Abstract
The nanofiltration (NF) membrane technology implemented in the recirculating aquaculture system (RAS) suffers from fouling issues, which cause a decrease in the flux value and performance of the NF membrane. Fouling can be reduced by incorporating pre-treatment units such as the multimedia filter unit (MMF) and activated carbon filter (CF). Therefore, this research aims to minimize fouling and maintain optimal membrane flux and performance by integrating pre-treatment units such as MMF and CF, as well as conducting detailed flux modeling. The models simulate transport phenomena within the NF membrane system, emphasizing the impact of concentration polarization and fouling on flux and overall performance across various pre-treatment configurations under non-steady state conditions. Configuration of MMF-CF-NF (V1) demonstrated the highest overall efficiency in reducing ammonia concentration by 97.5%, nitrite by 100%, and 100% by TSS. Furthermore, the values of the solvent transport coefficient (Aw), the solute transport coefficient (Bs), and the contaminant accumulation volume (X) varied for each configuration due to different treatments, leading to varying membrane workloads. Validation and statistical analysis of the permeate concentration model demonstrated that the developed model has a strong capability to accurately describe and predict the filtration process using the NF membrane, with Bs contributing most significantly to influencing the concentration permeate NF.