One- and Three-Dimensional CFD Study of an Onboard Refuelling Vapour Recovery System

Abstract

The human impact of the environmental development is without doubt one of the biggest challenges in modern society. The influence from cars is highly debated and especially the impact from fossil fuels like gasoline. At the Fuel System Department at Volvo Cars in Torslanda, the refuelling process of the cars is tested with physical experiments on a daily basis. A system called Onboard Refuelling Vapour Recovery (ORVR) works within the fuel system to prevent hydrocarbons and other harmful particles to escape during refuelling. Countries set different limitations on what quantity of particles that may be emitted and it is sometimes difficult to explain what causes variations in the results of the experiments. Computational fluid dynamics (CFD) is used in order to get a more detailed view of the ORVR system, which could be a great complement to physical experiments. Decisive parameters for the flow field could then be evaluated and possible sources of errors of the testing could be discovered. In this thesis the ORVR system is studied through both one- and three-dimensional computational fluid dynamics. The studied refuelling system corresponds to the fuel system for a Volvo XC90 certified for the US market, with a 71 liters fuel tank. The three dimensional case is simulated in SIEMENS Star-CCM+. The simulation uses the Volume of Fluid (VOF) model for surface tracking of the multiphase problem. An evaporation model is implemented to deal with the mass transfer between the phases. The accuracy of the rate of evaporation of the gasoline is studied by both experimental and computational testing. Since the three-dimensional simulations are computationally expensive and time consuming, a one-dimensional study is also performed, where the software GT-Suite is used. The geometry is discretized from the three-dimensional CAD models to a one-dimensional system using softwares like ANSA, GT SpaceClaim and GEM3D before being imported into GT-Suite. The results from the different approaches are then evaluated and compared with experimental data. Possible advantages and drawbacks with the respective methods are analyzed and presented in a perspicuous way to the company.