System Identification of a DC Motor Using Different Variants of Particle Swarm Optimization Technique

System identification is a ubiquitous necessity for successful applications in various fields. The area of system identification can be characterized by a small number of leading principles, e.g. to look for sustainable descriptions by proper decisions in the triangle of model complexity, information contents in the data, and effective validation. Particle Swarm Optimization (PSO) is a stochastic, population-based optimization algorithm and many variants of PSO have been developed since, including constrained, multi objective, and discrete or combinatorial versions and applications have been developed using PSO in many fields. The basic PSO algorithm implicitly utilizes a fully connected neighborhood topology. However, local neighborhood models have also been proposed for PSO long ago, where each particle has access to the information corresponding to its immediate neighbors, according to a certain swarm topology. In this local neighborhood model of PSO, particles have information only of their own and their nearest neighbors' bests, rather than that of the entire population of the swarm. In the present work basic PSO method and two of its local neighborhood variants are utilized for determining the optimal parameters of a dc motor. The result obtain from the simulation study demonstrate the usefulness of the proposed methodology.