Design of a self-tuning controller for local water quality adjustment

Abstract

This paper presents the design of a self-tuning control scheme for local adjustment of water quality, through treatment by using Generalized Minimum-Variance Control (GMVC). It incorporates a self-tuning estimator with Recursive Extended Least-Squares (RELS) method to approximate the unknown parameters of a discrete-time single-input single-output local water quality model. The proposed self-tuning scheme can be implemented on-line to automatically tune the parameters and derive the controller using an indirect self-tuning procedure. When deriving the controller, the closed-loop tracking performance in achieving the desired level of the particular water quality parameter at the measured node is the primary consideration. Additionally, the influence of noise and disturbance on the local water quality model is considered. The good performances of the set-point tracking and disturbance rejection in the simulation examples show the effectiveness and reliability of the designed self-tuning controller in improving the local quality of water through automated water treatment.