Experimental investigation and artificial neural networks ANNs modeling of electrically-enhanced membrane bioreactor for wastewater treatment

• Membrane bioprocess is hybridized with electrokinetics for wastewater treatment.
• 99, 99, and 98% removal efficiency of COD, PO43−-P, and NH4+-N is achieved.
• Modeling of electrically enhanced MBR using artificial neural networks (ANNs).

In this work, a new configuration of an electrically-enhanced membrane bioreactor has been introduced to treat medium strength wastewater at Masdar City, Abu Dhabi, United Arab Emirates (UAE). The integrated setup enhanced the reduction of wastewater contaminant concentrations. The investigated components in this study were chemical oxygen demand (COD), orthophosphates (PO43−-P) and ammonium (NH4+-N). The percentages of COD, PO43−-P, and NH4+-N removal obtained were 98, 99, and 98%, respectively. Variation in environmental compositions such as mixed liquor dissolved oxygen (DO), volatile suspended solids (MLVSS), pH, and electrical conductivity influenced the effluent concentration of wastewater components. Artificial neural networks (ANNs) based ensemble model was used to model the experimental findings of COD, PO43−-P and NH4+-N removal given the initial mixed liquor compositions. Comparison between the model results and experimental data set gave high correlation coefficients for COD (r = 0.9942), PO43−-P (r = 0.9998) and NH4+-N (r = 0.9955).