Runge-Kutta Model-Based Adaptive PID Controller Applied to a 4 Coupled-tank System

Abstract

In this paper an adaptive PID controller is presented, which updates its parameters by minimizing a cost function composed of the system outputs predictions over a prediction horizon. A model of the nonlinear system based on the fourth-order Runge-Kutta method is used to calculate the predictions. The proposed method is tested by simulations in a 4 coupled-tank MIMO system which has a non-minimum phase configuration. The results show that with the proposed approach the desired reference is reached in a more optimal way at both outputs of the system compared to a standard PID. This is evidenced by a reduction of up to 15% in the RMSE of both outputs with the proposed approach, which is achieved by increasing the prediction horizon.