Predicting Distribution Transformer Failures Through Efficiency-Loss Trend Analysis on a Live Web Monitoring Platform

Abstract

Power transformers are critical components in electrical power distribution systems, and their reliable operation is essential for maintaining grid stability. Conventional transformer monitoring methods primarily depend on periodic manual inspections, which may lead to delayed fault detection and increased maintenance costs. This paper presents a low-cost real-time transformer monitoring and predictive fault detection system based on the ESP32 microcontroller and Internet of Things (IoT) technology. The proposed system continuously monitors key transformer parameters, including voltage, current, temperature, and power factor, using dedicated sensors. The collected data is processed locally and transmitted to a Firebase cloud database for remote access and storage. A web-based dashboard developed using HTML, CSS, and JavaScript provides real-time visualization, historical data analysis, and status alerts categorized as Normal, Warning, and Critical. In addition, a linear trend forecasting algorithm is implemented to predict future parameter variations, enabling proactive maintenance and reducing the risk of unexpected failures. Experimental results demonstrate the accuracy, reliability, and scalability of the proposed system under varying operating conditions, making it suitable for smart grid and industrial applications.