Heavy goods vehicles (HGVs) are extensively used in the UK transport sector, and an ever-increasing volume of traffic on the roads has led to an increase in total fuel consumption. Over the years, different strategies have been employed in the design of HGVs including shape changes or use of add-on devices at various positions. More recently, a popular option has been to change the traditional rectangular longitudinal sectional shape of typical HGVs and to modify the over-body shape with curvature. However, there is no clear guidance on how to do this effectively and the variety of approaches seen on the roads suggests that there are no universally accepted design rules. Furthermore, most aerodynamics studies in the literature focus primarily on minimising drag; far too little attention has been paid to enhance stability, handling and safety operation due to shape changes, especially due to side winds or gusts. The purpose of this work is to address these issues using Computational Fluid Dynamics (CFD) simulations with aerodynamic shape optimisation.
A simplified generic HGV, the Ground Transportation System (GTS), is investigated in this thesis because it closely resembles a typical European HGV and there is an abundance of high-quality experimental data available. A rigorous verification and validation study using the commercial CFD package, ANSYS Fluent, found agreement to within 15% between experiments and the numerical results. Following this, an aerodynamic shape optimisation study was formulated. The shape of the over-body profile was parameterised using a 3rd order polynomial with three design variables determining the shape of the vehicle, namely, the height of the base region, the angle of the trailing edge of the roof and the radius of curvature of the longitudinal roof edges. Using a 125-point Design of Experiments (DoE) approach, high-fidelity CFD simulations were carried out for yaw angles of 0 ͦ, 5 ͦ, 6 ͦ and 8 ͦ. These angles were determined from analysis of a typical Leeds-London-Leeds motorway journey.
Moving Least Squares (MLS) metamodels were used in conjunction with Genetic Algorithms (GAs) and gradient-based techniques to identify optimum HGV designs. Results show that a minimum-drag design can accomplish drag reduction of around 40% compared to a baseline (rectangular) vehicle, however, weathercock stability is 11% poorer at a slip angle of 5 ͦ increasing to 35%, at 8 ͦ. The best stability design was found to achieve a 33% drag improvement, compared to the baseline and weathercock stability is between 8% and 25% worse, for 5 ͦ and 8 ͦ of yaw, respectively. However, the reduction in weathercock stability, compared to the baseline design, naturally leads to better directional stability and road holding capability. The height of the base of the vehicle is the dominant design parameter with small values leading to improved drag but large values inducing greater weathercock stability due to increased rear side area. From this, design guidelines are proposed.
