基于BP神经网络模型的催化湿式氧化正丁酸反应条件的优化  

Reaction Condition Optimization of Butyric Acid under Catalytic Wet Air Oxidation Based on BP Artificial Neural Network Model

作　　者：李鱼[1,2] 郑爽[1] 王江玲[2] 陈宇[1] 

机构地区：[1]吉林大学环境与资源学院,长春130012 [2]华北电力大学能源与环境研究中心,北京102206

出　　处：《吉林大学学报（理学版）》2009年第2期397-402,共6页Journal of Jilin University:Science Edition

基　　金：吉林省环境保护项目基金(批准号:吉环科字第03-07号)

摘　　要：研究正丁酸原液TOC质量浓度、催化剂用量、反应温度、氧气分压对Mn-Ce催化剂催化湿式氧化正丁酸恒温反应过程的影响,建立了催化湿式氧化降解正丁酸恒温反应过程的BP神经网络模型.经计算,模型的模拟效率系数NSC=0.973 2>0.80,表明所建模型可以较准确地预测催化湿式氧化过程中残余的正丁酸质量浓度;在BP神经网络模型上研究了催化湿式氧化降解正丁酸恒温反应过程影响因素的最优取值,结果表明,在最优反应条件下,催化湿式氧化降解正丁酸的效率可提高约20%.Effects of initial TOC of butyric acid, catalyst dosage, reaction temperature, and initial oxygen pressure on the degradation reaction process of butyric acid under catalytic wet air oxidation were investigated, and a BP artificial neural network model for the degradation process of butyric acid under catalytic wet air oxidation was established. The simulated efficiency factor of the model was calculated to be 0. 973 2 （ more than 0.80）, indicating the model established could well and truly forecast the residual butyric acid in the degradation reaction process of catalytic wet air oxidation. Meanwhile, the reaction condition optimization of butyric acid under catalytic wet air oxidation based on the BP artificial neural network model was carried out, and the results show that the degradation efficiency of butyric acid under catalytic wet air oxidation was nearly enhanced by 20%.

关 键 词：BP神经网络 催化湿式氧化 正丁酸 降解 数学模拟与优化 

分 类 号：X132[环境科学与工程—环境科学]