Research on Preparation of C4 Olefins Based on Multiple Regression Analysis and Genetic Algorithm Optimization

Abstract

In this paper, aiming at the process of preparing C4 olefins from ethanol through coupling reaction, a multi-objective optimization model is established with the goal of C4 olefin yield, and the system parameters such as catalyst combination mode and temperature are searched and solved. Firstly, it is found that there is a certain relationship between ethanol conversion, C4 olefin selectivity and temperature change, and multiple functions are used for regression analysis. Taking catalyst A1 as an example, the best regression model between ethanol conversion and temperature is y1 = 0.003x² – 1.193x + 141.807. The best regression model between C4 olefin selectivity and temperature is y2 = 0.001x² – 0.681x + 88.406. Ethylene selectivity time optimal model: y3 = – 8.936e – 05x² + 0.030x + 2.252. Then, the influence model of ethanol conversion rate is established, the model is solved by the least square method, and the genetic algorithm is used to optimize the model. The optimal model of ethanol conversion rate is:w1 = 83.9812pethanol + 0.9895t + 0.4925βCo/Sio2 – 84.2251. The optimal model of C4 olefin absorption is: w2 = –2.2047λco + 0.6778t + 0.5841βCo/Sio2 – 44.4423.