A Health Condition Assessment and Safety Early Warning Framework for Hydropower Equipment Using Multi-Source Data Fusion and Decision Support Systems
DOI:
https://doi.org/10.13052/dgaej2156-3306.4153Keywords:
Multi-source data fusion, CNN–LSTM, hydropower equipment monitoring, structural risk index, safety early warning systemAbstract
Hydropower plants heavily depend on the performance of turbine generators, where unexpected equipment failure can cause downtime, loss, and even safety issues in the process. The traditional condition monitoring methods mainly focus on vibration signals detected by a single sensor, which is not integrated with multiple sources and lacks the ability to perceive the risk at the plant level. In addition, the majority of the studies do not consider the equipment level health condition and structural risk factors to develop unified decision support for safety issues. To address the issues, the paper proposes a framework for the integrated health assessment and safety early warning framework, which is developed by integrating the equipment level fault diagnosis model based on the CNN-LSTM network and the structural risk assessment model developed by the GloHydroRes dataset. The novelty of the proposed multi-level data fusion framework integrates equipment-level and plant-level information for safety assessment that incorporates multi-channel vibration and torque signals along with structural attributes such as dam height, reservoir volume, and installed capacity to derive a unified safety risk index. Experimental validation using the proposed approach on the SEU multi-sensor dataset achieved high classification performance with Accuracy of 0.9634, Precision of 0.8642, Recall of 0.9613, F1-score of 0.9065, and MCC of 0.8662. Moreover, multi-class ROC analysis indicated high performance with AUC values of 0.9936 (Low Risk), 0.9930 (Medium Risk), and 0.9997 (High Risk), outperforming individual CNN, LSTM, and Random Forest approaches. The unified safety index revealed that 46.6% of samples were classified as Low Risk, 34.1% of samples were classified as Medium Risk, and 19.3% of samples were classified as High Risk, thus validating the efficacy of the proposed decision support mechanism. Compared with traditional fault detection methods, the proposed approach improves the reliability of fault prediction results, reduces false alarm rates, and enables proactive maintenance decisions based on risk considerations.
Downloads
References
Su Y, Gan H, Ji Z (2024) Research on multi-parameter fault early warning for marine diesel engine based on PCA-CNN-BiLSTM. J Mar Sci Eng 12:965. https://doi.org/10.3390/jmse12060965.
He X et al. (2023) Situation awareness of energy internet of things in smart city based on digital twin: from digitization to informatization. IEEE Internet Things J 10(9):7439–7458. https://doi.org/10.1109/JIOT.2022.3203823.
Lu J et al. (2025) Multi-modal intelligent fault detection of screening equipment using sparse sensing and deep visual analysis: field application in coal preparation plants. Int J Coal Prep Util: 1–28. https://doi.org/10.1080/19392699.2025.2584172.
Wang Z, Wang S, Chen Y, Zhan M, Wang Y, Sun C (2025) A pattern-guided CIM vulnerability diagnosis framework for multi-sensor thermal management system in energy storage stations. Energies 18:6158. https://doi.org/10.3390/en18236158.
Peng J (2025) Digital twin technology and its application in water governance: China’s practices and achievements. Int J Water Resour Dev 41(5–6):974–989. https://doi.org/10.1080/07900627.2025.2544810.
Wu X et al. (2025) From forecasting to foresight: building an autonomous O&M brain for the new power system based on a cognitive digital twin. Electronics 14:4537. https://doi.org/10.3390/electronics14224537.
Tang T, Zhao C, Liu X, Sun S (2025) Digital twin-enabled deep learning for real-time fire situation awareness. Vis Comput 41(14):11857–11869. https://doi.org/10.1007/s00371-025-04131-0.
Wei A, Ke H, He S, Jiang M, Yao Z, Yi J (2025) Enhanced landslide risk evaluation in hydroelectric reservoir zones utilizing an improved random forest approach. Water 17:946. https://doi.org/10.3390/w17070946.
Dui H, Dong X, Wu X, Chen L, Bai G (2024) IoT-enabled risk warning and maintenance strategy optimization for tunnel-induced ground settlement. IEEE Internet Things J 11(13):22966–22981. https://doi.org/10.1109/JIOT.2024.3377440.
