Attention-Gaussian-LSTM-Wiener based remaining useful life prediction method

Abstract

Most machine learning-based remaining useful life (RUL) prediction methods only yield point predictions, and their “black-box” nature results in low interpretability. Stochastic process-based modeling can predict RUL probability density function (PDF), yet it often suffers from inaccurate modeling and failure to fully utilize historical degradation data of the same equipment type. To overcome these limitations, this paper integrates the two approaches and proposes an Attention-Gaussian-LSTM-Wiener (AG-LSTM-Wiener)-based RUL prediction method, enabling dynamic weighted fusion of predicted PDFs. An AG-LSTM-Wiener model with a two-branch structure is constructed. Health indicator (HI) is fed into the corresponding branch models to generate two different PDF curves. Decision blocks are employed to estimate RUL, from which weights are derived to achieve dynamic weighted fusion of the PDFs. Experiments on the CMPASS turbofan engine degradation dataset validate the proposed method’s effectiveness. Results demonstrate that the proposed method not only prevents PDF curve distortion but also improves the prediction accuracy compared with other methods. With the root mean squared error (RMSE) and Score reduced by 32.8% and 46.1% on average, and the mean squared error of PDF (

{\mathrm{MSE}}_{\mathrm{PDF}}

) improved by 99.3% compared to AG-LSTM, which exhibits the best performance among the contrast methods.

Keywords

RUL prediction / Attention-Gaussian-LSTM-Wiener / Probability density function / Decision blocks / Dynamic weighted fusion

References

[1] Zeng S., Pecht M.G., Wu J.. Status and perspectives of prognostics and health management technologies. Acta Aeronaut. Astronaut. Sin., 2005, 26(5626

[2] Ye Z., Xie M.. Stochastic modelling and analysis of degradation for highly reliable products. Appl. Stoch. Models Bus. Ind., 2015, 31(1): 16-32.

[3] Li H., Zhang Z., Li T., Si X.. A review on physics-informed data-driven remaining useful life prediction: challenges and opportunities. Mech. Syst. Signal Process., 2024, 209. 111120

[4] Liu Y., Wang J., Shen Y.. Research on verification of sensor fault diagnosis based on bp neural network. 2020 11th International Conference on Prognostics and System Health Management (PHM-2020 Jinan), 2020Los AlamitosIEEE456-460.

[5] Zhan H.. Application of rough set and support vector machine in fault diagnosis of power electronic circuit. 2010 2nd IEEE International Conference on Information Management and Engineering, 2010Los AlamitosIEEE289-292.

[6] Yang S., Li W., Wang C.. The intelligent fault diagnosis of wind turbine gearbox based on artificial neural network. 2008 International Conference on Condition Monitoring and Diagnosis, 2008Los AlamitosIEEE1327-1330.

[7] Sateesh Babu G., Zhao P., Li X.. Deep convolutional neural network based regression approach for estimation of remaining useful life. Database Systems for Advanced Applications: 21st International Conference, DASFAA 2016, 2016BerlinSpringer214-228. 21

[8] Noussis A., O’Neil R., Saif A., Khatab A., Diallo C.. A hybrid Bi-LSTM model for data-driven maintenance planning. Auton. Intell. Syst., 2025, 5(1): 1-23.

[9] Fu H., Liu Y.. A deep learning-based approach for electrical equipment remaining useful life prediction. Auton. Intell. Syst., 2022, 2(1. 16

[10] Jin Z., Chen C., Syntetos A., Liu Y.. Enhanced bearing rul prediction based on dynamic temporal attention and mixed MLP. Auton. Intell. Syst., 2025, 5(1. 2

[11] P. Pethkar, Rul estimation of N-CMAPSS turbofan engines using deep learning with customized penalty and expanded sensors (2025)

[12] Wang C., Jiang W., Shi L., Zhang L.. Rolling bearing remaining useful life prediction using deep learning based on high-quality representation. Sci. Rep., 2025, 15(1. 8228

[13] Z. Zhang, D. Li, S. Wu, J. Cai, B. Zhang, S.K. Ng, Z. Zheng, Bace-rul: a bi-directional adversarial network with covariate encoding for machine remaining useful life prediction. arXiv preprint arXiv:2503.11730 (2025)

[14] G. Muthukumar, J. Philip, CNN-LSTM hybrid deep learning model for remaining useful life estimation. I: arXiv preprint (2024)

[15] Si X.-S., Wang W., Hu C.-H., Zhou D.-H., Pecht M.G.. Remaining useful life estimation based on a nonlinear diffusion degradation process. IEEE Trans. Reliab., 2012, 61(1): 50-67.

[16] Salinas D., Flunkert V., Gasthaus J., Januschowski T.. DeepAR: probabilistic forecasting with autoregressive recurrent networks. Int. J. Forecast., 2020, 36(3): 1181-1191.

[17] Ramasso E., Saxena A.. Performance benchmarking and analysis of prognostic methods for CMAPSS datasets. Int. J. Progn. Health Manag., 2014, 5(2): 1-15

[18] Lei Y., Li N., Guo L., Li N., Yan T., Lin J.. Machinery health prognostics: a systematic review from data acquisition to RUL prediction. Mech. Syst. Signal Process., 2018, 104: 799-834.

[19] Cao S., Zhang H., Xia K., Meng G., Wang A., Ji Y.. Remaining useful life prediction for aero-engine based on Gaussian-LSTM. 2024 4th International Conference on Electronic Information Engineering and Computer (EIECT), 2024Los AlamitosIEEE383-387.

Funding

National Natural Science Foundation of China(62273030)

Innovative Research Group Project of the National Natural Science Foundation of China(62303353)

Interdisciplinary Research Project for Young Teachers of USTB (Fundamental Research Funds for the Central Universities)(FRF-IDRY-GD21-004)

Open Fund of Intelligent Control Laboratory

RIGHTS & PERMISSIONS

The Author(s)