Optimizing Firefly Algorithm for Directional Overcurrent Relay Coordination: A case study on the Impact of Parameter Settings

Abstract

The Firefly Algorithm is a nature-inspired optimization algorithm that has been shown to be effective for solving a variety of problems. In this paper, we study the Firefly Algorithm for the IEEE 3-bus network coordination problem. We examine the impact of key parameters on the algorithm's performance, including generation number, population size, absorption coefficient, and randomization parameter. The results showed that increasing the generation number improves the quality of the solution, but the benefits decrease at a certain point. Feasibility also improves with higher generations, but a balance between solution quality and feasibility becomes apparent at very high generations. Objective function evaluations and computation time increase linearly with the number of generations. Larger population sizes yield better solution quality and feasibility, but a balance is observed at very high population sizes. The randomization parameter has a modest influence on the results, but extreme values can impact solution quality, feasibility, and computation time. The absorption coefficient, on the other hand, has a significant impact on convergence and quality. The study offers guidance for parameter optimization and adaptive techniques.