Prediction Method of Snow Ingress Amount into the Engine Air Intake Duct Employing LES and Detailed Snow Accumulation Model

When a vehicle is driven in snowy conditions, if a proper air intake design is not adopted, the snow lifted by the leading vehicles may penetrate into the engine air intake, in case of large snow ingress amount, causing a power drop. The evaluation of such risk for the intake is carried out through climatic wind tunnel tests, which cannot be conducted at the early stage of vehicle development when the prototype vehicle does not exist. In order to study that risk prior to the prototype vehicle delivery, computational fluid dynamics (CFD) which predicts the snow ingress amount accurately was established with taking into account unsteady air flow and snow accumulation. Large Eddy Simulation (LES) was used to reproduce the unsteady flow field, leading to a good agreement of the flow downstream from the snow generator with the experimental one measured by Particle Image Velocimetry (PIV). As for the snow particle behavior model, the Lagrangian method was chosen. In order to study the snow tracking property, the velocity difference between air flow and snow particles was measured by PIV, whilst super-slow snow particle motions were visualized with a high speed camera. Through comparison between experimental results and customized CFD results, non-spherical drag coefficient and turbulent dispersion are found to be crucial to reproduce the tracking properties. Regarding the restitution coefficient at the walls, elastic-plastic deformation and surface friction were taken into account. In addition, an accumulation model was developed, in order to stop the particles and force them to accumulate at the wall when a certain criterion is satisfied. The accumulation model also takes into account the accumulation height limitation by recognizing the height of the flow path. The snow ingress amounts of CFD implementing those new techniques showed good correlation with experimental results for multiple vehicle categories.