New publication about a machine learning framework for studying of spray and combustion processes


10th November 2022

A machine learning model for predicting the thermophysical properties at a wider range of temperature (300 to 700 K) that are essential for spray, combustion, and emission modeling  is developed in this work. The model uniqueness is that, it can predict the various properties of biodiesel fuels based on only their composition without using the complex analytical correlations. A good agreement is observed for the model results with a maximum mean absolute percentage error of 10 %.  Hence, the model is robust in terms of the fuel property predictions over a broad range of temperature.