Advanced regression modeling for correlating transformer oil electrical properties with thermal aging trends

Abdi, Sifeddine, Besseri, Boubakar Achraf, Haddad, Manu ORCID: https://orcid.org/0000-0003-4153-6146, Harid, Noureddine and Boubakeur, Ahmed 2025. Advanced regression modeling for correlating transformer oil electrical properties with thermal aging trends. Presented at: UPEC 2024, Cardiff, Wales, 02-06 September 2024. Proceedings 59th International Universities Power Engineering Conference (UPEC). IEEE, 10.1109/upec61344.2024.10892438

Full text not available from this repository.

Abstract

In this study, we investigate the impact of thermal aging on crucial electrical properties of mineral oil widely used in transformers, namely the breakdown voltage, dielectric dissipation factor, and resistivity. Our research stands out for its innovative use of an exponential regression model to compre-hensively analyze these properties. Through a meticulous experi-mental approach, we subject transformer oil to a rigorous aging protocol spanning 5000 hours at temperatures of 80°C, 100°C, 120°C, and 140°C. At regular 500-hour intervals, we meticulously evaluate the breakdown voltage, dielectric dissipation factor, and resistivity to track any changes induced by aging. Leveraging advanced regression analysis techniques, particularly employing the exponential model, we accurately characterize the evolving electrical properties of the oil samples. Our findings reveal distinct alterations in breakdown voltage, dielectric dissipation factor, and resistivity following thermal aging, with more signifi-cant degradation observed at higher temperatures. Moreover, our regression analysis closely aligns with experimental results across all samples and characteristics studied, boasting high correlation coefficients. These results affirm the reliability of our model in predicting the electrical properties of aged transformer oil accurately, providing valuable insights for assessing transformer performance.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Published Online
Status:	In Press
Schools:	Schools > Engineering
Publisher:	IEEE
ISBN:	979-8-3503-7973-0
Last Modified:	10 Mar 2025 15:00
URI:	https://orca.cardiff.ac.uk/id/eprint/176777

Actions (repository staff only)

Edit Item