The design and optimization stages of combustion systems for modern Heavy Duty Diesel engines must be supported by reliable CFD tools for the definition of the chamber geometry and injection strategy. To be fully predictive in terms of in-cylinder thermodynamics and flame structure, the employed combustion models must account for complex chemistry and turbulence-kinetics interactions. Within this context, the authors have implemented into an opensource code a model based on the multiple Representative Interactive Flamelets approach (mRIF ) and applied it to Diesel combustion simulations. New numerical techniques were integrated in the proposed mRIF model in order to speed up the CPU time in integrating chemistry and the β-pdf of the chemical species to compute composition in the CFD domain. A parallel validation was performed both with constant-volume and Heavy-Duty Diesel Engine experiments, selecting similar operating conditions. In such way, both flame structure and heat release rate predictions are analyzed and the model capabilities with respect to its set-up and mesh structure are assessed.
IntroductionDevelopment of heavy duty Diesel engines is strongly affected by more and more demanding requirements for a contemporary reduction of both fuel consumption and pollutant emissions. To fulfill such objectives, a combination and integration of different technologies is necessary: efficient combustion systems, engine downsizing, new after-treatment devices and heat recovery [7,11,8,19]. In particular, fuel-air mixing and combustion processes must be investigated and optimized in detail since they both affect the quality of exhaust gases and energy conversion efficiency. Within this context, promising solutions appear to be the use of very high injection pressures (up to 3000 bar), further increase of full-load bmep, extension of engine operating range with advanced combustion 1