Global optimality of hybrid MSF-RO seawater desalination processes

This article addresses the global optimization of hybrid MSF-RO desalination processes using differential evolution algorithm. All possible stream contacting options for MSF process with brine recycle (MSF-BR) and single stage RO system (SRO) have been considered to achieve five distinct alternate hybrid MSF-RO processes 1-5. For comparison purposes, sequential quadratic programming (SQP) algorithm available in the MATLAB optimization toolbox and its variants (multi-start SQP and DESQP) has been considered. Based on optimization studies, the hybrid 4 MSF-RO process has been identified to provide lowest freshwater production cost of 0.9703 $/m3 for an optimal decision variable value set of [9527.8 m3/h, 1.0439, 20, 3.5082 kW/m2·K, 3.2792 kW/m2·K, 3.2884 kW/m2·K, 8.62 m, 68.03 bar, 8054, 491.8 ppm, 0.8061] for variable set of [WMF, RH, NR, UB, UR, Uj, LT, P1F, NM1, C1RP, r1]. Further, inequality constraint resolution has been evaluated to be the best for the DE in comparison to MS-SQP, SQP and DE-SQP. Parametric sensitivity of various process and operating cost parameters indicated non-linear variation of the hybrid MSF-RO process cost profiles which are significantly different from the linear profiles available in the literature.