IDEAS home Printed from https://ideas.repec.org/a/eee/renene/v256y2026ipbs0960148125016337.html

MSTI-GNN: A multi-scale spatiotemporal interactive graph neural network method for precise hydropower unit status prediction

Author

Listed:
  • Yi, Tongqiang
  • Guo, Jiang
  • Meng, Yang
  • Ling, Yuewei
  • Ke, Yiming
  • Guo, Zhilong

Abstract

As hydropower continues to increase its share in the energy structure, accurately predicting the operational status trends of hydropower units is essential for ensuring their reliability and optimizing maintenance strategies. However, existing methods exhibit significant shortcomings in capturing dynamic dependencies, long-term evolution patterns, and multi-scale feature fusion, thereby limiting their prediction accuracy and generalization capabilities. To address these challenges, this paper proposes a multi-scale spatiotemporal interactive graph neural network (MSTI-GNN). This network integrates long-term stable dependencies with short-term dynamic features through a dynamic adaptive graph generation module, precisely capturing the time-varying relationships among sensor nodes and overcoming the limitations of traditional static graph construction. Subsequently, the spatiotemporal interaction block (STIB) combines the dynamic MixHop spatial layer (DMSL) with the CSA-TCN temporal layer (CTTL), achieving an efficient fusion of spatial and temporal information and comprehensively reflecting the complex interactions within the sensor network. Finally, the multi-scale spatiotemporal interaction module (MSTIM) employs skip connections, residual connections, and a feature fusion block (CSFblock) to dynamically integrate local short-term features with global long-term characteristics, completing hierarchical aggregation of features across different time scales and spatial hierarchies, thereby enabling precise trend prediction. Through a series of comparative experiments, MSTI-GNN achieves RMSE values of 0.3229, 0.2239, and 0.5347 for the UGX, WGX, and FLX monitoring targets, respectively, outperforming the best baseline models by 13.94 %, 14.64 %, and 9 %. It also achieves low MAE values (0.2360, 0.1904, and 0.4267, respectively) and stable R2 values (0.9518, 0.9540, and 0.9598, respectively), demonstrating its effectiveness. The model further exhibits excellent stability in medium- and long-term prediction tasks, validating its robustness.

Suggested Citation

  • Yi, Tongqiang & Guo, Jiang & Meng, Yang & Ling, Yuewei & Ke, Yiming & Guo, Zhilong, 2026. "MSTI-GNN: A multi-scale spatiotemporal interactive graph neural network method for precise hydropower unit status prediction," Renewable Energy, Elsevier, vol. 256(PB).
    • Handle: RePEc:eee:renene:v:256:y:2026:i:pb:s0960148125016337
    DOI: 10.1016/j.renene.2025.123969
    as

    Download full text from publisher

    File URL: http://www.sciencedirect.com/science/article/pii/S0960148125016337
    Download Restriction: Full text for ScienceDirect subscribers only
    File URL: https://libkey.io/10.1016/j.renene.2025.123969?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    As the access to this document is restricted, you may want to

    for a different version of it.

