Generalized predictive control based on particle swarm optimization for linear/nonlinear process with constraints

Abstract

This paper presents an intelligent generalized predictive controller (GPC) based on particle swarm optimization (PSO) for linear or nonlinear process with constraints. We propose several constraints for the plants from the engineering point of view and the cost function is also simplified. No complicated mathematics is used which originated from the characteristics of PSO. This method is easy to be used to control the plants with linear or/and nonlinear constraints. Numerical simulations are used to show the performance of this control technique for linear and nonlinear processes, respectively. In the first simulation, the control signal is computed based on an adaptive linear model. In the second simulation, the proposed method is based on a fixed neural network model for a nonlinear plant. Both of them show that the proposed control scheme can guarantee a good control performance.