Parameter identification of Box-Jenkins systems based on the particle swarm optimization

This paper considers the parameter estimation of Box-Jenkins systems. The particle swarm optimization method is adopted to identify the parameters of the Box-Jenkins systems. In order to improve the convergence speed and identification accuracy, some strategies have been applied to the basic particle swarm optimization algorithm, and the synchronous changing learning factor particle swarm optimization identification algorithm, the linearly decreasing weight particle swarm identification algorithm and the improved particle swarm optimization identification algorithm are presented respectively. The simulation results indicate that all the three algorithms can identify the Box-Jenkins systems effectively.