Search Results

The target of substantial CO₂ reductions in the spirit of the Kyoto Protocol as well as higher engine efficiency requirements has increased research efforts into hybridization of passenger cars. In the frame of this hybridization, there is a real need to develop small Internal Combustion Engines (ICE) with high power density. The two-stroke cycle can be a solution to reach these goals, allowing reductions of engine displacement, size and weight while maintaining good NVH, power and consumption levels. Reducing the number of cylinders, could also help reduce engine cost. Taking advantage of a strong interaction between the design office, 0D system simulations and 3D CFD computations, a specific methodology was set up in order to define a first optimized version of a two-stroke uniflow diesel engine. The main geometrical specifications (displacement, architecture) were chosen at the beginning of the study based on a bibliographic pre-study and the power target in terms.

A full 3D Large Eddy Simulation (LES) of a four-stroke, four-cylinder engine, performed with the AVBP-LES code, is presented in this paper. The drive for substantial CO₂ reductions in gasoline engines in the light of the global energy crisis and environmental awareness has increased research into gasoline engines and increased fuel efficiencies. Precise prediction of aerodynamics, mixing, combustion and pollutant formation are required so that CFD may actively contribute to the improvement/optimization of combustion chamber, intake/exhaust ducts and manifold shapes and volumes which all contribute to the global performance and efficiency of an engine. One way to improve engine efficiency is to reduce the cycle-to-cycle variability, through an improved understanding of their sources and effects. The conventional RANS approach does not allow addressing non-cyclic phenomena as it aims to compute the average cycle.

Dedicated experiments were performed in an optically-accessible, constant volume combustion vessel whose geometry and aerodynamic flow was representative of a pentroof SI engine combustion chamber. A detailed characterization of the flowfield was conducted in several near-wall regions where flame-wall interaction occurs using high-speed Particle Image Velocimetry (PIV). Simultaneous heat flux measurements were also performed at these same spatial locations. From a numerical point of view, current Reynolds Averaged Navier Stokes (RANS) or Large Eddy Simulation (LES) models take into account the effects of the wall on the flame however the effects of the turbulent flame-wall interaction on wall heat flux are not accounted for. Direct Numerical Simulations (DNS) of a 2D, premixed, steady-state V-flame were performed in order to aid the development and validation of a new model based on the flame surface density concept in order to take into account flame-wall interaction effects [1].