A Coupled Unstructured-Adaptive Cartesian CFD Method for Hover Prediction

An innovative computational fluid dynamics (CFD) approach is employed to predict the aerodynamic performance of hovering rotors. Two different CFD solvers are applied in different parts of the computational domain: a body-fitted unstructured solver near the blade surface to capture complex geometry and viscous effects, and a high-order block-structured Cartesian solver away from the blade to capture the wake. The Cartesian solver applies Adaptive Mesh Refinement (AMR) to resolve tip vortices. Results are demonstrated on the TRAM isolated rotor. The results show the approach is able to achieve aerodynamic figure of merit performance predictions to within 2% of experiment on a 64-core distributed parallel Linux cluster. Solution driven AMR is effective for resolving the vortex wake at significantly reduced computational cost over fixed-grid calculations with similar resolution.