Robust Parameter Estimation and Output Prediction for Nonlinear Water Quality Control in Water Distribution Systems

Abstract

An increasing shortage of drinking water resources around the world has been observed, especially for people living in absolute poverty. Hence, meeting demand for drinking water with required quality needs advanced control technology to operate the drinking water distribution systems (DWDSs). This paper aims to implement robust parameter estimation and an advanced output prediction algorithm on a nonlinear water quality model for the purpose of monitoring water quality in DWDSs. The advanced algorithm was tested for robustly predicting the outputs of the water quality model in DWDSs incorporating disinfectant by-products (DBPs). The application of the point-parametric model (PPM) for robustly estimating parameters of the nonlinear water quality model considering DBPs by utilizing a multi-input multioutput (MIMO) model structure is stretched. The piecewise constant parameters within the time-varying parameter model structure explaining the dynamics of water quality considering DBPs in DWDSs under input disturbance are acquired by applying modified piecewise continuity algorithms. In addition, this paper for the first time describes the robustness of the PPM framework on a MIMO nonlinear water quality model under multihydraulic operation status with advanced operating procedures. Simulation results suggest that the PPM technology and its piecewise continuity algorithm are appropriate for robustly feasible model predictive control (RFMPC) on water quality in DWDSs with DBPs involved.