.

[6] Yan T, Wang D, Zheng M, Xia T, Pan E, Xi L. Fisher’s discriminant ratio based health indicator for locating informative frequency bands for machine performance degradation assessment. Mech. Syst. Signal Process. 2022, 162, 108053. doi:10.1016/j.ymssp.2021.108053.

[7] Gu Y, Bi Q, Qiu G. Practical health indicator construction methodology for bearing ensemble remaining useful life prediction with ISOMAP-DE and ELM-WPHM. Meas. Sci. Technol. 2021, 32, 25007. doi:10.1088/1361-6501/ac3855.

[8] Chen D, Qin Y, Wang Y, Zhou J. Health indicator construction by quadratic function-based deep convolutional auto-encoder and its application into bearing RUL prediction. ISA Trans. 2021, 114, 44-56. doi:10.1016/j.isatra.2020.12.052.

[9] Kunzelmann B, Rycerz P, Xu Y, Arakere NK, Kadiric A. Prediction of rolling contact fatigue crack propagation in bearing steels using experimental crack growth data and linear elastic fracture Mechanics. Int. J. Fatigue 2023, 168, 107449. doi:10.1016/j.ijfatigue.2022.107449.

[10] Zaid M, Doquet V, Chiaruttini V, Depouhon P, Bonnand V, Pacou D. On the influence of secondary branches on crack propagation in rolling contact Fatigue. Int. J. Fatigue 2024, 182, 108211. doi:10.1016/j.ijfatigue.2024.108211.

[11] Voskamp AP, Mittemeijer EJ. State of residual stress induced by cyclic rolling contact loading. Mater. Sci. Technol. 1997, 13, 430-438. doi:10.1179/mst.1997.13.5.430.

[12] Tsutsumi M, Watankui D, Miyamoto Y, Asa N. Basic study on evaluation method for fatigue cack initiation and propagation in carburized steel for large rolling bearings considering residual stress in three directions. J. Soc. Mater. Sci. Jpn. 2024, 73, 793-800. doi:10.2472/jsms.73.793.

[13] Sun C, Xiu S, Xing C, Lu H, Wang D, Wang R. Influence of prestress grinding hardening residual stress on rolling contact Fatigue. Mater. Sci. Technol. 2022, 38, 716-729.doi:10.1080/02670836.2022.2062881.

[14] You S, Tang J, Zhou W, Zhou W, Zhao J, Chen H. Research on calculation of contact fatigue life of rough tooth surface considering residual Stress. Eng. Fail. Anal. 2022, 140, 106459. doi:10.1016/j.engfailanal.2022.106459.

[15] Liu Y, Kang G, Wang C.Effect of residual stresses on the crack initiation life for rolling contact fatigue. J. Tsinghua Univ. Nat. Sci. Ed. 2025, 12, 1664-1667. doi:10.3321/j.issn:1000-0054.2005.12.023.

[16] Dai W, Zhang C, Yue H, Li Q, Guo C, Zhang J, et al. A review on the fatigue performance of Micro-arc oxidation coated Al alloys with micro-defects and residual stress. J. Mater. Res. Technol. 2023, 25, 4554-4581. doi:10.1016/j.jmrt.2023.06.244.

[17] Xia Z, Wu D, Zhang X, Wang J, Han E. Rolling Contact Fatigue Failure Mechanism of Bearing Steel on Different Surface Roughness Levels Under Heavy Load. Int. J. Fatigue 2024, 182, 107310. doi:10.1016/j.ijfatigue.2023.108042.

[18] Pape F, Neubauer T, Maiß O, Denkena B, Poll G. Influence of residual stresses introduced by manufacturing processes on bearing endurance time. Tribol. Lett. 2017, 65, 70. doi:10.1007/s11249-017-0855-3.

[19] Denkena B, Poll G, Maiß O, Pape F, Neubauer T. Enhanced boundary zone rolling contact fatigue strength through hybrid machining by hard turn-rolling. Bear. World J. 2016, 1, 87-102.

