Event-driven enabled regression aided multi-loop control for SEC minimisation in SWRO desalination considering salinity variation

Abstract

This paper addresses the energy minimised operation of seawater reverse osmosis (SWRO) desalination process by simultaneous manipulation of feed pressure and reject valve opening. The specific energy consumption (SEC) analysis of SWRO desalination process for maintaining constant permeate flow during feed salinity variation is performed. The analysis is carried out to identify the suitable manipulating variables that can reduce the energy requirement for regulating permeate flow during feed salinity variation. Based on the analysis, a multi-loop control strategy using event-driven programming paradigm aided by regression based predictive model is proposed. The proposed methodology is more desirable than traditional event-driven multi-loop PID control due to smoother control transition and energy reduction. The investigation of SEC and analysis of proposed control strategy were performed using a previously validated dynamic model for SWRO desalination process. The simulation results show that the proposed methodology is superior to conventional PID control by enabling energy-minimised operation of RO process with significant reduction of feed pressure. The analysis shows that the proposed control approach reduces the feed pressure requirement by approximately 300 kPa during feed salinity variation.