Simulation of flow phenomena at the helicopter rotor with high accuracy

点击翻译按钮获取中文摘要

The global goal of the project is the improvement of aerodynamic and aeroacoustic simulation capabilities for the helicopter. Advanced numerics - in this case Discontinuous Galerkin discretisations of high order - have already proven successful in simpler generic problems, and the aim is now to bring their value together with the existing possibilities of established flow solver like FLOWer. In consequence, a new generation of technologically advanced simulation tools will emerge. Taking into account the specific features of current and future computing technology we want to establish new levels of accuracy, reliability and efficiency. Therefore it is to be expected, that additional and more problematic parts of the helicopter flight envelope like highly loaded or sideward flight or even tail shake conditions can be simulated successfully and so the vision of a "virtual first flight" comes closer than ever.In a previous part of the project new DES (Detached Eddy Simulation) models have been successfully implemented and validated. Furthermore, run times of the then existing DG flow solver have been reduced by nearly an order of magnitude, while maintaining and even improving parallel efficiency up to thousands of processors. In the currently proposed project these efforts shall be continued towards the simulation of the complex flow phenomena at the helicopter rotor.In order to reach that goal, the physical model and its formulation have to be extended to cover the highly instationary flow features of dynamic stall, which happens in forward flight at the retreating blade and may significantly influence the entire flow field. On the other hand, it is inevitable to handle the elasticity of the rotor blades, where their deformation retroacts on the aerodynamics. Therefore the movement and deformation of the meshes have to be considered and - using the Chimera technology of overlapping grids - the flow state between them exchanged. Although turbulence models in general as well as grid deformation and Chimera transfers are well established technology, their usage in high order discretisations has to be observed very closely and possibly some modifications may turn up as appropriate. Finally, the success of our accuracy improvements is to be evaluated not only in global values as thrust and power consumption, but also in the acoustic post processing of descent flight encountering BVI (blade vortex interaction) events.To sum it up, such novel tools shall help to answer challenging technical questions dependably with sufficient accuracy, in order to significantly extend the capabilities of established methods in the medium term.