Controlled Comparison of Machine Learning Models for Fault Classification and Localization in Power System Protection
arXiv:2510.00831v2 Announce Type: replace-cross Abstract: The increasing complexity of modern power systems, driven by the integration of inverter-based and distributed energy resources, challenges the reliability of conventional protection schemes and motivates the use of machine learning for protection tasks. However, published results are often difficult to compare because datasets, sensing assumptions, and decision horizons vary across studies. This paper presents a controlled comparison of machine learning models for fault classification (FC) and fault localization (FL) under identical se
The increasing integration of inverter-based and distributed energy resources is creating new complexities in power systems, necessitating advanced protection mechanisms beyond conventional methods.
This research highlights the critical role of machine learning in ensuring grid stability and reliability, which becomes paramount as energy grids modernize and integrate more intermittent-renewable sources.
The paper provides a standardized methodology for comparing ML models in power system protection, which should accelerate the adoption of effective AI solutions for grid management.
- · Grid operators
- · Machine learning developers
- · Renewable energy sector
- · Power system engineers
- · Manufacturers of conventional protection schemes
- · Grids reliant on outdated infrastructure
Improved fault classification and localization in power systems will lead to reduced outages and more efficient grid operation.
Enhanced grid reliability through AI will support faster and broader adoption of distributed and renewable energy sources, increasing grid resilience.
The standardized comparison framework could foster a competitive landscape for AI solutions in critical infrastructure, driving innovation and cost reduction.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG