Evolutionary computing techniques for model order formulation of linear time invariant discrete systems

The authors present an algorithm for order formulation of linear time invariant discrete systems using a new version of evolutionary computing technique namely, modified particle swarm optimization (MPSO). In the modified PSO, the movement of a particle is governed by three behaviors namely, inertia, cognitive and social. The cognitive behavior helps the particle to remember its previous visited best position. This paper proposes to split the cognitive behavior into two sections. This modification helps the model order formulation with minimum integral squared error. The results are shown in the form of unit step and unit impulse time response curves and compared with Genetic algorithm and general PSO based optimization techniques. The proposed method is illustrated through numerical example from literature.