Ashraf, Syed Abdullah, Ashraf, Syed Mohammad and Abdullah, Syed Ahmad (2026) Comprehensive Fault Analysis in Transmission Lines Using Artificial Intelligence. In: 2025 IEEE DELCON - International Conference on Recent Smart Technologies in Engineering for Sustainable Development, 31 Oct - 02 Nov 2025, New Delhi, India.
Comprehensive_Fault_Analysis_in_Transmission_Lines_Using_Artificial_Intelligence.pdf - Published Version
Restricted to Repository staff only
Download (792kB) | Request a copy
Abstract
Reliable fault detection, classification, and localization are critical for ensuring the stability and protection of high-voltage power transmission systems. This paper presents a unified multilabel framework that addresses these tasks simultaneously, providing a practical alternative to approaches that treat them independently. Using stratified evaluation to ensure balanced performance across fault types, the study demonstrates near-perfect accuracy and F1-scores in detecting and classifying faults, with equally strong performance in identifying faulted lines. Fault position localization, while more challenging due to overlapping feature distributions, achieves promising accuracy and highlights areas for further improvement. The proposed framework offers a scalable and efficient solution suitable for real-time deployment, contributing to enhanced grid reliability, reduced downtime, and improved operational efficiency in modern power systems.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Uncontrolled Keywords: | Fault analysis | fault detection | machine learning | artificial intelligence | transmission line |
| Subjects: | Social Sciences and humanities > Business, Management and Accounting > Management Information Systems |
| Divisions: | Jindal Global Business School |
| Depositing User: | Mr. Arjun Dinesh |
| Date Deposited: | 29 Mar 2026 13:33 |
| Last Modified: | 29 Mar 2026 13:33 |
| Official URL: | https://doi.org/10.1109/DELCON68055.2025.11400026 |
| URI: | https://pure.jgu.edu.in/id/eprint/11078 |
Downloads
Downloads per month over past year
Dimensions
Dimensions