应用神经网络评价地下水脱氮效果

Assessment of groundwater denitrification by artificial neural networks

为解决地下水硝酸盐污染,应用人工神经网络评价生物脱氮工艺硝酸盐的去除效率.采用工艺参数COD、NO3—N、NO2—N、MLSS、DO等作为输入节点,COD、NO3—N、NO2—N作为输出节点.神经网络实验结果显示神经网络能够较好地预报出水的水质参数,其预测结果与试验结果符合得较好.

Ground water contamination by nitrate is a globally growing problem. Biological denitrification is a simple and effective method. However,this process is non-linear,complex and multivariable. This paper presents the application of artificial neural network （ANN） in denitrification process in ground water. Experimental results show that the ANN is 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 are in the range of ±10% and ±5% respectively. The ANN model of nitrate removal in ground water prediction results produced good agreement with experimental data.