Taritas, I., Kozarac, D., Sjeric, M., Sierra Aznar, M. et al., "Development and Validation of a Quasi-Dimensional Dual Fuel (Diesel – Natural Gas) Combustion Model," SAE Int. J. Engines 10(2):483-500, 2017, doi:10.4271/2017-01-0517.
This paper presents a newly developed quasi-dimensional multi-zone dual fuel combustion model, which has been integrated within the commercial engine system simulation framework. Model is based on the modified Multi-Zone Combustion Model and Fractal Combustion Model. Modified Multi-Zone Combustion Model handles the part of the combustion process that is governed by the mixing-controlled combustion, while the modified Fractal Combustion Model handles the part that is governed by the flame propagation through the combustion chamber. The developed quasi-dimensional dual fuel combustion model features phenomenological description of spray processes, i.e. liquid spray break-up, fresh charge entrainment, droplet heat-up and evaporation process. In order to capture the chemical effects on the ignition delay, special ignition delay table has been made. Additionally, to capture the effect of entrained methane on the chemical reaction rate, special table that features chemical reaction time scale has also been made. The start of flame propagation is calculated through the newly developed sub-model based on knock integral calculation. The existing k-ε turbulence model has been extended to account for the effect of diesel pilot injection on the increase of in-cylinder turbulence level. Since in the conventional dual fuel combustion process multiple flames propagate through the combustion chamber, multiple flame propagation model has been developed. This model can describe arbitrary number of flames that propagate through the combustion chamber. The validation data for the developed quasi-dimensional dual fuel combustion model have been acquired on a 2.0 liter Diesel engine, which has been modified to operate in the conventional dual fuel combustion mode. The presented model has been validated at different loads and diesel substitution ratios, and there is a good fit between the measured and cycle-simulation data.