Numerical Simulation of an Adaptive Wall in a Virtual Transonic Wind Tunnel

Abstract

COMPLEX adaptive wind tunnel walls provide a dilemma: because the wall adaptation can never fully compensate for all the interference of the walls, particularly separations on the side walls and gap flow for wall-mounted models, then computing the wind tunnel model in the wind tunnel environment is desirable. If the full experiment is computed using the experimentally measured wind tunnel wall shapes, then any errors in the computation of the model lift or blockage will be amplified by the wall that is now “wrong” for the computed aerodynamics [1]. The solution is to use the same method to compute wall shapes in the virtual wind tunnel as in the experiment, and this will lead to results that are 1:1 comparable. The same method can also be used for models in the design phase that do not have experimental wind tunnel wall contours associated with them. This note shows a method of providing the adaptive wall contours numerically, including comparisons with experiment. The primary objective of the presented investigation was to build up a process chain implementing the same wall adaptation algorithm as found in the experiment and combine it with a grid adaption tool to morph the computational fluid dynamics (CFD) mesh like the real test section. This process starts at straight walls without any deflection. Reference points are obtained from measurements in the Transonic Wind Tunnel Goettingen (DNW-TWG). A secondary objective was an assessment of the convergence behavior of the adaptation algorithm at transonic flow.