An Improved Particle Swarm Optimization with Multiple Strategies for PID Control System Design

Abstract

In this paper, an improved particle swarm optimization (PSO) with multiple subpopulations is developed for PID control system designs. The original single population needs to be divided into several subpopulations, and each subpopulation then tackles a corresponding performance index of the system. Under this proposed structure, several PID controllers can be simultaneously designed to meet different performance indexes when the algorithm is executed only one time. It is a great improvement because the general PSO algorithm with a single population can only deal with one performance index. To demonstrate the feasibility of the proposed scheme, a complicated chemical nonlinear process called the continuously stirred tank reactor (CSTR) is illustrated. Three different kinds of control operations are simulated including the step response control, set-point tracking control, and unstable equilibrium point control. For each control case five different performance indexes are assigned to guide the PID controller design combined with the nonlinear CSTR system. Simulation results will sufficiently confirm the superiority of the proposed algorithm.