In order to analyze the effects of different catalyst combinations and temperatures on ethanol conversion and C4 olefin selectivity, BP neural network was trained to predict ethanol conversion and C4 olefin selectivity. Then, using the control variable method, the ethanol conversion and C4 olefin selectivity were predicted by neural network. Finally, the xgboost model is trained according to the data, and the particle swarm optimization algorithm is used to optimize the xgboost parameters in the training process, so that the prediction accuracy is improved from 65.97% to 93.75%, and the prediction accuracy is improved by 27.28%. Using the trained xgboost model to input the combination of different characteristic indexes, the final result is more than 350 degrees, and the optimal C4 olefin yield is 27.823%. The final result is that the yield of C4 olefin is 177.252%.

Jianxun Feng. Research on C4 Olefins Prepared by Ethanol Coupling Based on BP Neural Network. Academic Journal of Engineering and Technology Science (2021) Vol. 4, Issue 6: 46-50. https://doi.org/10.25236/AJETS.2021.040608.

[1] LV Shaopei. Preparation of butanol and C4 olefins by ethanol coupling [D]. Dalian University of technology, 2018.

[2] Liu Fang, Xu long, Ma Xiaoxun. Development of BP neural network and its application in chemistry and chemical engineering [J]. Chemical progress, 2019, 38 (6): 2559-2573.

[3] Wang Ning, Cao Cuiwen. Dynamic multi output prediction model of refinery hydrogen network based on xgboost model [J]. Journal of East China University of Technology (NATURAL SCIENCE EDITION), 2020, 46 (1): 77-83.