Performance Evaluation Of A Wiener Model Predictive Control For A Coagulation Chemical Dosing Unit In Water Treatment Plants

Coagulation process control is an essential operation in water treatment plants. It is a challenging control problem due to the nonlinear and physicochemical nature of the coagulation process. The paper presents the application of the Wiener model predictive control (WMPC) algorithm to a coagulation chemical dosing unit for water treatment plants in order to keep the surface charge and pH level of the dosed water at the reference trajectory set by the operator of the plant. Wiener models with different nonlinear estimators are compared and evaluated. A Wiener model with a wavelet network estimator presented the highest goodness of fit (98%), and was thus selected as the best prediction model. Simulation results show that the proposed control strategy has good set-point tracking, as well as noise and disturbance rejection performances.