Author(s)

Yihang Luan, Zhangming Zhao

Corresponding Author

Yihang Luan

ABSTRACT

In this paper, a statistical regression model is established to analyze and study the relationship between temperature, catalyst combination and catalytic performance, and a multi-objective programming model is established to obtain the optimal process conditions. Firstly, after filtering and completing the data, it is found that there is a nonlinear correlation with the change of temperature. Then one-way ANOVA was used to calculate that the effect of temperature on ethanol conversion and C4 olefin selectivity was significantly positively correlated. Then, using the method of controlling variables, taking different catalyst combinations and temperature as independent variables, ethanol conversion and C4 olefin selectivity as dependent variables, and further splitting different catalyst combinations into three independent variables. The data were analyzed by two-way variance model to obtain the specific significance. Finally, taking the maximum C4 olefin yield as the goal, it is decomposed into two objective functions: the maximum ethanol conversion and the maximum C4 olefin selectivity, and a multi-objective programming model for the optimal C4 olefin yield is established. Compared with other traditional solving methods, evolutionary algorithm does not need the reference of weight dimension or ideal value, which is more suitable for the situation of this problem. The fast non dominated sorting genetic algorithm based on elite strategy (NSGA-II) is used to solve the problem.

KEYWORDS

Nonlinear regression, Control variables, Two factor analysis of variance, Multi objective programming model