Research Projects

High-voltage power transformers are essential and costly apparatus, expected to operate continuously to maintain a stable power supply. Their reliable performance is critical, as they often encounter lightning and switching overvoltages/overcurrents due to electrical disturbances and also short-circuits, which stress the transformer insulation and windings. Cumulative effect of this stress can lead to insulation degradation or winding deformation, initially causing subtle and undetectable damage and that, if left unmonitored, may evolve into more severe faults, resulting in outages and costly repairs. On the other hand, India is a power-hungry country, where power demand is rising due to urbanization, industrial growth, and increasing focus on electric vehicles. With rising power demand and the integration of renewable energy sources, power fluctuations on transformers are expected to grow. Given these conditions, an improved diagnostic tool would empower utility engineers with an effective means to monitor and maintain transformers, minimizing downtime and prevent costly failures. While traditional diagnostics such as gas-in-oil analysis and impedance measurements provide some clues, they often fall short of detecting early-stage or minor faults. Frequency Response Analysis (FRA) – a non-intrusive and highly sensitive technique has emerged as a preferred monitoring method to detect such winding related issues early on. However, despite their sensitivity, current FRA technique primarily serves as a monitoring tool rather than a diagnostic method, as they can certainly detect even the minutest of winding changes but cannot as yet succeed in identifying the fault type, extent of severity or location. The objective of this project is to enhance FRA’s diagnostic capabilities and enable fault localization. This research aims to develop a novel method that can determine the location of the fault. This shall be achieved by analyzing the characteristics and salient features of different frequency response functions under various fault scenarios. The project will employ symbolic computation techniques for circuit analysis, advanced data analysis, and experimental validation to assess the effectiveness of different FRA responses. The study also seeks to understand inter-phase coupling in multi-phase transformers, which will aid in improving FRA’s ability to localize faults within a 3-ph transformer. By successfully developing and implementing this method, the project promises significant progress toward improving the diagnostic capabilities of FRA. This will not only prolong the lifespan and reliability of the transformer, but also contribute to a consistent power supply throughout the grid, thereby enhancing the overall power supply reliability. In summary, this project aims to enhance the capability of the FRA as a diagnostic tool that can localize faults in high-voltage transformers, thus reducing unexpected transformer failures.