Chaotic sequence to improve particle swarm optimization

Abstract

Particle Swarm Optimization (PSO) is a population-based stochastic algorithm. It uses to optimize some continuous nonlinear functions by simulating movement of particles with social behaviours. It optimizes a problem by trying different values with particles’ movement randomly, with social behaviours that particles will move toward to the best optima at that moment. The particles do that iteratively until meeting the goal we given. According to the research of Particle Swarm Optimization, control variable is import in convergence of Particle Swarm Optimization. So, in this project, we propose to replace the use of random number to simulate the social behaviours of particles by chaotic sequences in order to improve the performance of PSO.