乙醇制备C4烯烃收率的优化建模

Optimization Modeling of Yield of C4 Olefins with Ethanol

摘要: C4烯烃是重要的基础化工原料，以乙醇为原料生产C4烯烃具有良好的应用前景及经济效益。在乙醇偶合制备C4烯烃的过程中，不同的催化剂组合和温度将对C4烯烃的选择性和乙醇转化率产生重大影响，进一步影响C4烯烃收率。为在减少实验量的基础上探寻多因素下使C4烯烃收率最大的反应条件，本文采用Lasso回归将不同催化剂条件量化表示为多个变量，联合温度条件，分别对乙醇转化率和C4烯烃选择性进行回归，筛选对二者影响显著的变量，分别得到回归方程式；利用回归方程得到C4烯烃收率表达式，并作为目标函数，采用模拟退火算法，解得最优化反应条件；并在现有实验数据及分析的基础上，设计探究更优反应条件的正交试验。

Abstract: C4 olefin is a fundamental chemical raw material. The production of C4 olefin with ethanol as raw material enjoys good application prospect and economic benefits. In the process of producing C4 olefins by ethanol coupling, both various catalyst combinations and temperatures will have a significant impact on C4 olefin selectivity and ethanol conversion, further influencing yield of C4 olefins. In order to explore the reaction conditions that maximize the yield of C4 olefins under multiple factors on the basis of reducing experiments, this paper uses Lasso regression to regress the ethanol conversion rate and C4 olefin selectivity respectively by quantifying different catalyst conditions as multiple variables and combining with temperatures, and then the regression equations are obtained respectively through screening the variables that influence ethanol conversion rate and C4 olefin selectivity significantly. Furthermore, the yield expression of C4 olefin is derived from the regression equations, which is taken as the objective function. Consequently, the simulated annealing algorithm is adopted to obtain the optimal reaction condition, and based on existing experimental data and analysis, an orthogonal test for exploring the optimal reaction condition is designed.