Novel non-linear PID based multiple-controller incorporating a neural network learning sub-model

Abstract

The paper proposes a new non-linear adaptive PID based multiple-controller incorporating a neural network learning sub-model. The unknown non-linear plant is represented by an equivalent model consisting of a linear time-varying sub-model plus a non-linear neural-networks based learning sub-model. The proposed multiple-controller methodology provides the designer with a choice of using either a conventional PID self-tuning controller, a PID based pole-placement controller, or a newly proposed PID based pole-zero placement controller through the flick of a switch. Simulation results using a non-linear plant model demonstrate the effectiveness of the proposed multiple-controller, with respect to tracking set-point changes with the desired speed of response, penalising excessive control action, and its application to non-minimum phase and unstable systems.