Skip to main content
Springer Nature Link
Log in

A novel spatio-temporal hybrid neural network for remaining useful life prediction

  • Published:
The Journal of Supercomputing Aims and scope Submit manuscript

Abstract

Remaining useful life (RUL) prediction is a crucial mission for the prognostic and health management (PHM) of machinery equipment in modern industry. Due to ignorance of the spatial relationship among multiple sensors, the current RUL prediction methods do not perform adequately in terms of accuracy and comprehensive performance. In order to improve the performance, this paper proposes a novel spatio-temporal hybrid neural network (STHNN) method to capture spatio-temporal features of the time sequence signals from multi-sensors at different locations on mechanical equipment. In this method, the spatial relationship among sensors is constructed by a sensor graph based on the cosine similarity among them. Spatial features of the sensor graph are then extracted by the improved residual graph attention module, which is the first part of the spatio-temporal hybrid neural network. As the second part of the STHNN, the Transformer encoder module is employed to obtain the temporal features. At the same time, this study proposes a new evaluation metric named as comprehensive evaluation function (CEF) to assess overall performance of the STHNN on the entire dataset. To confirm the accuracy and validity of the method, experiments are conducted on the commercial modular aero-propulsion system simulation (C-MAPSS) dataset. The comparison with existing state-of-the-art methods on the same dataset demonstrates performance of the provided STHNN is significantly greater than the current methods.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13

Similar content being viewed by others

Data and materials availability

The data used to support the findings of this study are available from the corresponding author upon request.

References

  1. Saidi L, Benbouzid M (2021) Prognostics and health management of renewable energy systems: state of the art review, challenges, and trends. Electronics 10(22):2732

    Article  Google Scholar 

  2. Mo Y, Wu Q, Li X, Huang B (2021) Remaining useful life estimation via transformer encoder enhanced by a gated convolutional unit. J Intell Manuf 32:1997–2006

    Article  Google Scholar 

  3. Saxena A, Goebel K, Simon D, Eklund NHW (2008) Damage propagation modeling for aircraft engine run-to-failure simulation. In: 2008 International Conference on Prognostics and Health Management. 1–9

  4. Kuncham E, Sen S, Kumar P, Pathak H (2022) An online model-based fatigue life prediction approach using extended Kalman filter. Theor Appl Fract Mech 117:16

    Article  Google Scholar 

  5. Nejjar I, Geissmann F, Zhao M, Taal C, Fink O (2023) Domain adaptation via alignment of operation profile for remaining useful lifetime prediction

  6. Gu X, Shang YL, Kang YZ, Li JL, Mao ZH, Zhang CH (2023) An Early minor-fault diagnosis method for lithium-ion battery packs based on unsupervised learning. IEEE-CAA J Automatica Sin 10(3):810–812

    Article  Google Scholar 

  7. Chen Y, Peng G, Zhu Z, Li S (2020) A novel deep learning method based on attention mechanism for bearing remaining useful life prediction. Appl Soft Comput 86:105919

    Article  Google Scholar 

  8. Cameron ZA, Krantz TL (2023) Statistical distribution of gear surface fatigue lives at high reliability. Int J Fatigue 167:10

    Article  Google Scholar 

  9. Wei H, Zhang QH, Gu Y (2023) Remaining useful life prediction of bearings based on self-attention mechanism, multi-scale dilated causal convolution, and temporal convolution network. Meas Sci Technol 34(4):16

    Article  Google Scholar 

  10. Li Y, Huang X, Ding P, Zhao C (2021) Wiener-based remaining useful life prediction of rolling bearings using improved Kalman filtering and adaptive modification. Measurement 182:109706

    Article  Google Scholar 

  11. Babu GS, Zhao P, Li X (2016) Deep convolutional neural network based regression approach for estimation of remaining useful life. DASFAA

  12. Li X, Ding Q, Sun J (2018) Remaining useful life estimation in prognostics using deep convolution neural networks. Reliab Eng Syst Saf 172:1–11

    Article  Google Scholar 

  13. Zhang C, Lim P, Qin AK, Tan KC (2017) Multiobjective deep belief networks ensemble for remaining useful life estimation in prognostics. IEEE Trans Neural Netw Learn Syst 28:2306–2318

