AI and Abductive Inference-Based Techniques for Power-System Abnormal Alarm Identification and On-Site Grid Operational Behavior Analysis
DOI:
https://doi.org/10.4108/ew.13844Keywords:
Abnormal alarm identification, event aggregation, Hidden Markov Model, root cause localization, on-site operational behavior analysis;, power systemAbstract
During the operation of power-system industrial control and monitoring platforms (e.g., SCADA/EMS and substation automation), strong coupling among components and multi-source heterogeneous data often lead to alarm flooding and complicate root cause identification. To address this, this paper proposes an intelligent abnormal-alarm identification and on-site operation behavior analysis method, combining artificial intelligence with an abductive inference framework. Under fixed parameters, the method first aggregates raw alarm streams by events to enhance structure and interpretability. Then, a diagonal-covariance Gaussian Hidden Markov Model (HMM) is trained with normal data, and a path-deviation metric ranks root cause candidates. Multi-source evidence chains—integrating temporal, network, and semantic features—further improve inference interpretability for grid operation and maintenance. Using annotated operation logs, four quantitative metrics (MTTA, MTTR, action rate, consistency) assess the link between model outputs and actual handling behaviors. Experiments on five test sets show the method achieves a 78% alarm compression rate and a 0.43 average silhouette coefficient. Top-1 and Top-3 root cause localization hit rates are 71.8% and 88.5%, with path score fluctuations under 0.05 nats. The average MTTA and MTTR are 186s and 792s, with an 84% action rate and 72% consistency. These results confirm the method’s effectiveness in mitigating alarm flooding, improving root cause localization, and supporting on-site decision-making in power-grid operational scenarios.
Downloads
References
[1] Yang T, Zhang J Y, Huang Z Q, et al. A survey on industrial control system security[J]. Journal of Computer Research and Development, 2022, 59(05): 1035-1053.
[2] Shang S Y, Li H J, Song C, et al. A survey of KPI anomaly detection based on machine learning in Internet service scenarios[J]. Journal of Computer Research and Development, 2025, 62(01): 207-231.
[3] Kargaran A H, Neshastegaran A, Izadi I, et al. Analytical derivation and comparison of alarm similarity measures[J]. IFAC-PapersOnLine, 2021, 54(3): 360-365.
[4] Li D, Cheng X. A first-out alarm detection method via association rule mining and correlation analysis[J]. Entropy, 2023, 26(1): 30.
[5] Khaleghy H, Izadi I. Detection of correlated alarms using graph embedding[C]//2021 7th International Conference on Signal Processing and Intelligent Systems (ICSPIS). IEEE, 2021: 1-7.
[6] Medianovskyi K, Pietarinen A V. On explainable AI and abductive inference[J]. Philosophies, 2022, 7(2): 35.
[7] Farahnakian F, Nicolas F, Farahnakian F, et al. A comprehensive study of clustering-based techniques for detecting abnormal vessel behavior[J]. Remote Sensing, 2023, 15(6): 1477.
[8] Panza M A, Pota M, Esposito M. Anomaly detection methods for industrial applications: A comparative study[J]. Electronics, 2023, 12(18): 3971.
[9] Rahaman M H, Alinezhad H S, Chen T. Real-time classification and early warning of industrial alarm floods using modified TF-IDF methods[J]. Control Engineering Practice, 2025, 164: 106485.
[10] Jadidi Z, Pal S, Hussain M, et al. Correlation-based anomaly detection in industrial control systems[J]. Sensors, 2023, 23(3): 1561.
[11] Wang J, Sun Q, Zhou C. Insider threat detection based on deep clustering of multi-source behavioral events[J]. Applied Sciences, 2023, 13(24): 13021.
[12] Cui J, Zhang G, Chen Z, et al. Multi-homed abnormal behavior detection algorithm based on fuzzy particle swarm cluster in user and entity behavior analytics[J]. Scientific Reports, 2022, 12(1): 22349.
[13] Manca G, Dix M, Fay A. Clustering of similar historical alarm subsequences using alarm series and characteristic coactivations[C]//Proc. 32nd Int. Workshop on Principles of Diagnosis. 2021: 2-9.
