Voltage Stability Enhancement in North West Ethiopia’s Power Grid Using Contingency Analysis and PSO-Based Load Shedding

This study presents an integrated methodology to enhance voltage stability and operational security in the North West Ethiopian power grid by addressing system vulnerabilities and optimizing load-shedding strategies. A hybrid simulation-optimization framework is developed, combining DIgSILENT PowerFactory-based contingency analysis with Particle Swarm Optimization (PSO)-driven intelligent load shedding in MATLAB. The modeled power network comprises 2 generators, 15 buses, 15 transmission lines, and 3 external grids. Vulnerability assessment is conducted using Voltage and Overload Performance Indices (PIV and PIP) to rank critical elements under various contingencies. The analysis identifies the BDII-230 to GOII transmission line as the most vulnerable, where its outage causes voltage drops to 0.908 p.u. at Gondar and 0.923 p.u. at Metema, both falling below acceptable operational limits. Under increased load conditions (50% and 75% load factor), the Gondar bus experiences severe voltage deterioration, dropping to 0.8879 p.u., which significantly stresses system stability. Through PSO-based load shedding, the voltage at Gondar is restored to 0.9538 p.u., demonstrating a 7.4% improvement, while excessive real power (e.g., at BDII230) is reduced from 1.3235 to 0.8958 p.u. under 50% load increase, effectively stabilizing the system. The proposed PSO framework outperforms conventional load-shedding mechanisms by factoring in bus sensitivity and minimizing overall power shed while maximizing voltage recovery. The results provide actionable insights for grid operators and planners, establishing a reliable and scalable method for real-time contingency response in developing power networks.