[20] Cui L, Su Y. Contact fatigue life prediction of rolling bearing considering machined surface Integrity. Ind. Lubr. Tribol. 2021, 74, 73-80. doi:10.1108/ilt-08-2021-0345.

[21] Choi Y, Yang X. Impact evaluation of rolling contact fatigue life Models. J. Mech. Sci. Technol. 2012, 26, 1865-1873. doi:10.1007/s12206-012-0431-6.

[22] Lei Y, Li N, Gontarz S, Lin J, Radkowski S, Dybala J. A model-based method for remaining useful life prediction of machinery. IEEE Trans. Reliab. 2016, 65, 1314-1326. doi:10.1109/TR.2016.2570568.

[23] Hu C, Youn BD, Wang P. Ensemble of data-driven prognostic algorithms for robust prediction of remaining useful life. Reliab. Eng. Syst. Saf. 2012, 103, 120-135. doi:10.1016/j.ress.2012.03.008.

[24] Kumar P, Kumaraswamidhas L, Laha S. Selection of efficient degradation features for rolling element bearing prognosis using Gaussian Process Regression method. ISA Trans. 2021, 112, 386-401. doi:10.1016/j.isatra.2020.12.020.

[25] Cui L, Wang X, Wang H, Ma J. Research on remaining useful life prediction of rolling element bearings based on time-varying kalman filter. IEEE Trans. Instrum. Meas. 2020, 69, 2858-2867. doi:10.1109/TIM.2019.2924509.

[26] Li X, Zhang W, Ding Q. Deep learning-based remaining useful life estimation of bearings using multi-scale feature extraction. Reliab. Eng. Syst. Saf. 2019, 182, 208-218. doi:10.1016/j.ress.2018.11.011.

[27] Xia M, Li T, Shu T, Wan J, De Silva C, Wang Z. A two-stage approach for the remaining useful life prediction of bearings using deep neural networks. IEEE Trans. Ind. Inform. 2019, 15, 3703-3711. doi:10.1109/TII.2018.2868687.

[28] Cheng C, Ma G, Zhang Y, Sun M, Teng F, Ding H, et al. A deep learning-based remaining useful life prediction approach for bearings. IEEE/ASME Trans. Mechatron. 2020, 25, 1243-1254. doi:10.1109/TMECH.2020.2971503.

[29] Yang L, Wang J. GA-BP Algorithm Predicts the Degrading Trend of Rolling Bearings. In Proceedings of the 2021 IEEE International Conference on Sensing, Diagnostics, Prognostics, and Control (SDPC), Weihai, China, 13-15 August 2021; pp. 69-73.

[30] Guo L, Li N, Jia F, Lei Y, Lin J. A recurrent neural network based health indicator for remaining useful life prediction of bearings. Neurocomputing 2017, 240, 98-109. doi:10.1016/j.neucom.2017.02.045.

[31] Hochreiter S, Schmidhuber J. Long short-term memory. Neural. Comput. 1997, 9, 1735-1780. doi:10.1162/neco.1997.9.8.1735.

[32] Cho K, et al.Van Merriënboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv 2014, arXiv:1406.1078. doi:10.48550/arXiv.1406.1078.

[33] Hu Y, Wei R, Yang Y, Li X, Huang Z, Liu Y, et al. Performance degradation prediction using LSTM with optimized parameters. Sensors 2022, 22, 2407. doi:10.3390/s22062407.

[34] Zheng X, Qian Y, Wang S. GRU prediction for performance degradation of rolling bearings based on optimal wavelet packet and Mahalanobis distance. J. Vib. Shock. 2020, 39, 39-49. doi:10.13465/j.cnki.jvs.2020.17.006

[35] Huang C, Huang H, Li Y. A Bidirectional LSTM prognostics method under multiple operational conditions. IEEE Trans. Ind. Electron. 2019, 66, 8792-8802. doi:10.1109/TIE.2019.2891463.

