Risk Surveillance for Backup Power Systems Using Rule-Based Fuzzy Inference and Dependency Graphs
Article Sidebar
Main Article Content
Abstract
Continuous power supply is critical for factories in Thailand’s Andaman region because even short interruptions can stop production and affect business continuity. Conventional UPS monitoring with fixed thresholds often creates too many alarms or alerts too late during critical events. This paper presents a fuzzy–graph framework with two main parts: (1) a linear overlap fuzzy inference system (LO-FIS) that uses threshold mapping instead of defuzzification to produce binary outputs (0 = normal, 1 = critical), and (2) a dependency graph that shows how risk can propagate from the UPS to other devices. The framework is implemented on a raspberry Pi and connected to a Syndome HE-RT-1-3K UPS. We inject controlled faults at different levels and record data continuously for 72 hours, covering a wide range of loads (4,321 samples). The results show that our method achieves precision 99.32%, recall 97.14%, and F1-score 98.07%, with only one false alarm, and provides an average lead time of about 30 seconds before actual outages. These outcomes indicate improved reliability and resilience for industrial backup power systems.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
- The original content that appears in this journal is the responsibility of the author excluding any typographical errors.
- The copyright of manuscripts that published in PKRU SciTech Journal is owned by PKRU SciTech Journal.
References
(1) Decker, L., Leite, D., Giommi, L., & Bonacorsi, D. (2020). Real-time anomaly detection in data centers for log-based predictive maintenance using an evolving fuzzy-rule-based approach. arXiv preprint, arXiv:2004.13527.
(2) Zhang, H., Li, Z., & Ren, Z. (2020). Data-driven construction of data center graph of things for anomaly detection. arXiv preprint, arXiv:2004.12540.
(3) Shrestha, B. R., Tamrakar, U., Hansen, T. M., Bhattarai, B. P., James, S., & Tonkoski, R. (2018). Efficiency and reliability analyses of AC and 380 V DC distribution in data centers. IEEE Access, 6, 63305–63315.
(4) Zakharov, A. V., Gusev, O. Y., & Cho, G. C. (2018). Reliability assessment of data centers power system. Proceedings of the 2018 International Conference on Information Technologies (INFORINO), 1–4.
(5) O’Halloran, B. M., Papakonstantinou, N., Giammarco, K., & Van Bossuyt, D. L. (2017). A graph theory approach to functional failure propagation in early complex cyber-physical systems. INCOSE International Symposium, 27(1), 164–178.
(6) Fliss, I., & Tagina, M. (2012). A novel fault detection approach combining adaptive thresholding and fuzzy reasoning. arXiv preprint, arXiv:1203.5454.
(7) Xu, J., Zhao, M., Fortes, J. A. B., Carpenter, R., & Yousif, M. (2007). On the use of fuzzy modeling in virtualized data center management. Proceedings of the 4th IEEE International Conference on Autonomic Computing (ICAC), 25–34.
(8) Fu, Y., Zhou, C., Li, J., & Chen, L. (2025). Long-term evolutionary patterns matter: Self-supervised anomaly detection on dynamic graphs. Knowledge-Based Systems, 311, 113049.
(9) Wagner, S. M., Bode, C., & Koziol, P. (2009). Supplier default dependencies: Empirical evidence from the automotive industry. European Journal of Operational Research, 199(1), 150–161.
(10) Prasath, S. C., Darwin, N., Ramkumar, R. S., Nithishkumar, S., & Somasundharam, P. L. (2023). IoT-powered UPS battery monitoring: Ensuring high availability and reliability for critical systems. E3S Web of Conferences, 399, 04007.
(11) Rao, Y., Chen, X., & Wang, Y. (2024). Fuzzy-coalition graphs: A framework for understanding multi-agent cooperation in uncertain environments. Mathematics, 12(22), 3614.
