Performance investigations of APSO tuned linear and nonlinear PID controllers for a nonlinear dynamical system

Abstract

This paper investigates the performance of adaptive particle swarm optimization (APSO) algorithm to find the best optimal gain parameters (Kp, Ki, Kd) in linear PID (L-PID) and nonlinear PID (NL-PID) controllers for the trajectory control of a nonlinear dynamical system (NLDS). Mathematical modeling, stability and performance analysis of a NLDS are being carried out with L-PID and NL-PID controllers. It is found that the APSO based tuning of NL-PID controller provides better tracking of NLDS than L-PID controller in respect of its search domain, premature convergence, and computational efficiency. Simulation analysis of various variables of controllers and system response such as error, cost function, and response of NLDS and their comparison to that of APSO tuned L-PID controller is described. Keywords: Linear PID, Non-linear PID, PID tuning, Swarm intelligence, Nonlinear dynamical system