Abstract

When it comes to the problem of expressing intricate non-linear interactions, one relatively recent development in the field of mathematical modelling is the application of artificial neural networks, which are also abbreviated as ANNs in some instances. In this paper, we develop a machine learning prediction model for predicting the flow of mass transfer in an alumina matrix porous media. Consider of a cylinder with a catalyst layer on its surface and a porous media surrounding it that is completely filled with fluid except for the one end. This cylindrical device is typical of a catalytic reactor. When the cylinder is heated to a constant temperature, the chemically reactive zeroth-order material is predicted to completely coat the outside of the vessel. Reinforced porous materials undergo a continual, temperature-dependent chemical reaction in their fluid phase. The model shows an improved predictive performance in all its experimentation.

Key words: Mass Transfer, ANN, Machine Learning, Prediction