Advanced Control Synthesis for Reverse Osmosis Water Desalination Processes

• Published in: Water environment research Vol. 89; no. 11; pp. 1932 - 1941
• Main Authors: Phuc, Bui Duc Hong, You, Sam-Sang, Choi, Hyeung-Six, Jeong, Seok-Kwon
• Format: Journal Article
• Language: English
• Published: United States John Wiley and Sons, Inc 01-11-2017; Water Environment Federation; Blackwell Publishing Ltd
• Subjects: Computer simulation; Control; Decoupling method; Desalination; desalination plant; Desalination plants; Distillation; Distilling; disturbance; Dynamics; Energy consumption; Interactions; MIMO (control systems); Model reduction; Noise reduction; Osmosis; parametric uncertainty; Reverse osmosis; Robust control; Robust design; Salinity; Seawater; Simulation; structured singular value; Structured singular values; Synthesis; Uncertainty; Variables; Water desalting; Water flow; Water Purification - methods; water quality; water supply
• ISSN: 1061-4303; 1554-7531
• DOI: 10.2175/106143017X15054988926316

In this study, robust control synthesis has been applied to a reverse osmosis desalination plant whose product water flow and salinity are chosen as two controlled variables. The reverse osmosis process has been selected to study since it typically uses less energy than thermal distillation. The aim of the robust design is to overcome the limitation of classical controllers in dealing with large parametric uncertainties, external disturbances, sensor noises, and unmodeled process dynamics. The analyzed desalination process is modeled as a multi-input multi-output (MIMO) system with varying parameters. The control system is decoupled using a feed forward decoupling method to reduce the interactions between control channels. Both nominal and perturbed reverse osmosis systems have been analyzed using structured singular values for their stabilities and performances. Simulation results show that the system responses meet all the control requirements against various uncertainties. Finally the reduced order controller provides excellent robust performance, with achieving decoupling, disturbance attenuation, and noise rejection. It can help to reduce the membrane cleanings, increase the robustness against uncertainties, and lower the energy consumption for process monitoring.