Sensitivity analysis of design variables in semi-batch reverse osmosis for energy-efficient water treatment

Stricter water quality regulations have increased the demand for advanced water treatment technologies. Reverse osmosis (RO) is widely applied to remove emerging contaminants but remains limited by water recovery constraints. To address this limitation, semi-batch reverse osmosis (SBRO) has been developed to enhance recovery. Despite growing interest in SBRO, optimal design variables for large-scale water treatment applications have not been sufficiently studied. This study determined optimal SBRO design variables through sensitivity analysis for a large-scale system (100,000 m³/d) to optimize specific energy consumption (SEC). The analysis focused on the number of RO elements per pressure vessel (PV), feed ratio, plug-flow (PF) recovery, and closed-circuit (CC) recycled concentrate flow rate. Cost efficiency was evaluated by determining the optimal number of RO elements per PV, indicating that four elements per PV minimized total costs by balancing PV capital cost and the increase in SEC. Adjusting the feed ratio from 100 % to 200 % reduced the SEC by 6.2 % due to lower operating pressure during the CC phase. PF recovery from 0 % to 50 % improved water production, reducing SEC by 4.0 %. Further analysis of the CC recycled concentrate flow rate, from 54,000 m³/d to 162,000 m³/d, resulted in a 19.3 % increase in SEC. The CC recycled concentrate flow rate was identified as the most influential factor through Shapley additive explanations, and optimized design variables were proposed based on varying feed conditions and overall recovery. This sensitivity analysis provides an analytical basis for optimizing large-scale SBRO design to achieve energy-efficient water treatment, thereby resulting in cleaner water production.