Development of artificial intelligence-based models for the prediction of filtration performance and membrane fouling in an osmotic membrane bioreactor

Currently, osmotic membrane bioreactor (OMBR) is actively employed for municipal and industrial wastewater treatment. However, there are still some barriers limit the application of the OMBR process on a real scale. This study aimed to develop artificial intelligence (AI)-based models for the early prediction of OMBR system performance in order to increase the system efficiency, then reducing the environmental impacts of wastewater. Based on systematic evaluation of the operational parameters, appropriate input datasets were recommended for each simulation model, in which conductivity and pH parameters played an essential role in all but one simulation. Model structures for all simulations were evaluated for the optimal number of hidden layers (2-6 layers) and the appropriate number of neurons in each layer (5-30 neurons). The developed models demonstrated extremely good performance, with R-2 values of 0.92 and 0.93 reported for the prediction of water flux and membrane fouling simulations, respectively. The root mean square errors (RMSEs) for these predictions were 1.4 LMH and 0.8 x 10(14) m(-1), respectively. Moreover, the developed model for TOC removal also displayed very high efficiency, with R-2 and RMSE values of 0.98% and 0.3%, respectively. Strong predictions were obtained for the simulation of nutrient rejections as well (NH4-N, TN, PO4-P, TP), with all R-2 values > 0.84 and RMSEs < 1.1%. These results indicate that artificial intelligence-based models are highly functional for predicting early controls and optimizing OMBR systems.