[36] Lei N, Huang F, Li C. Rolling bearing fault diagnosis based on variational mode decomposition and weighted multidimensional feature entropy Fusion. J. Vibroeng. 2024, 26, 590-614. doi:10.21595/jve.2023.23673.

[37] Wang G, Chen J, Zhong Z, Hu S. Multi-source heterogeneous fusion entropy ratio distance feature of bearing performance degradation based on DTW. Vibroeng. Procedia 2021, 39, 17-23. doi:10.21595/vp.2021.22269.

[38] Zhu J, Jiang Q, Shen Y, Xu F, Zhu Q. Res-HSA: Residual hybrid network with self-attention mechanism for RUL prediction of rotating Machinery. Eng. Appl. Artif. Intell. 2023, 124, 106491. doi:10.1016/j.engappai.2023.106491.

[39] Liao X, Chen S, Wen P, Zhao S. Remaining useful life with self-attention assisted physics-informed neural network. Adv. Eng. Inform. 2023, 58, 102195. doi:10.1016/j.aei.2023.102195.

[40] Li T, Shi H, Bai X, Li N, Zhang K. Rolling bearing performance assessment with degradation twin modeling considering interdependent fault evolution. Mech. Syst. Signal Process. 2025, 224, 112194. doi:10.1016/j.ymssp.2024.112194.

[41] Akpudo UE, Hur JW. A feature fusion-based prognostics approach for rolling element bearings. J. Mech. Sci. Technol. 2020, 34, 4025-4035. doi:10.1007/s12206-020-2213-x.

[42] Li X, Gong Y, Liu W, Yin Y, Zheng Y, Nie L. Dual-track spatio-temporal learning for urban flow prediction with adaptive Normalization. Artif. Intell. 2024, 328, 104065. doi:10.1016/j.artint.2024.104065.

[43] Li Y, Zeng X, Shi Y. A spatial and temporal signal fusion based intelligent event recognition method for buried fiber distributed sensing System. Opt. Laser. Technol. 2023, 166, 109658. doi:10.1016/j.optlastec.2023.109658.

[44] Wang Y, Jiang Y, Kang S. The applications of time domain and frequency domain statistical factors on rolling bearing performance degradation assessment. Comput. Model. New Technol. 2014, 18, 192-198.

[45] Qiu G, Gu Y, Chen J. Selective health indicator for bearings ensemble remaining useful life prediction with genetic algorithm and Weibull proportional hazards model. Measurement 2020, 150, 107097. doi:10.1016/j.measurement.2019.107097.

[46] Zhou F, Park JH, Liu Y. Differential feature based hierarchical PCA fault detection method for dynamic fault. Neurocomputing 2016, 202, 27-35. doi:10.1016/j.neucom.2016.03.007.

[47] Li P, Chen S. A review on Gaussian process latent variable models. CAAI Trans. Intell. Technol. 2016, 1, 366-376. doi:10.1016/j.trit.2016.11.004.

[48] Li X, Jiang H, Xiong X, Shao H. Rolling bearing health prognosis using a modified health index based hierarchical gated recurrent unit network. Mech. Mach. Theory 2019, 133, 229-249. doi:10.1016/j.mechmachtheory.2018.11.005.

[49] Trifa A, Sbai AH, Chaari WL. Enhancing assessment of personalized multi-agent system through convlstm. Procedia Comput. Sci. 2017, 112, 249-259. doi:10.1016/j.procs.2017.08.239.

[50] Li G, Wei J, He J, Yang H, Meng F. Implicit Kalman filtering method for remaining useful life prediction of rolling bearing with adaptive detection of degradation stage transition point. Reliab. Eng. Syst. Saf. 2023, 235, 109269. doi:10.1016/j.ress.2023.109269.