πŸ“ž +91-7667918914 | βœ‰οΈ ijireeice@gmail.com
International Journal of Innovative Research in Electrical, Electronics, Instrumentation and Control Engineering
International Journal of Innovative Research in Electrical, Electronics, Instrumentation and Control Engineering A monthly Peer-reviewed & Refereed journal
ISSN Online 2321-2004ISSN Print 2321-5526Since 2013
IJIREEICE meets the suggestive parameters outlined in the latest University Grants Commission (UGC) for peer-reviewed journals, ensuring high standards of research integrity, publication ethics, and academic excellence.
← Back to VOLUME 14, ISSUE 10, OCTOBER 2026

Smart Transformer Health Monitoring System Using Multi-Parameter Sensing and Early Anomaly Detection

Agnivo Chowdhury, Saumya Pankaj Kumar, Jatin Satapathy, Ayan Saikia, Hareesh Dash, Karthikeya Bhandari, A. Thamilamaran

πŸ‘ 42 viewsπŸ“₯ 21 downloads
Share: 𝕏 f in ✈ βœ‰
Abstract: This paper presents a low-cost, non-invasive health monitoring system for transformers. Conventional protection relies on static, per-parameter alarm limits that react only after a quantity is already dangerously high, while commercial diagnostic equipment is expensive. The prototype is built around an Arduino Uno that measures voltage, current and surface temperature using a 0–25 V voltage-sensor module, ACS712 Hall-effect current sensors and an analog temperature sensor, and displays the readings on a 16Γ—2 LCD. Beyond display, the measurements are interpreted jointly: a baseline operating signature relating temperature, vibration and voltage to load current is established, and the normalized deviations from it are fused into one anomaly score and a 0–100 health index. A simulation framework (first- order thermal model, IΒ² vibration model, sensor noise and ADC quantization) is used to compare this approach with conventional threshold alarms. Over 300 Monte-Carlo runs per scenario, the fusion method confirmed a developing cooling fault 19.5 min after onset against 39.5 min for static thresholds, detected a developing mechanical fault that the thresholds missed in 100% of runs, and raised no false anomaly alarm during a load-explained overload. The work provides the foundation for experimental validation, machine-learning and edge-based extensions.

Keywords: Transformer condition monitoring, sensor fusion, anomaly detection, Arduino, ACS712, health index, measurement and instrumentation.

How to Cite:

[1] Agnivo Chowdhury, Saumya Pankaj Kumar, Jatin Satapathy, Ayan Saikia, Hareesh Dash, Karthikeya Bhandari, A. Thamilamaran, β€œSmart Transformer Health Monitoring System Using Multi-Parameter Sensing and Early Anomaly Detection,” International Journal of Innovative Research in Electrical, Electronics, Instrumentation and Control Engineering (IJIREEICE)

Creative Commons License This work is licensed under a Creative Commons Attribution 4.0 International License.