Unsteady Transonic Wing-Pylon-Nacelle Interference - Investigation and CFD Validation by the WIONA Test

Abstract

Wing nacelle interference can play a significant role in transport aircraft aeroelasticity. Particularly, in transonic flows a remarkable impact of interference induced unsteady airloads on the flutter behavior has been observed. Recent wind tunnel tests on the generic WIONA (Wing with Oscillating Nacelle) configuration conducted in the DNW-TWG confirm the crucial influence of the transonic channel flow in the local interference region. The WIONA model is composed of a flow-through nacelle mounted via a pylon on an unswept rectangular wing. The model geometry is based on the DLR-F6 configuration. The wing was derived from a supercritical airfoil of 9.2% thickness. The model possesses three degrees of freedom, namely nacelle roll, nacelle yaw and rigid body pitch, which were excited harmonically during the experiment. Pressure transducers were arranged in streamwise sections in and out of the channel region and on the nacelle. The particle image velocimetry technique (PIV) has been applied to investigate the flow field in the channel region. With PIV recordings from subsequent cycles a resolution of 32 equally spaced samples per period was achieved, which clearly resolve the shock oscillations in the channel flow. The TAU code is applied for steady and unsteady RANS computations on a hybrid grid with 1.9 million vertices around the WIONA geometry. For convenience, wind tunnel walls were replaced by symmetry and farfield boundary conditions respectively. Steady computations were performed adopting both the 1-equation turbulence models of Spalart-Allmaras (S-A) and the 2-equation Linearized Explicit Algebraic (LEA) k-omega turbulence model. Despite differing topologies in separated regions no significant differences were encountered. The flow conditions for all cases were chosen Ma=0.82, AoA=-0.6 deg, Re=2.2 Mio., where the computed steady transonic flow field depicts shocks on both upper wing surface and in the interference channel between the lower wing surface and nacelle. Mild shock induced separation occurs. Despite small discrepancies due to non-smooth model surface the agreement with the test results is very good. For unsteady computations the forced oscillations of the WIONA model are reproduced either by rigid body rotation of the whole grid or grid deformation in case of relative motion of the nacelle w.r.t. the wing. Unsteady results for forced harmonic motion were achieved with the S-A model alone. The computed harmonic unsteady pressure distribution is in good agreement with the test results. The computed unsteady flow fields correlate well with PIV measurements.