A 0D phenomenological turbulence model, based on the K-k and k- ɛ approaches, was coupled with a predictive turbulent combustion model using the commercial code GT-Suite, and its predictive capabilities were assessed for a downsized turbocharged SI engine.
Differently from the 3D-CFD approach which is typically utilized to describe the evolution of the in-cylinder flow field, and which has very high computational requirements, the 0D phenomenological approach adopted in this work gives the opportunity to predict the evolution of the in-cylinder charge motion and the subsequent combustion process by means of a turbulent combustion model, with a significantly reduced computational effort, thus paving the way for the simulation of the whole engine operating map.
Moreover, a procedure has been adopted to calibrate the turbulent combustion model parameters by means of a Design of Experiments (DoE) coupled with Genetic Algorithm (GA) approach, in order to predict the burn rate at various engine operating points.
Finally, a detailed validation process, based on an extensive experimental data set, was carried out concerning the predicted burn rates and the in-cylinder pressure traces for several engine operating points, including load, speed and spark timing sweeps, achieving a satisfactory agreement and thus confirming the reliability of the proposed approach.