Validation of CFD simulations for X-31 wind-tunnel models

Abstract

AbstractComputational Fluid Dynamics (CFD) has become an attractive method of choice in the design of many aerospace vehicles because of advances in numerical algorithms and convergence acceleration methods. However, the flow around an advanced fighter aircraft is complicated and usually unsteady due to the presence of vortex-dominated flows. The accuracy and predictability of conventional turbulence models for these applications may be questionable and therefore results obtained from these models must be validated and evaluated on the basis of experimental data from wind tunnels and/or flight tests. This work aims to validate CFD simulations of X-31 wind-tunnel models with and without a belly-mounted sting. The sting setup facilitates forced sinusoidal oscillations in one of three modes of: pitch, yaw, and roll. However, the results show that measured aerodynamic data are altered by the turbulent wake behind the sting, even at small angles of attack. The high angle-of-attack flow around the X-31 is also very complicated and unsteady due to canard and wing vortices. Therefore, validation of CFD models for predicting these complex flows can be a very challenging task. The X-31 wind-tunnel experiments were carried out in the German Dutch low-speed wind tunnel at Braunschweig and include aerodynamic force and moment measurement as well as span-wise pressure distributions at locations of 60% and 70% chord length. This data set is used to validate the Cobalt and Kestrel flow solvers and the results are similar and match quiet well with experiments for small to moderate angles of attack. The main discrepancies between CFD and measurements occur close to the wing tip, where leading-edge flaps are located.