Particle Swarm Optimization with Adaptive Inertia Weight using Fuzzy Logic for Large-Scale Problems

Abstract

In this paper an alternative approach is proposed to improve the convergence of Particle Swarm Optimization (PSO) algorithm by adapting the inertial weight parameter with a fuzzy logic system to solve large-scale optimization problems. The PSO algorithm is a population-based metaheuristic inspired by the social behavior of birds, and it has been applied to numerous optimization problems successfully. However, one of its main disadvantages is the decaying performance when applied to complex and large-scale problems. The proposed algorithm uses the fuzzy system to dynamically calculate a value of the Inertia Weight parameter during the search process to find better solutions. After carrying out experiments on a well-known benchmark for large-scale optimization, the proposed approach provides a competitive performance.