利用神经网络动态模拟污水生物脱氮过程

Dynamic Modeling of Wastewater Bio-nitrogen-removal Process Using Neural Network

用以活性污泥数学模型 (ASM)为基础的EFOR软件的模拟输入输出数据训练设计好的神经网络 ,然后同时变化输入数据 ,比较输出数据。结果表明神经网络用于污水生物脱氮过程的动态的模拟 ,预测出水COD的误差可控制在± 3%以下 ,出水TN的误差可控制在± 6 %以下 ,出水SS的误差可控制在± 10 %以下。

In this paper,designed neural networks trained by i np ut and output data from EFOR were introduced.Then change the input data at the s ame time,and compared the output data of EFOR and neural networks.It could be co ncluded that neural networks can be used to model wastewater bio-treatment for nitrogen removal.The errors respectively were COD<±3%;TN<±6%;SS<±10%.