Enhanced Fault Diagnosis and Optimal Protection Coordination in Radial Distribution Systems Using Adaptive Differential Evolution

Abstract

The reliability of radial distribution systems is critically affected by the frequency and severity of electrical faults, which often result in voltage instability, supply interruptions, and equipment degradation. Effective fault diagnosis and protection coordination therefore remain essential components of modern distribution network operation. This study presents an enhanced Adaptive Differential Evolution -based framework for fault diagnosis and optimal protection coordination using the IEEE 33-bus radial distribution system as a case study. A mathematical model that integrates post-disturbance voltage and current signatures with network impedance characteristics was formulated to estimate fault distance and classify fault severity. The Forward and Backward Sweep method was applied to compute pre- and post-fault system states under multiple symmetrical and asymmetrical fault scenarios. The ADE algorithm was then employed to optimize protection indices, minimizing both fault location error and cumulative clearing time while ensuring selective relay coordination. Simulation results demonstrate that ADE significantly improves fault diagnosis accuracy compared to Genetic Algorithm (GA) and Political Optimization (PO). For the IEEE 33-bus system, ADE accurately identified critical faulted buses (10, 27, 26, 28, and 6) while achieving the lowest location error and the shortest fault clearing interval of 85–92 ms. Post-fault stability analysis further revealed that ADE maintained higher voltage recovery (0.8071 p.u) and lower short-circuit current magnitudes (14.1651 p.u) relative to GA and PO. These improvements directly enhance relay selectivity, reduce miscoordination risk, and minimize stress on feeder equipment. The findings confirm that the ADE-based approach offers a robust, fast, and intelligent protection strategy for radial distribution networks, and establishes its potential for integration into future smart-grid automation and self-healing protection schemes

Keywords

References