Study on the preparation of C4 olefins by ethanol coupling based on multiple regression analysis

Abstract

With the continuous development of the current chemical industry, the demand for olefins as raw materials for the production of chemical products and pharmaceutical intermediates is increasing. Compared with traditional fossil energy, ethanol is an excellent green raw material for the production of C4 olefins. In the preparation process, the selectivity of C4 olefins and the conversion rate of ethanol depend to a certain extent on the reaction temperature and the catalyst of the reaction, so the reasonable selection of catalyst and temperature becomes very important. In order to study the effects of different catalyst combinations and temperatures on ethanol conversion and C4 olefin selectivity, a multiple regression model was established and its endogeneity was tested by Monte Carlo simulation. The stepwise regression method was used to select the appropriate independent variables, and then the model was optimized to reduce the dimension, and the first optimized linear regression equation was obtained. Finally, in order to test and eliminate the heteroscedasticity of the model, this paper uses the combination of White test and Wls least squares estimation and robust variance error to obtain the second optimized linear regression equation.