A modified particle swarm optimization for lower order model formulation of linear time invariant continuous systems

This paper proposes a new version of the classical particle swarm optimization (PSO), namely, MPSO used to formulate the lower order model for the linear time invariant continuous systems. 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 particle to search the target very effectively. In order to minimize the integral squared error of the lower order model MPSO is proposed and results are shown in the form of unit step response curves and are compared with the response of the original higher order model and with the other model formulation methods. The proposed method is illustrated through numerical example from literature.