Auto-selection of quasi-components/components in the multi-dimensional quasi-discrete model

A new algorithm for the auto-selection of quasi-components and components (QC/Cs) in the ‘multi-dimensional quasi-discrete’ model is suggested. This algorithm is applied to the analysis of heating and evaporation of multi-component fuel droplets. It allows one to automatically select QC/Cs and update the initial selection during droplet evaporation. The new algorithm is expected to be applicable to the analysis of a wide range of fuels and fuel blends. It can be directly implemented into CFD codes with minimal intervention by end-user. Using this algorithm, the effects of transient diffusion of species on droplet lifetimes are investigated for mixtures of Diesel and E85 (85% vol. ethanol and 15% vol. gasoline) fuels. It is shown that the new algorithm can reduce the analysis of the E85-Diesel fuel droplets, taking into account the contributions of up to 119 components at the initial stage of heating and evaporation, to that based on 5 QC/Cs, near the end of droplet evaporation, with up to 1.9% errors in predicted droplet temperatures and radii. The CPU time needed to perform calculations using the new algorithm is shown to be 80% less than that when considering the full composition of fuel.