    Article  Google Scholar 

  14. Wu Y, Yuan M, Dong S, Lin L, Liu Y (2018) Remaining useful life estimation of engineered systems using vanilla LSTM neural networks. Neurocomputing 275:167–179

    Article  Google Scholar 

  15. Miao H, Li B, Sun C, Liu J (2019) Joint learning of degradation assessment and RUL prediction for aeroengines via dual-task deep LSTM networks. IEEE Trans Ind Inf 15:5023–5032

    Article  Google Scholar 

  16. Xu M, Bai YY, Qian PJ (2022) Remaining useful life prediction based on improved LSTM hybrid attention neural network. In: 18th International Conference on Intelligent Computing (ICIC); 2022 Aug 07–11. Xian, Peoples R China. Springer International Publishing Ag, Cham

  17. Wu Z, Yu S, Zhu X, Ji Y, Pecht MG (2019) A weighted deep domain adaptation method for industrial fault prognostics according to prior distribution of complex working conditions. IEEE Access 7:139802–139814

    Article  Google Scholar 

  18. Al-Dulaimi A, Zabihi S, Asif A, Mohammadi A (2019) A multimodal and hybrid deep neural network model for remaining useful life estimation. Comput Ind 108:186–196

    Article  Google Scholar 

  19. Yu W, Kim IY, Mechefske CK (2020) An improved similarity-based prognostic algorithm for RUL estimation using an RNN autoencoder scheme. Reliab Eng Syst Saf 199:106926

    Article  Google Scholar 

  20. Zhu Q, Xiong Q, Yang Z, Yu Y (2022) A novel feature-fusion-based end-to-end approach for remaining useful life prediction. J Intell Manuf. https://doi.org/10.1007/s10845-022-02015-x

    Article  Google Scholar 

  21. Li T, Zhao Z, Sun C, Yan R, Chen X (2021) Hierarchical attention graph convolutional network to fuse multi-sensor signals for remaining useful life prediction. Reliab Eng Syst Saf 215:107878

    Article  Google Scholar 

  22. Velickovic P, Cucurull G, Casanova A, Romero A, Lio’ P, Bengio Y (2018) Graph attention networks. ArXiv. 2018;abs/1710.10903

  23. He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 770–778

  24. Vaswani A, Shazeer NM, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. (2017) Attention is All you Need. In: NIPS

  25. Heimes F (2008) Recurrent neural networks for remaining useful life estimation. In: 2008 International Conference on Prognostics and Health Management. 1–6

  26. Zhao CY, Huang XZ, Li YX, Li SJ (2021) A novel cap-LSTM model for remaining useful life prediction. IEEE Sens J 21(20):23498–23509

    Article  Google Scholar 

Download references

Acknowledgements

This study was supported by The Guizhou Science and Technology Support Project (QKHZC〔2022〕YB074) and Science and Technology Foundation of Guizhou Province (QKHJC-ZK〔2021〕YB271).

Funding

The Guizhou Science and Technology Support Project (QKHZC〔2022〕YB074) and Science and Technology Foundation of Guizhou Province (QKHJC-ZK〔2021〕YB271).

Author information

Authors and Affiliations

  1. Key Laboratory of Advanced Manufacturing Technology of the Ministry of Education, Guizhou University, Guiyang, 550025, China

    Tao Wang

  2. State Key Laboratory of Public Big Data, College of Computer Science and Technology, Guizhou University, Guiyang, 550025, China

    Xianghong Tang & Jianguang Lu

  3. Chongqing Industrial Big Data Innovation Center Co., Ltd., Chongqing, 400707, China

    Jianguang Lu

  4. Guizhou Xin Si Wei Technology Co. Ltd., Guiyang, 550001, China

    Fangjie Liu

Authors
  1. Tao Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Xianghong Tang
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Jianguang Lu
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Fangjie Liu
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Data analysis and writing contributed by Tao Wang; Supervision contributed by Xianghong Tang, Jianguang Lu, Fangjie Liu.

Corresponding author

Correspondence to Xianghong Tang.

Ethics declarations

Ethical approval

Not applicable.

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, T., Tang, X., Lu, J. et al. A novel spatio-temporal hybrid neural network for remaining useful life prediction. J Supercomput 79, 19095–19117 (2023). https://doi.org/10.1007/s11227-023-05404-y

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11227-023-05404-y

Keywords