Application of Artificial Intelligence in Pre-defect Identification and Hidden Danger Early Warning for Distribution Substation Fire and Flood Risks

DOI: https://doi.org/10.62517/jes.202602239

Author(s)

Sikun Yuan*, Bingbing Shang, Jianing Liang

Affiliation(s)

Henan Jiuyu Enpai Electric Power Technology Co., Ltd., Zhengzhou, Henan, China *Corresponding Author

Abstract

In our modern day and age, how well the power distribution equipment works is connected with what we do every single day and even with just living. The endpoint of the power system, it is very important to make distribution stations run safely and healthily. Natural hazards like fires and floods that can directly cause dangers on power grids. Traditional prediction and disaster avoidance uses manual judgments which leads us to delays and not good early warning. AI tech developing speed brings up an option that could help spot substation danger signs before anything terrible happens with disaster protection. It systematically reviews the use of AI tech in finding substation equipment flaws and warning about fire and water risks in advance. Risk modeling and anomaly detection and decisions etc. The paper focuses on research achievements of such as mult-sourced data gathering, computer vision, time sequence forecast, multitype information fusion, knowledge graph. And analyze through typical example from home and abroad of practical effects brought by AI improvement on early warning correctness, decreasing answer time, minimizing false alarm frequency. Also, talk about current difficulties and restrictions for technology. Then look ahead towards where we hope it will go so we might have something to offer in terms of theory about how substations could grow healthily and stably.

Keywords

Distribution Substation; Fire Early Warning; Flood Risk and Hazard Early Warning; Multimodal Fusion; Artificial Intelligence

References

[1] Huang J, He R, Zhang X, et al. Adaptive Flood Prevention System for Substations Using Integrated Fuzzy Bayesian Network and MILP Optimization. International Journal of Reliability, Quality and Safety Engineering, 2025. [2] Mahajan D, Sharma S, Saini K. Rising with Intelligence: A Systematic Review of AI Transforming Smart Flood Management. Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA), 2026, 40(8): 1-31. [3] Xing D, Shi Q P, Yuan H, et al. Intelligent inspection and defect identification method for substation equipment based on deep learning. Electrical Engineering Technology, 2025(22): 167-169. [4] Chen Z, Wen J M, Li Y Z, et al. Research on prevention and control of active fire early warning system in intelligent substation. Science & Technology Information, 2025, 23(15): 186-188. DOI: 10.16661/j.cnki.1672-3791.2411-5042-0813. [5] Ning X, Ding T L, Zhu H W. Research on Low-Voltage Arc Fault Based on CNN-Transformer Parallel Neural Network. Sensors, 2024, 24(20). [6] Guo X, Ding L, Ji J. Method for Fire Risk Assessment of Urban Power Substation Considering Data Uncertainty. Fire Technology, 2025, 61(7): 5291-5317. [7] Shalaby A A, Abdeltawab H, Mohamed Y A R I. Safe deep reinforcement learning for resilient self-proactive distribution grids against wildfires. IEEE Transactions on Smart Grid, 2025. [8] Liasi S, Ghiasi N S, Hadidi R. Enhancing Power Grid Resilience Against Flooding: A Practical Optimization Approach for Substation Protection. IEEE Transactions on Industry Applications, 2025. [9] Le C, McDonald K, Qu Z. Predicting Extreme Weather-Induced Outages in Distribution Networks Using Graph-Based Neural Network Models//2025 IEEE Conference on Artificial Intelligence (CAI). IEEE, 2025: 1443-1448. [10] Baghkarvasef M, Bidram A, Farrokhabadi M, et al. Leveraging Artificial Intelligence for Enhancing Power Grid Resilience to Extreme Weather Events: Applications and Challenges. IEEE Energy Sustainability Magazine, 2025, 1(3): 85-97. [11] Moloi K, Addo K, Mnkandla E. Spatio-Temporal Artificial Intelligence for Multi-Hazard-Aware Renewable Energy Site Selection Using Integrated Geospatial and Climate Data. Processes, 2025, 13(11): 3728. [12] Hallioui A, Pedroni N. Enhancing the Resilience and Sustainability of Integrated Energy Systems Exposed to Extreme Natural Hazards by Means of Artificial Intelligence, Advanced Simulation, and Optimization Methods, Within an Integrative Systems Framework: A Critical Review of Literature. Energies, 2026, 19(4): 957. [13] Esparza M, Li B, Ma J, et al. AI meets natural hazard risk: A nationwide vulnerability assessment of data centers to natural hazards and power outages. International Journal of Disaster Risk Reduction, 2025, 126: 105583. [14] Cao H, An H, Bai X, et al. Research and Design of an Artificial Intelligence-Based Inspection System for Substation and Distribution Rooms//2024 4th International Symposium on Artificial Intelligence and Intelligent Manufacturing (AIIM). IEEE, 2024: 367-370. [15] Liu H, Su G, Huang M, et al. Design and application of an automatic inspection system for smart distribution station based on cloud-edge collaboration//International Conference on Artificial Intelligence and Intelligent Information Processing (AIIIP 2022). SPIE, 2022, 12456: 13-18. [16] Fu B, Yuan B H. Integration of embedded systems and artificial intelligence in electrical automation for smart grid management. International Journal of Emerging Electric Power Systems, 2026, 27(1): 13-24. [17] Liu G T, Zhu Y X, Chen S J, et al. MWSN: A Multi-Head Wide Kernel Deep Convolutional Siamese Network for residual current fault noise-resistant diagnosis. [18] Zhao Z, Yang J, Su X, et al. Applications of AI for the Optimal Operations of Power Systems Under Extreme Weather Events: A Task-Driven and Methodological Review. Energies, 2026, 19(2): 506. [19] Maralapalle V, Iyer R, Kumavat H R. Leveraging artificial intelligence for climate-resilient infrastructure management//Climate Change Effects on Infrastructure. Woodhead Publishing, 2026: 213-236.