Predicting Adsorption Behavior in Engineered Floodplain Filtration System Using Backpropagation Neural Networks

Abstract

Engineered floodplain filtration (EFF) system is an eco-friendly low-cost water treatment process wherein water contaminants can be removed, by adsorption and-or degraded by microorganisms, as the infiltrating water moves from the wastewater treatment plants to the rivers. An artificial neural network (ANN) based approach was used in this study to approximate and interpret the complex input/output relationships, essentially to understand the breakthrough times in EFF. The input parameters to the ANN model were inlet concentration of a pharmaceutical, ibuprofen (ppm) and flow rate (md– 1), and the output parameters were six concentration-time pairs (C, t). These C, t pairs were the times in the breakthrough profile, when 1%, 5%, 25%, 50%, 75%, and 95% of the pollutant was present at the outlet of the system. The most dependable condition for the network was selected by a trial and error approach and by estimating the determination coefficient (R2) value (>0.99) achieved during prediction of the testing set. The proposed ANN model for EFF operation could be used as a potential alternative for knowledge-based models through proper training and testing of variables.