Application of Neural Network in Groundwater Denitrification Process

Biological denitrification is a simple and cost effective method to treat groundwater contaminated by nitrate. However, this process is non-linear, complex and multivariable. In order to tackle the problem to remove nitrate from polluted groundwater, based on artificial neural network (ANN) technique and GUI(graphical user interfaces) function of MATLAB, a groundwater denitrification process forecasting model was built. Experimental results showed that the ANN was able to predict the output water quality parameters-including nitrate as well as nitrite and COD. Most of relative error of NO₃ ^--N and COD were in the range of plusmn10% and plusmn5% respectively. The ANN model of nitrate removal in groundwater prediction results produced good agreement with experimental data. Simulation testing proved that the model could forecast well and which could be used in groundwater quality forecast.