Proposed new fouling monitoring indices for seawater reverse osmosis to determine the membrane cleaning interval

This study focuses on how to systematically determine the membrane cleaning interval of the seawater reverse osmosis (SWRO) desalination process by proposing two new fouling indices as well as comparing the fouling model with the fouling-free model. A fouling-free mathematical model of the SWRO desalination process based on the effects of temperature on flux as well as operation data obtained from a full-scale SWRO desalination plant in Korea was developed in order to normalize the permeate flow rate and feed pressure values. The normalized permeate flow rate was monitored in order to determine the proper cleaning interval of the RO membrane based on the membrane manufacturing company's cleaning protocol. The normalized feed pressure was modeled based on the linear cake filtration theory and dynamic fouling indices including an updated slope β and intercept γ that were proposed using the recursive least square (RLS) method. In addition, statistical control charts for the dynamic fouling indices were proposed in order to determine the proper membrane cleaning interval based on standard rules. The proposed fouling model was calibrated by the Monte-Carlo and principal component analysis (PCA). The experimental results from a full-scale SWRO plant in Geojae Island, Korea, indicate that the proposed fouling model and fouling monitoring indices were able to analyze the fouling mechanism of the RO membrane and were also able to determine the optimal membrane cleaning intervals.