Process schemes of ethanol coupling to C4 olefins based on a genetic algorithm for back propagation neural network optimization
Minghan Li,
Lingling Zhao,
Shuo Jin,
Danlu Li,
Jingyi Huang,
Jiaxin Liu
Affiliations
Minghan Li
School of Life and Pharmaceutical Sciences, Dalian University of Technology, 2 Dagong Road, Panjin, Liaoning 124221, PR China; Panjin Campus, Dalian University of Technology, 2 Dagong Road, Panjin, Liaoning 124221, PR China
Lingling Zhao
Panjin Campus, Dalian University of Technology, 2 Dagong Road, Panjin, Liaoning 124221, PR China
Shuo Jin
School of Life and Pharmaceutical Sciences, Dalian University of Technology, 2 Dagong Road, Panjin, Liaoning 124221, PR China; Panjin Campus, Dalian University of Technology, 2 Dagong Road, Panjin, Liaoning 124221, PR China
Danlu Li
School of Life and Pharmaceutical Sciences, Dalian University of Technology, 2 Dagong Road, Panjin, Liaoning 124221, PR China; Panjin Campus, Dalian University of Technology, 2 Dagong Road, Panjin, Liaoning 124221, PR China
Jingyi Huang
Panjin Campus, Dalian University of Technology, 2 Dagong Road, Panjin, Liaoning 124221, PR China
Jiaxin Liu
China Nuclear Power Engineering Co., Ltd., 117 North West Third Ring Road, Beijing 100840, PR China; School of Chemical Engineering, Dalian University of Technology, 2 Linggong Road, Dalian, Liaoning 116024, PR China; Corresponding author at: China Nuclear Power Engineering Co., Ltd., 117 North West Third Ring Road, Beijing 100840, PR China.
C4 olefin is an important feedstock for the chemical industry. Designing an effective and stable industrial process for preparing C4 olefin from renewable ethanol is crucial for further sustainable chemical production. In this study, a comprehensive evaluation system of an experimental scheme was constructed based on the Analytic Hierarchy Process/Entropy Weight Method-Technique for Order Preference by Similarity to Ideal Solution (AHP/EWM-TOPSIS) and Chemical production indicators. Using this evaluation system, a Back Propagation Neural Network (BPNN) based on a Genetic Algorithm (GA) was constructed after simulating C4 olefin production conditions using the Improved Mixed Congruential method. Subsequently, the production scheme with the highest evaluation score was determined when the temperature was not limited and when the temperature was lower than 350°C through a series of mathematical models. Overall, our mathematical models provide guidance for the commercial production of ethanol to butene and effectively reduce the risk of scaling up the chemical process to pilot or industrial scale.
