Mathematical optimization model on BP neural network for the preparation of C4 olefins by ethanol coupling

Abstract

With industrial production, the continuous consumption of resources, people’s use, and consumption of resources make the community concerned about resources. C4 olefins are used as significant raw materials in chemical production, and the use of ethanol coupling to prepare C4 olefins will become an important route in industrial production in the future. The data of ethanol conversion and C4 olefin selectivity can reflect the utilization of resources in the reflective process.With the rapid growth of machine learning and intelligent computing, a large number and different kinds of models have been constructed for the utilization and consumption of resources, which makes it possible to optimize the process of ethanol coupling for the preparation of C4 olefins. The objective of this paper is to construct a mathematical optimization model for the preparation of C4 olefins by ethanol coupling. Firstly, the catalyst combination was split, and secondly, the prediction model for ethanol conversion and C4 selectivity was constructed based on BP neural network. Finally, the feasibility of the model was tested by determining three indicators, Correlation Coefficient (R), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE). Meanwhile, the PCA algorithm was applied to rank the degree of influence of catalyst and temperature on the conversion of ethanol and C4 olefin selectivity.