Theoretical and experimental aerodynamic analysis for high-speed ground vehicles

An improved understanding of the aerodynamics of high-speed ground vehicles can lead to significant reductions in the energy consumption required for propulsion, an increase of vehicle cruising speed, and an increase in the safety and comfort of passengers. To contribute to these goals, this thesis employs theoretical and experimental techniques to investigate the air flow around a proposed geometry for a high-speed electromagnetic suspension (EMS) train. Train motion at normal cruising speed in still air and in crosswind conditions are studied, considering aerodynamic forces and moments, the wake in the lee side of the train and the turbulent boundary layer development. The theoretical prediction work may be conveniently divided into two parts, for inviscid flow, and with viscous effects included. In the first, a numerical technique called the panel method has been applied to the representation of the body shape and the prediction of the potential flow and pressure distribution. Two computer programmes have been written, one for a single vehicle in the presence of the ground at different yaw angles, and the second for application to two body problems, e.g. a train passing a railway station or a train passing the central part of another train. Both programmes have been developed in fully three-dimensional form, but are currently based purely on the source distribution method. This limits the applicability of the method, in particular to small angles of yaw, but useful results are still obtainable. In the second part of the theoretical prediction work, two methods based on the momentum integral equations for three-dimensional boundary layer flow have been developed for use with the aforementioned potential flow analysis; these predict the development of the three-dimensional turbulent boundary layer (i) on the central section (for the analysis of crosswind conditions) and (ii) on the nose of the train. The primary interest of the experimental programme was to provide qualitative and quantitative results for comparison with the theoretical predictions as well as to give insight into the flow behaviour around the train. The experimental tests also provided the first results for the influence of both stationary and moving ground planes on the EMS train. Extensive wind tunnel tests were performed on four purpose-made models of the high-speed train to measure aerodynamic forces, moments and pressures to establish ground effect characteristics. The experimental results demonstrated the importance of ground clearance. Flow visualisation showed that the wake vortices were both stronger and larger in the presence of a ground. At small yaw angles ground clearance had little effect, but as yaw increased, larger ground clearance led for example to substantial increase in lift and side force coefficients. The wind tunnel tests also identified the differences between a moving and a fixed ground plane. The measured data showed that the type of ground simulation was significant only in the separated region. A comparison of the results predicted using potential flow theory for an EMS train model and the corresponding results from wind tunnel tests indicated good agreement in regions where the flow is attached. For small yaw angles, not more than 15° , predicted pressure distributions reproduced measured behaviour. For greater angles, the shed vorticity (associated with flow separation) has a strong effect on the surface pressure field and this would have to be introduced into the panel method to improve prediction. The turbulent boundary layer calculations for the train in a crosswind condition showed that the momentum thickness along the crosswind surface distance co-ordinate increased slowly at the beginning of the development of the boundary layer but then increased sharply at the side top roof on the lee side. The sharp increase is believed to indicate a tendency for flow separation as the solution procedure exhibits signs of failure in this region. Suggestions are made in the thesis for ways of improving both this and other aspects of the theoretical approach.