Multi-Objective Particle Swarm for Optimal Load Shedding Remedy Strategies of Power System

Abstract

Load shedding is an emergency strategy that mitigates substantial mismatch between the generation and the loads. It generally sustains system stability during/after severe disturbances. This article proposes robust, simple, and innovative Under-Frequency Load Shedding (UFLS) technique based on Multi-Objective Particle Swarm Optimization (MOPSO). MOPSO has the objectives of: minimizing the amount of the dropped load and maximizing the lowest swing frequency. The functionality and feasibility of the proposed MOPSO are corroborated via comprehensive comparison with traditional, adaptive, Single Objective PSO (SOPSO), and Genetic Algorithm (GA) UFLS schemes. IEEE 9-bus and 39-bus systems are used cases for examining the reliability, applicability, and viability of the proposed MOPSO. Different scenarios as: outage of single, multiple generating plants and load increase are applied in the test systems, while load shedding is executed via MOPSO, SOPOS, GA, traditional, and adaptive UFLS approaches. The DigSilent power factor software is used for simulating the test systems while subjected to the different disturbance levels. MATLAB is used for coding SOPSO, MOPSO, adaptive, and GA algorithms. The results show that MOPSO-based UFLS relay produces higher lowest swing frequency and lower amount of dropped load than SOPSO and GA. MOPSO requires less computation requirements than GA approach.