[ONLINE](RTM) simulation of constrained model predictive control of reverse osmosis desalination plant.

摘要

The study presented here extends the investigation of the feasibility of applying an advanced process control strategy to a reverse osmosis (RO) desalination pilot plant. A simulation study was conducted to assess the benefits of Constrained Model Predictive Control (CMPC) in a hollow fiber RO membrane desalination system. The models required for CMPC control were obtained by testing step changes in manipulated variables in the pilot plant. The performance of CMPC was then examined and optimized using [ONLINE]RTM software. Furthermore, comparison was made with previous experimental results of the application of CMPC to a spiral wound RO unit. The simulation results showed that: the inherent interaction compensation and constraint handling capability of CMPC permitted, this simulated hollow fiber RO system to achieve the goals of producing the desired water flow with satisfactory salt content. At the same time CMPC kept the trans-membrane pressure and inlet pH within specified limits. CMPC also demonstrated that the operation of this RO desalination pilot plant can be more conveniently optimized due to the various flexible features and advanced control schemes provided by CMPC.