Modeling and optimization of process parameters of biofilm reactor for wastewater treatment

Abstract

The efficiency of heavy metal in biofilm reactors depends on absorption process parameters, and those relationships are complicated. This study explores artificial neural networks (ANNs) feasibility to correlate the biofilm reactor process parameters with absorption efficiency. The heavy metal removal and turbidity were modeled as a function of five process parameters, namely pH, temperature(°C), feed flux(ml/min), substrate flow(ml/min), and hydraulic retention time(h). We developed a standalone ANN software for predicting and analyzing the absorption process in handling industrial wastewater. The model was tested extensively to confirm that the predictions are reasonable in the context of the absorption kinetics principles. The model predictions showed that the temperature and pH values are the most influential parameters affecting absorption efficiency and turbidity.

Keywords: Artificial neural networks (ANN); Biofilm reactor; Heavy metal removal; Index of relative importance (I(RI)); Wastewater treatment; Weight distribution.