POWERGUARD AI: LEARNING-BASED PROTECTION COORDINATION FOR SMART GRIDS

Abstract

The increasing penetration of renewable energy sources and the dynamic operating conditions of modern smart grids have significantly complicated protection relay coordination. Conventional relay coordination methods, which rely on fixed settings and offline analysis, are often unable to respond effectively to real-time changes in network configuration and fault characteristics. This paper presents a learningbased relay coordination strategy that employs data-driven techniques to optimize relay settings adaptively. Machine learning models are trained using historical and real-time system data to identify optimal coordination parameters under varying operating scenarios. The proposed approach enables faster fault isolation, improved selectivity, and enhanced system reliability. Simulation results demonstrate that the learningbased strategy outperforms traditional coordination methods in terms of reduced fault clearing time and improved protection accuracy. The findings highlight the potential of intelligent relay coordination to support secure and resilient smart grid operation.