Combustion in a lean atmosphere diluted with a large amount of air can greatly improve fuel efficiency by reducing cooling loss [1, 2]. On the other hand, when air-fuel mixture in cylinder becomes lean, the turbulent combustion speed will decrease, resulting in problems such as the generation of unburned hydrocarbon (HC) and combustion instability [3, 4]. In order to solve these problems, it is important to increase the turbulence intensity and combustion speed [5, 6, 7, 8, 9, 10].
When designing combustion in cylinder by using Computational Fluid Dynamics (CFD), K-epsilon model is widely used for a turbulence model, and the calculated turbulence energy k or turbulence intensity u’ have been used as important indices of combustion velocity [11, 12]. However, it has been confirmed by measurements that the flow will conversely weaken near the top dead center and the combustion duration will become longer when the air flow in the cylinder is extremely strengthened by improved intake port. This phenomenon can be expressed by using Large Eddy Simulation (LES) instead of K-epsilon model as a turbulence model, so K-epsilon model cannot be used for future combustion development including lean burn, and LES is critical. This paper proposes a model-based development method for designing engine combustion by using LES and Kolmogorov scale.