1D and 3D CFD Investigation of Burning Process and Knock Occurrence in a Gasoline or CNG fuelled Two-Stroke SI Engine 2011-32-0526
The paper presents a combined experimental and numerical investigation of a small unit displacement two-stroke SI engine operated with gasoline and Natural Gas (CNG).
A detailed multi-cycle 3D-CFD analysis of the scavenging process is at first performed in order to accurately characterize the engine behavior in terms of scavenging patterns and efficiency. Detailed CFD analyses are used to accurately model the complex set of physical and chemical processes and to properly estimate the fluid-dynamic behavior of the engine, where boundary conditions are provided by a in-house developed 1D model of the whole engine. It is in fact widely recognized that for two-stroke crankcase scavenged, carbureted engines the scavenging patterns (fuel short-circuiting, residual gas distribution, pointwise lambda field, etc.) plays a fundamental role on both of engine performance and tailpipe emissions.
In order to assess the accuracy of the adopted numerical approach, comparisons between numerical forecasts and experimental measurements of instantaneous in-cylinder pressure history for steady-state operations of the engine are at first performed and shown in the paper.
Subsequently, results from 3D simulations are used to improve the scavenging characterization within the 1D model, where particular emphasis is now devoted to the investigation of the knock occurrence. In order to limit the computational cost of the simulations, the activity is at first carried out within the experimental and 1D modeling frameworks, where a quasi-dimensional combustion and knock model is used.
The 1D model is used to compute a numerical knock index which can be useful to address the tuning of the spark advance, given a prescribed and controlled percentage of knock released heat. At the end of the simulation process, the 1D knock index is qualitatively compared to results from full 3D knocking analyses for different in-cylinder compositions and spark timings.
The intrinsic knock-resistance of the CNG fuel is finally numerically exploited, through variations of both compression ratio and spark advance.