A parametric study on the influence of boundary conditions on the longitudinal pressure gradient in CFD simulations of an automotive wind tunnel
Artikel i vetenskaplig tidskrift, 2017
Computational fluid dynamics (CFD) is an important and extensively used tool for aerodynamic development in the vehicle industry today. Validation of virtual methods by comparison to wind tunnel experiments is a must because manufacturers aim to substitute physical tests on prototype vehicles with virtual simulations. An appropriate validation can be performed only if the wind tunnel geometry with representative boundary conditions is included in the numerical simulation, and if the flow of the empty wind tunnel is accurately predicted. One of the important flow parameters to predict is the longitudinal pressure distribution in the test section, which is dependent on both the wind tunnel geometry and the settings of the boundary layer control systems. This study investigates the effects of flow angularity at the inlet and different boundary layer control systems, namely, basic scoop suction, distributed suction, and moving belts, on the longitudinal pressure distribution in the full-scale aerodynamic wind tunnel of Volvo Cars using CFD and a systematic design of experiments approach. The study shows that the different suction systems used to reduce boundary layer thickness upstream of the vehicle have statistically significant effects on the longitudinal pressure distribution in the test section. However, the estimated drag difference induced on a typical vehicle by the difference in horizontal buoyancy between the tested settings is within the test-to-test uncertainty of the physical wind tunnel, thereby leading to the conclusion that force calculations in simulations are fairly insensitive to the tested parameters on the investigated intervals.
Boundary layer control
Design of experiments
CFD
Automotive wind tunnel