    References listed on IDEAS

    as
    1. Omar Farhan Al-Hardanee & Hüseyin Demirel, 2024. "Hydropower Station Status Prediction Using RNN and LSTM Algorithms for Fault Detection," Energies, MDPI, vol. 17(22), pages 1-23, November.
    2. Abdo Y. Alfakih, 2018. "Euclidean Distance Matrices and Their Applications in Rigidity Theory," Springer Books, Springer, number 978-3-319-97846-8, October.
    3. Li, Xiang & Ding, Qian & Sun, Jian-Qiao, 2018. "Remaining useful life estimation in prognostics using deep convolution neural networks," Reliability Engineering and System Safety, Elsevier, vol. 172(C), pages 1-11.
    4. Zhang, Yuning & Zheng, Xianghao & Li, Jinwei & Du, Xiaoze, 2019. "Experimental study on the vibrational performance and its physical origins of a prototype reversible pump turbine in the pumped hydro energy storage power station," Renewable Energy, Elsevier, vol. 130(C), pages 667-676.
    5. Pham, Quang Hung & Gagnon, Martin & Antoni, Jérôme & Tahan, Antoine & Monette, Christine, 2022. "Prediction of hydroelectric turbine runner strain signal via cyclostationary decomposition and kriging interpolation," Renewable Energy, Elsevier, vol. 182(C), pages 998-1011.
    6. Betti, Alessandro & Crisostomi, Emanuele & Paolinelli, Gianluca & Piazzi, Antonio & Ruffini, Fabrizio & Tucci, Mauro, 2021. "Condition monitoring and predictive maintenance methodologies for hydropower plants equipment," Renewable Energy, Elsevier, vol. 171(C), pages 246-253.
    7. Boyi Xiao & Yun Zeng & Yidong Zou & Wenqing Hu, 2023. "Hydropower Unit State Evaluation Model Based on AHP and Gaussian Threshold Improved Fuzzy Comprehensive Evaluation," Energies, MDPI, vol. 16(15), pages 1-19, July.
    8. Zhang, Zhiyang & Bu, Yifeng & Wu, Haitao & Wu, Linyan & Cui, Lin, 2023. "Parametric study of the effects of clump weights on the performance of a novel wind-wave hybrid system," Renewable Energy, Elsevier, vol. 219(P1).
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Basora, Luis & Viens, Arthur & Chao, Manuel Arias & Olive, Xavier, 2025. "A benchmark on uncertainty quantification for deep learning prognostics," Reliability Engineering and System Safety, Elsevier, vol. 253(C).
    2. Fatemeh Hajimohammadali & Emanuele Crisostomi & Mauro Tucci & Nunzia Fontana, 2024. "Evaluating Deep Learning Networks Versus Hybrid Network for Smart Monitoring of Hydropower Plants," Energies, MDPI, vol. 17(22), pages 1-16, November.
    3. Xu, Zhiqiang & Zhang, Yujie & Miao, Qiang, 2024. "An attention-based multi-scale temporal convolutional network for remaining useful life prediction," Reliability Engineering and System Safety, Elsevier, vol. 250(C).
    4. Lin, Yan-Hui & Chang, Liang & Guan, Lu-Xin, 2024. "Enhanced stochastic recurrent hybrid model for RUL Predictions via Semi-supervised learning," Reliability Engineering and System Safety, Elsevier, vol. 248(C).
    5. Chen, Sheng & Wang, Jing & Zhang, Jian & Yu, Xiaodong & He, Wei, 2020. "Transient behavior of two-stage load rejection for multiple units system in pumped storage plants," Renewable Energy, Elsevier, vol. 160(C), pages 1012-1022.
    6. Li, Yuanfu & Chen, Yifan & Shao, Haonan & Zhang, Huisheng, 2023. "A novel dual attention mechanism combined with knowledge for remaining useful life prediction based on gated recurrent units," Reliability Engineering and System Safety, Elsevier, vol. 239(C).
    7. Xiao, Lei & Tang, Junxuan & Zhang, Xinghui & Bechhoefer, Eric & Ding, Siyi, 2021. "Remaining useful life prediction based on intentional noise injection and feature reconstruction," Reliability Engineering and System Safety, Elsevier, vol. 215(C).
    8. Othman, Mohd Edzham Fareez & Sidek, Lariyah Mohd & Basri, Hidayah & El-Shafie, Ahmed & Ahmed, Ali Najah, 2025. "Climate challenges for sustainable hydropower development and operational resilience: A review," Renewable and Sustainable Energy Reviews, Elsevier, vol. 209(C).
    9. Yuyan Yin & Jie Tian & Xinfeng Liu, 2025. "Remaining useful life prediction based on parallel multi-scale feature fusion network," Journal of Intelligent Manufacturing, Springer, vol. 36(5), pages 3111-3127, June.
    10. Duan, Juan & Peng, Zeyu & Chen, Luyang & Zeng, Yun, 2025. "A review of OMA parameter identification for hydro-turbine unit: Challenges in condition monitoring," Renewable and Sustainable Energy Reviews, Elsevier, vol. 217(C).
    11. Li, Yuanfu & Chen, Yao & Hu, Zhenchao & Zhang, Huisheng, 2023. "Remaining useful life prediction of aero-engine enabled by fusing knowledge and deep learning models," Reliability Engineering and System Safety, Elsevier, vol. 229(C).
    12. Marcin Witczak & Marcin Mrugalski & Bogdan Lipiec, 2021. "Remaining Useful Life Prediction of MOSFETs via the Takagi–Sugeno Framework," Energies, MDPI, vol. 14(8), pages 1-23, April.
    13. Liu, Lu & Song, Xiao & Zhou, Zhetao, 2022. "Aircraft engine remaining useful life estimation via a double attention-based data-driven architecture," Reliability Engineering and System Safety, Elsevier, vol. 221(C).
    14. Zhang, Jiusi & Li, Xiang & Tian, Jilun & Luo, Hao & Yin, Shen, 2023. "An integrated multi-head dual sparse self-attention network for remaining useful life prediction," Reliability Engineering and System Safety, Elsevier, vol. 233(C).
    15. Yaping Li & Enrico Zio & Ershun Pan, 2021. "An MEWMA-based segmental multivariate hidden Markov model for degradation assessment and prediction," Journal of Risk and Reliability, , vol. 235(5), pages 831-844, October.
    16. Zhang, Mingyuan & He, Chen & Huang, Chengxuan & Yang, Jianhong, 2024. "A weighted time embedding transformer network for remaining useful life prediction of rolling bearing," Reliability Engineering and System Safety, Elsevier, vol. 251(C).
    17. Zhuang, Jichao & Jia, Minping & Ding, Yifei & Ding, Peng, 2021. "Temporal convolution-based transferable cross-domain adaptation approach for remaining useful life estimation under variable failure behaviors," Reliability Engineering and System Safety, Elsevier, vol. 216(C).
    18. Khaled Akkad & David He, 2023. "A dynamic mode decomposition based deep learning technique for prognostics," Journal of Intelligent Manufacturing, Springer, vol. 34(5), pages 2207-2224, June.
    19. Weeber, Christoph, 2024. "Development of a cost optimal predictive maintenance strategy," Junior Management Science (JUMS), Junior Management Science e. V., vol. 9(3), pages 1805-1835.
    20. Nadire Cavus & Yakubu Bala Mohammed & Mohammed Nasiru Yakubu, 2021. "An Artificial Intelligence-Based Model for Prediction of Parameters Affecting Sustainable Growth of Mobile Banking Apps," Sustainability, MDPI, vol. 13(11), pages 1-21, May.

    More about this item

    Keywords

    ;
    ;
    ;
    ;

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:eee:renene:v:256:y:2026:i:pb:s0960148125016337. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Catherine Liu (email available below). General contact details of provider: http://www.journals.elsevier.com/renewable-energy .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.