Versatility of Particle Swarm Optimization in identification for continuous-time systems

This paper shows the versatility of the Particle Swarm Optimization, which attracts a lot of attention recently in the evolutionary computation area due to its empirical evidence of its superiority, in the area of continuous-time system identification. First, a method to identify (possibly nonlinear) continuous-time systems is shown, which uses the Particle Swarm Optimization to minimize the mean square prediction error. Second, its effectiveness is demonstrated through various numerical examples which include time-delay systems, closed loop ones and nonlinear ones.