基于多元回归的对苯二甲酸加氢精制宏观动力学模型  

Macrokinetic Model for Hydropurification of Terephthalic Acid Based on Multivariate Regression Model

作　　者：胡春平[1] 颜学峰[1] 

机构地区：[1]华东理工大学自动化研究所,上海200237

出　　处：《化学反应工程与工艺》2008年第5期468-471,共4页Chemical Reaction Engineering and Technology

基　　金：国家自然科学基金(20506003;20776042);国家863项目(2007AA042164;2007AA042171)

摘　　要：针对在间歇反应釜钯/炭催化剂条件下对苯二甲酸(TA)加氢精制反应过程中,各反应因素对反应过程影响复杂、呈高度非线性的特征、且各种反应过程假设难免与真实机理存在出入,提出一种基于多元回归速率常数模型的加氢精制宏观动力学模型方法。该动力学模型,首先基于反应步骤确定动力学模型框架;然后,应用非线性偏最小二乘回归,基于实验数据,以预测性能为指标,挖掘各反应因素与反应速率常数之间的关联模型,进而建立加氢精制反应过程宏观动力学模型。实际应用表明,所建立动力学模型具有良好的拟合精度,其平均相对误差比文献报道值低31.6%。A novel approach based on multivariate regression model of rate constants was proposed to develop the kinetic model of terephthalic acid hydrogenation. First, the framework of kinetic model was built by the reaction mechanism of hydrogenation. Second, based on experiments date, a model of the rate constants with the reaction factors was developed by nonlinear partial least squares regression method in such a way that it minimized the prediction error of the model. And then, a macrokinetic model of terephthalic acid hydrogenation was established. The results showed that the good predicting ability of the model was obtained, and the absolute value of model＇s analog relative total error decreases by 31.6% compared with the reported literature data.

关 键 词：对苯二甲酸 加氢精制 偏最小二乘回归 动力学模型 

分 类 号：TQ013.2[化学工程] TQ245.1