چكيده به لاتين
The sustainable production process of dimethyl ether includes the optimal use of carbon dioxide and makes it an attractive way to reduce environmental pollutants. The dimethyl ether production reactor is a fixed-bed tubular cocurrent flow system with membrane walls to remove water from the reaction medium and increase carbon dioxide conversion. In this research, a mathematical model was created and the equations of energy, mass and organ were solved simultaneously using the Rangkota technique (4,5). Validation of the model was confirmed according to three previous studies. After verifying the validation, it was shown that the SBWR state equation is the most suitable for modeling by adding water membrane, hydrogen membrane and one charge of both membranes at the same time. In addition, the results showed that the molar fraction of dimethyl ether in the membrane reactor WMR, HMR and WHMR increased by 13.64%, 8.86% and 23.33%, respectively, compared to CR. Also, black box model methods, response surface methodology (RSM) and artificial neural networks (ANN), including multilayer perceptron (MLP) and radial basis function (RBF) neural networks, were used to comprehensively investigate the relationships between key parameters. Input parameters such as gas-hour space velocity (GHSV), pressure, temperature, length-to-diameter ratio, mole ratios of hydrogen to carbon oxides, and overall membrane permeation with outputs being the mole fractions of dimethyl ether, carbon dioxide, and carbon monoxide, in were considered. Exceptional performance was shown by SLP model through hyperparameter optimization and neural network modeling, achieving R2 0.9998 and MSE 1.57×10-5. The optimal RSM model, quadratic regression, has been identified, which reaches R2ave 0.9949 and MSEave 1.14×10-2. 3D diagrams were used to show the effects of different parameters on the output. In addition, three functional equations of six key parameters were obtained using RSM to facilitate output generation.