Zou Y et al. (2025) Multi-stakeholder risk assessment of a waterway engineering project during the decision-making stage from the perspective of sustainability. Sustainability 17:5372. https://doi.org/10.3390/su17125372.
Ahmed S, Alnaqbi AH, Al Hemairy M, Al Ahmad M (2018) ECG abnormality detection algorithm. Int J Adv Comput Sci Appl 9(8):215–221. https://doi.org/10.14569/ijacsa.2018.090827.
Al Hemairy M, Amin S, Hijji M, Serhani M, Al Ahmad M (2016) Integrated and scalable architecture for providing cost-effective remote health monitoring. In: Proc 9th Int Conf Dev eSystems Eng (DeSE), pp. 74–77. https://doi.org/10.1109/DeSE.2016.14.
Zhang P, Wang Q, Xu S, Zhu J, Zhong S, Zhang Y (2025) Early warning of coal mine production environment safety risks based on multi-source information feature fusion. Sustainability 17:2085. https://doi.org/10.3390/su17052085.
Plakantara SP, Karakitsiou A (2025) Transforming agrifood supply chains with digital technologies: a systematic review of safety and quality risk management. Oper Res Forum 6:113. https://doi.org/10.1007/s43069-025-00511-3.
Jiaxiang O et al. (2025) Intelligent hardware design for wire status monitoring integrating Beidou GNSS and inertial measurement technology in machine learning environment. Aust J Electr Electron Eng 22(4):575–586. https://doi.org/10.1080/1448837X.2024.2423982.
Li H, Zhao H, Shen Y, Zheng S, Zhang R (2024) Structural health monitoring and failure analysis of large-scale hydro-steel structures based on multi-sensor information fusion. Water 16:3167. https://doi.org/10.3390/w16223167.
Feng Z, Liang Q, Wei M, Li L, Bu Y, Xin Y (2025) Management system and optimal control for three-dimensional visualization and maintenance of thermal power plant. Energy Inform 8:71. https://doi.org/10.1186/s42162-025-00491-y.
Xin L, Zhao Y, Zhao Q (2025) BeiDou and SAR fusion technology with AI in reservoir dam monitoring for climate-based disaster mitigation. Discov Artif Intell 5:100. https://doi.org/10.1007/s44163-025-00320-x.
Shao S, Zhou F, Wang Y, Wu J (2025) An integrated multi-sensor information system for real-time reservoir monitoring and management. Sensors 25:5730. https://doi.org/10.3390/s25185730.
Sun J, Fei L (2025) A generative AI-enhanced case-based reasoning method for risk assessment: ontology modeling and similarity calculation framework. Mathematics 13:2735. https://doi.org/10.3390/math13172735.
Bai J et al. (2024) Fault diagnosis of pumped storage units—a novel data-model hybrid-driven strategy. Processes 12:2127. https://doi.org/10.3390/pr12102127.
Zhou X et al. (2025) A fault diagnosis method for centrifugal compressors based on ontology and Bayesian network fusion reasoning. Concurr Comput Pract Exp 37:e70278. https://doi.org/10.1002/cpe.70278.
Li S et al. (2025) Key technologies and applications of digital twin hydraulic engineering. Digit Twin 2:2486862. https://doi.org/10.1080/27525783.2025.2486862.
Xie Y, Zeng W, Qiu J (2023) Integrated water risk early warning framework of the semi-arid transitional zone based on the water environmental carrying capacity (WECC). J Arid Land 15(2):145–163. https://doi.org/10.1007/s40333-022-0083-z.
Zhao T et al. (2025) Leveraging immersive digital twins and AI-driven decision support systems for sustainable water reserves management: a conceptual framework. Sustainability 17:3754. https://doi.org/10.3390/su17083754.
Negi P et al. (2024) Specific soft computing strategies for the digitalization of infrastructure and its sustainability: a comprehensive analysis. Arch Comput Methods Eng 31(3):1341–1362. https://doi.org/10.1007/s11831-023-10018-x.
