Wading – Evaluation of SPH-based simulations versus traditional Finite Volume CFD
Examensarbete för masterexamen
Going through deep water is a driving condition connected with many potential problems for ground transportation vehicles. It exerts additional forces on the external parts of the vehicle not present during normal driving. In combustion vehicles, it may hinder the required constant supply of air to the engine, or it might even ﬂood. Simulations to study this driving condition are normally conducted at Volvo Car Corporation (VCC) using a traditional Finite Volume Method (FVM) namely, Volume Of Fluid (VOF), either conducted in StarCCM+ or OpenFOAM. These simulations are performed in an Eulerian framework requiring a computational mesh. This requirement aggravates the replication of real life conditions. A new alternative to the current method has been developed, Smooth Particle Hydrodynamics (SPH). SPH is a particle method conducted solely in a Lagrangian framework removing the need of a computational volume mesh. The SPH method has many advantages over the one used today, most importantly having the potential of reducing simulation times drastically. The purpose of this master thesis is to investigate the potential use of a SPH solver as a substitute or complement to VOF in wading simulations at VCC. Due to the versatility of the method, a setup that mimics real world phenomena and kinematics is sought. The methodology includes methods to do visual assessments together with force measurements, water ingress, as well as performance in terms of computational cost. SPH has proven to be a reliable framework that is both capable of predicting phenomena in a frame of reference both agreeing with the existing FVM and real-world conditions. In terms of computational time, great reductions can be achieved through an adaptive domain size. The developed method is not complete, and there are still room for future improvement. At this date, SPH has proven to be a competitive alternative to VOF.
SPH , VOF , Wading , FVM , Evaluation , PreonLab , Simulation , Physical testing