Prediction of pollutants removal from cheese industry wastewater in constructed wetland by artificial neural network

Due to its flexibility in operation and maintenance, besides its efficiencies in wastewater decontamination, constructed wetland is considered as one of most sustainable significant solutions for pollutants elimination from different industrial effluents. In this study, real food industry wastewater was freshly collected from a cheese manufacturing factory and treated in a constructed wetland of type horizontal flow. A multi-layer artificial neural network model was applied in order to predict the efficiency of the constructed wetlands in regard to pollutants removal. Excellent fitting between the prophesy and experimental outcomes and results was observed with high correlation determination up to 0.9929 as well as small mean square error of 0.0028. The promising results from this study approved the validity of horizontal subsurface flow constructed wetlands in organic loading removal from real industrial cheese whey wastewater as a sustainable and cost-effective treatment technology.

Automated summary

Based on the experiment, it was found that horizontal flow constructed wetlands can effectively remove pollutants from food industry wastewater, and the artificial neural network model shows excellent performance in predicting efficiency.