ADAPTIVE POWER SYSTEM AUTOMATION FOR REAL-TIME FAULT DETECTION AND SELF-HEALING GRID OPERATION

Abstract

Smart automation systems are needed in modern power networks to identify electrical faults quickly, segment the affected network, and reconnect healthy loads with minimal disruption. This research involved designing and testing a power-system automation framework in order to support real-time power-system fault detection and self-healing operations. The IEEE 33-bus distribution system was simulated, and a quantitative, simulation-based experimental design method was applied. The 300 simulated fault-event observations included single line-to-ground, line-to-line, double line-to-ground, and three-phase fault conditions under various operating conditions. The framework combined real-time monitoring, adaptive fault detection, intelligent classification, automatic protection and network reconfiguration, and self-healing restoration. The overall fault-detection accuracy was 98.2%, and the average fault-detection time was reduced to 42ms. Fault-isolation accuracy was 98.7%, and the average fault-isolation time was reduced to 0.96s. The adaptive framework achieved an average restoration speed of 1.84 s (63.5% reduction), compared with conventional restoration of 5.04 s. The restored-load percentage reached 95.9%, with the power losses reduced by 27.8% post-restoration. The average voltage deviation dropped by 57.4%, while recovery efficiency rose to 91.4%. The results showed that the integrated adaptive automation enhanced fault detection, protection selectivity, network restoration and operational resilience. The proposed solution enabled the development of flexible, self-regulating power networks that can operate under disturbances and varying conditions.

Keywords: Adaptive Automation, Fault Detection, Real-Time Monitoring, Self-Healing Grid, Smart Grid, System Restoration