NUMERICAL STUDY OF HYPERSONIC FLUID-STRUCTURE INTERACTION ON A CANTILEVERED PLATE WITH SHOCK IMPINGEMENT

Abstract

Aeroelastic vibration of compliant wing panels and control surfaces is a major design concern in hypersonic vehicles. Impingement of shock waves of varying intensity also adds to the aeroelastic effect and vibration. UNSW Canberra’s Hypersonic Multibody Aeroelastic eXperiment (HyMAX) serves as a benchmark test case for Fluid-Structure Interaction (FSI) in hypersonic flows. This study carries out a numerical study of the HyMAX experimental setup at a flow deflection angle of 10° using both low-fidelity modeling (LFM) and high-fidelity modeling (HFM) approaches. In the LFM approach, an analytical SE-based Piston Theory and a CFD-Enriched Piston Theory were used. And, a two-way partitioned approach using OpenFOAM, Calculix, and PreCICE was used for the HFM. The cantilevered plate deformed around the first mode. Peak pressure variation and the trailing edge displacement history showed similar nature indicating that the FSI phenomenon was dominated mostly by the local pressure changes over the plate. The peak pressure approximations results of both the LFM and the HFM highlighted the quasi-steady nature of the problem. The peak pressure value predicted by the viscous CFD is around 4% (300 Pa) more when compared to the inviscid CFD result and the maximum trailing edge deflection (3.96 mm) predicted by the viscous FSI is about 4 % higher than the value predicted by the inviscid FSI (3.80 mm), which can be attributed to Shock-Wave Boundary Layer Interaction (SWBLI) phenomenon, leading edge shock wave, and other viscous effects. The Shock-Expansion based Piston Theory (PT) predicted the maximum trailing edge displacement with about 8 % error in HyFoil HH (with no shock impingements) but with about 14 % error in HyMAx (with shock impingement). In the case of the shock-impingement, CFD Enriched PT was found to make a better prediction with only about an 6.82% error with respect to the viscous FSI result. The computation time required for a flow duration of 200 ms for CFD Enriched PT was very short only about 4 hours as compared to about 24 hours for the inviscid FSI and 480 hours for viscous FSI. Hence, CFD Enriched PT can be an effective tool for preliminary aeroelastic analysis