BP网络-正交设计用于炼油厂废水处理中活性污泥的活化

BP Model & Orthogonal Design Applied to the Activation of Activated Sludge in Refinery Wastewater Treatment

为探讨生化法处理炼油厂废水中活性污泥的活化条件,本文利用BP神经网络的数值逼近方法和实验检测方法,选择COD、硫化物、温度、DO为影响因素,针对活性污泥浓度和指数,得到了处理条件下的预测数值和实验数值,二者接近,说明BP神经网络预测的精确度较高.通过对检测结果的正交分析,得到了处理炼油厂废水用活性污泥的活化条件.

This paper is to study the biochemical method used in the treatment of activated sludge in refinery wastewater under the activation conditions. The influence of such factors as the COD, sulfides, temperature and DO is investigated, with a view to studying the concentration and index of sludge, and the results show that the predic- tion value is close to that of the experiment. Using the optimized parameters by the BP, this model is proved to have a high level of accuracy. The activation conditions of activated sludge in refinery wastewater treatment were obtained by orthogonal analyses on the data of the experiment.