Calibration and assessment of the capability of the predictive combustion model SI-Turb

In the automotive field, the aim of these days is to reduce CO2 emissions of vehicles: the most promising solution for what concerns Internal Combustion Engines (ICE) and in particular Spark Ignition (SI) engines is the adoption of downsized turbocharged petrol engines. However this technology together with the continuously increasing levels of boost to achieve the desired performance requires a precise description of the combustion pro- cess so that the effects of parameters such as spark timing, EGR rate, compression ratio changes or knock level can be well evaluated. The development and the validation of reliable methods able to describe the condition inside the cylinders and the evolution of the combustion process have therefore acquired great importance in the most recent times. The common practice to properly evaluate the flow motion in the cylinder and the combustion process is to run 3D Computation Fluid Dynamics (CFD) simulations which are accurate but also very time-consuming. As a consequence, the CFD analysis cannot always be used during the design and calibration phases of a new engine development. Since during these steps fast and reliable answers are necessary, innovative 0/1 Dimensional (0/1D) methods for the prediction of in-cylinder flow conditions and the combustion evolution have been investigated. One of these approaches is the 0D phenomenological turbulence model, built inside the commercial code GT-Suite, which will be used and calibrated in this project to evaluate the flow conditions inside the cylinder. Then the turbulence model will be coupled with the predictive combustion model from GT-Suite and its predictive capability will be assessed. An interesting analysis about the potential benefits of the use of this predictive tool when moving to a new engine will be carried out: the goal is to evaluate what happens and which are the results when using the turbulence and combustion calibration from a previous engine for a new one, which presents a slightly changed geometry. As final step, a predictive model will be calibrated from scratch and the comparison with the calibration of the previous engine will be evaluated.