Computational Modeling of Combustion and Soot Evaluation in GDI engine
Two of the most common types of engines used in today’s cars are GDI (Gasoline Direct Injection) and PFI (Port Fuel Injection) engines. GDI engines offer increased fuel economy over PFI engine, but they produce high particulate matter emissions compared to PFI engines. So, it is essential to model soot production and control its root cause. Soot generation occurs over various stages viz., inception, surface reactions, coagulation and condensation. Soot modelling was carried out using ConvergeCFD software. Modeling involved specifying the reaction mechanisms and associated optimization algorithms, defining a fuel spray model, selecting combustion, emission and soot models. Particulate Mimic (PM) model, which uses moments of various soot source terms was used to model soot mass, number density and moments. The modeling replicated the experimental setup by Bock et al (2018). The pressure response from modeling reasonably matched with those from the experiment, though the peak occurred slightly earlier. Soot number density and soot mass for stochiometric and lean mixes were studied. It was observed that the soot mass and the number density were less for the lean mixture. Also, the soot mass and number density were found to be less in the exhaust region than in the cylinder. First three moments (M0, M1 and M2) were analyzed and interpreted using Tecplot. Log-normal probability density function was derived from the moments for soot particle size. Peak of the distribution occurred at class number of 8.