BP神经网络预测废水处理过程的研究

通过在实验室条件下进行追纸废水处理试验取得的数据对BP神经网络进行训练，建立了造纸废水处理过程的网络模型。该网络模型仿真实际废水处理过程的结果表明，BP神经网络具有很强的学习能力。利用BP神经网络模型实现了对造纸废水处理过程的预测，平均相对误差为19％，表明网络泛化能力不是很好。经过动态训练的BP神经网络模型能够比较准确的预测造纸废水处理过程，平均相对误差为1．9％，大大优于未经动态训练的网络模型。

Studies on predicting the effluent treatment process with BP neural network

The BP neural network trained with the data from the papermaking effluent treatment experiment under laboratory conditions and a network model was built for the papermaking effluent treatment process. The results of emulating practical wastewater treatment process of the network model show that the BP neural network has a strong learning capability. The effluent treatment process was predicted with this BP neural network model with the average relative error of 19%, which indicates that the generalization power of the network is not so desirable. While the BP neural network model after a dynamic training, it had a more precise prediction on the papermaking effluent treatment process with the average relative error of 1.9%, which indicates that the dynamic training BP neural network is greatly superior to the one without dynamic training.