SPEAKER:	Panagiotis Christofides Department of Chemical and Biomolecular Engineering University of California, Los Angeles
TITLE:	Optimal Operation and Control of Reverse Osmosis Desalination Systems
DATE:	April 19 2011
TIME:	10:30am
PLACE:	JHE 342

ABSTRACT

In recent years, the interest in the use of reverse osmosis  membrane desalination has increased due to the energy efficiency  and versatility of this process relative to other water  desalination technologies. Water shortages in various areas of the  world have necessitated further development of the reverse osmosis  desalination process in order to provide clean drinking water to  the people in these regions. When operating a reverse osmosis  desalination process, it is imperative that the system conditions  be monitored and maintained at appropriate set-points in order to  produce the required production rate of clean, potable water.Furthermore, with the rising cost of energy, it is also desired to  deploy operating methods to reduce the energy consumption of  reverse osmosis desalination, especially when confronted with  variability of feed water quality. This task requires the  development and implementation of effective feedback control  strategies. Traditionally, classical (i.e., proportional-integral(PI) or proportional-integral-derivative (PID)) control algorithms  have been used to regulate process flow rates and adjust the system  pressure in order to achieve the desired system operating  conditions. While the traditional control algorithms are able to  maintain a constant permeate flow rate, they do not directly  minimize energy usage by the reverse osmosis process and may cause  damage to the RO membranes in the event of a large feed salinity  fluctuation.This work focuses on the design and implementation of model-based  and optimization-based control systems on an experimental reverse  osmosis (RO) membrane water desalination process. These control  systems serve to address large set-point changes, variations in  feed water salinity, and to facilitate system operation at energy  optimal conditions. A nonlinear model for the RO process is derived  using first principles, and the model parameters are computed from  experimental data. This model is combined with appropriate  equations for reverse osmosis system energy analysis and used to  form the basis for the design of the nonlinear optimization-based  and model-based control systems. The proposed control systems are  implemented on UCLA's experimental RO desalination system and their  set-point tracking, disturbance rejection, and energy optimization  capabilities are evaluated. Recent efforts on the design and  construction of an integrated ultra-filtration/RO desalination  system for shipboard deployment will be also discussed.