Chaotic Particle Swarm Optimization for Parameter Estimation in Nonlinear Dynamic Systems

Abstract

The estimation of parameters for chaotic systems is an important area of study in nonlinear dynamics, and it has recently attracted attention from a number of other fields. A multi-dimensional optimization problem is the simplest way to characterize the current scenario. Chaotic particle swarm optimization (CPSO) is a novel technique in evolutionary computation that has attracted much attention and is extensively employed due to its simplicity, ease of implementation, and rapid convergence. To our knowledge, no published study exists on applying CPSO for parameter estimation in chaotic systems. This research employs a CPSO methodology to precisely ascertain the Lorenz system's parameters. The effectiveness and durability of CPSO are showcased through numerical simulations and comparisons. In addition, the study also investigates the impact of population size on optimization performance.