Search

In this paper we have used a detailed reaction network, and the reaction experimental data that has been obtained from a thorough literature survey is used as a basis for the comparison and selection of plug flow reactor (PFR) and continuous stirred tank reactor (CSTR) for the synthesis of methanol over Cu/ZnO/Al2O3 Catalyst. The experimental data that has been obtained is operating between 210°C and 260°C and at pressure up to 70 bar. A generalized Machine learning-based data-driven methodology principal component analysis (PCA) is used for dimensionality reduction of the parameter presented in experimental data; further linear regression model is presented to compare the accuracy of the model with industrial-scale application. The data obtained was also utilized for the prediction of variable for the model which is proposed in this paper. The performance of the model for two different reactors has been compared using the R-Square value. Based on this comparison it is found that the plug flow reactor (PFR) is more efficient, with 94.48% prediction accuracy (R-Square) as compared to continuous stirred tank reactor (CSTR), with 91.98% prediction accuracy. However, more experimental data Is needed, as the used dataset is unable to differentiate for the regression of the adsorption isotherms. Model used is still performing well which suggests that this model can be further used for predictive analysis for experiments outside the training set. [ABSTRACT FROM AUTHOR]