Modeling and optimization of a solar forward osmosis pilot plant by response surface methodology

Abstract

Forward osmosis (FO) is a water treatment/separation technology of emerging interest. Due to its complex nature involving various operating parameters, modeling of this separation process is challenging. A solar thermal and photovoltaic-powered FO pilot plant has been optimized by means of a statistical experimental design and response surface methodology. Predictive models were developed for simulation and optimization of different responses such as the water permeate flux, the reverse solute permeate flux and the FO specific performance index that includes the water and reverse solute permeate fluxes together with the energy consumption. The considered input variables of the FO pilot plant were the feed flow rate, the permeate flow rate and the temperature. The developed response models have been tested using the analysis of variance. A Monte Carlo Simulation method has been conducted to determine the optimum operating conditions of the FO pilot plant. The obtained optimum parameters were confirmed experimentally. Regeneration of the draw solution can be performed by means of an optimized solar powered reverse osmosis (RO) pilot plant with an optimum FO specific performance index ranging from 25.79 to 0.62L/gkWh achieved under the FO optimal conditions, 0.83L/min feed flow rate, 0.31L/min draw solution flow rate and 32.65°C temperature. The FO energy consumption is only 14.1% the total energy consumption of the FO/RO hybrid system.