[14] Liu Y, Garg S, Nie J, et al. Deep anomaly detection for time-series data in industrial IoT: A communication-efficient on-device federated learning approach[J]. IEEE Internet of Things Journal, 2020, 8(8): 6348-6358.
[15] Zhang S, Liu M J. Industrial image anomaly detection based on deviation-reduced coupled hypersphere[J]. Progress in Laser and Optoelectronics, 2025, 62(12): 176-183.
[16] Birihanu E, Lendák I. Explainable correlation-based anomaly detection for industrial control systems[J]. Frontiers in Artificial Intelligence, 2025, 7: 1508821.
[17] Abdelaty M, Doriguzzi-Corin R, Siracusa D. DAICS: A deep learning solution for anomaly detection in industrial control systems[J]. IEEE Transactions on Emerging Topics in Computing, 2021, 10(2): 1117-1129.
[18] Song X, Liu Q, Dong M, et al. Chemical process alarm root cause diagnosis method based on the combination of data-knowledge-driven method and time retrospective reasoning[J]. ACS omega, 2022, 7(24): 20886-20905.
[19] Rao H R M, Zhou B, Brown K, et al. Alarm correlation analysis with applications to industrial alarm management[J]. Control Engineering Practice, 2024, 143: 105812.
[20] Vogel-Heuser B, Fay A, Rupprecht B, et al. Exploring challenges of alarm root-cause analysis across varying production process types[J]. at-Automatisierungstechnik, 2024, 72(4): 369-386.
[21] Venkidasalapathy J A, Kravaris C. Hidden Markov model based fault diagnoser using binary alarm signals with an analysis on distinguishability[J]. Computers & Chemical Engineering, 2022, 160: 107689.
[22] Alinezhad H S, Roohi M H, Chen T. A review of alarm root cause analysis in process industries: Common methods, recent research status and challenges[J]. Chemical Engineering Research and Design, 2022, 188: 846-860.
[23] Rahaman M H, Alinezhad H S, Chen T. Identification of Most Critical Alarms for Alarm Flood Reduction[J]. IFAC-PapersOnLine, 2024, 58(14): 835-840.
[24] Ariamuthu Venkidasalapathy J, Kravaris C. Hidden Markov model based approach for diagnosing cause of alarm signals[J]. AIChE Journal, 2021, 67(10): e17297.
[25] Li Y, Liu Y, Zhang J, et al. Automated analysis and assignment of maintenance work orders using natural language processing[J]. Automation in Construction, 2024, 165: 105501.
[26] Sundaram S, Zeid A. Technical language processing for Prognostics and Health Management: applying text similarity and topic modeling to maintenance work orders[J]. Journal of Intelligent Manufacturing, 2025, 36(3): 1637-1657.
[27] Ahn J, Park J Y, Lee S S, et al. SafeFac: Video-based smart safety monitoring for preventing industrial work accidents[J]. Expert Systems with Applications, 2023, 215: 119397.
[28] Pazho A D, Neff C, Noghre G A, et al. Ancilia: Scalable intelligent video surveillance for the artificial intelligence of things[J]. IEEE Internet of Things Journal. 2023, 10(17): 14940-14951.
[29] Shaheen B W, Németh I. Integration of maintenance management system functions with industry 4.0 technologies and features—A review[J]. Processes, 2022, 10(11): 2173.
[30] Shin H K, Lee W, Yun J H, et al. Two ICS security datasets and anomaly detection contest on the HIL-based augmented ICS testbed[C]//Proceedings of the 14th USENIX Workshop on Cyber Security Experimentation and Test (CSET ’21). ACM, 2021: 36-40.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Kan Shi, Jiao Yang, Yongmei Chen, Shengyan Ye, Xuanli Xia, Jingrui Shang, Mian Li, Jianwei Yang, Tianguo Yang, Shuangwu Li
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open-access article distributed under the terms of the Creative Commons Attribution CC BY 4.0 license, which permits unlimited use, distribution, and reproduction in any medium so long as the original work is properly cited.