Liu H, Su H, Li H (2024) Study on digital twin technologies for watershed information modeling (WIM): a systematic literature review and bibliometric analysis. Arch Comput Methods Eng 31(1):263–289. https://doi.org/10.1007/s11831-023-09977-y.
Ling M et al. (2025) Framework for comprehensive risk assessment and factor diagnosis from the perspective of the water–energy–food–ecology–carbon complex system: a case study of the Yellow River ‘Ji’ Bay. Sustainability 17:9637. https://doi.org/10.3390/su17219637.
Ma L, Chen S, Wei D, Zhang Y, Guo Y (2024) A comprehensive hybrid deep learning approach for accurate status predicting of hydropower units. Appl Sci 14:9323. https://doi.org/10.3390/app14209323.
Zhang L, Hou Y, Deng K, Xin J (2025) Advancements toward a standard system for intelligent operation and maintenance of buildings and municipal facilities. Buildings 15:3965. https://doi.org/10.3390/buildings15213965.
Zhao J et al. (2025) Analysis of construction safety risk management for cold region concrete gravity dams based on fuzzy VIKOR-LEC. Buildings 15:1981. https://doi.org/10.3390/buildings15121981.
Zhang W, Meng H, Wang R, Wang P (2025) Intelligent optimized diagnosis for hydropower units based on CEEMDAN combined with RCMFDE and ISMA-CNN-GRU-attention. Water 17:2125. https://doi.org/10.3390/w17142125.
Liu G et al. (2025) Advances in artificial intelligence and digital twin for tunnel boring machines. Artif Intell Rev 58:257. https://doi.org/10.1007/s10462-025-11261-3.
Tian J, Zhang X, Zheng S, Liu Z, Zhan C (2024) Synergising an advanced optimisation technique with deep learning: a novel method in fault warning systems. Mathematics 12:1301. https://doi.org/10.3390/math12091301.
Zhou Y et al. (2025) IoT-driven intelligent curing of face slab concrete in rockfill dams based on integrated multi-source monitoring. Buildings 15:2344. https://doi.org/10.3390/buildings15132344.
Pi Y, Tan Y, Golmohammadi AM, Guo Y, Xiao Y, Chen Y (2023) A fault warning approach using an enhanced sand cat swarm optimization algorithm and a generalized neural network. Processes 11:2543. https://doi.org/10.3390/pr11092543.
Zhang S, Liu J, Li Z, Xiahou X, Li Q (2024) Analyzing critical factors influencing the quality management in smart construction site: a DEMATEL-ISM-MICMAC based approach. Buildings 14:2400. https://doi.org/10.3390/buildings14082400.
cathysiyu (2026) Mechanical-datasets/gearbox/gearset. GitHub. Available at: https://github.com/cathysiyu/Mechanical-datasets/tree/master/gearbox/gearset (Accessed 24 Feb 2026)
Shah J, Hu J, Edelenbosch O, van Vliet MTH (2024) GloHydroRes—a global dataset combining open-source hydropower plant and reservoir data. Zenodo. https://doi.org/10.5281/zenodo.14526360.
Liu B, Wang X, Zhang Z, Zhao Z, Wang X, Liu T (2025) Fault prediction of hydropower station based on CNN-LSTM-GAN with biased data. Energies 18:3772. https://doi.org/10.3390/en18143772.
Yang Z et al. (2025) Hybrid CNN-BiLSTM-MHSA model for accurate fault diagnosis of rotor motor bearings. Mathematics 13:334. https://doi.org/10.3390/math13030334.
Han K, Wang W, Guo J (2024) Research on a bearing fault diagnosis method based on a CNN-LSTM-GRU model. Machines 12:927. https://doi.org/10.3390/machines12120927.
He X (2026) Digital twin-enabled smart energy management for mega sports events. Distrib Gener Altern Energy J 41(2):245–270.
Liu Y, Bai H, Li W (2026) A distribution network operational situation perception technology based on graph convolutional neural network-enhanced digital twin model. Distrib Gener Altern Energy J 41(2):301–326.
