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

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%.