NUMERICAL SIMULATION AND VALIDATION OF HELICOPTER BLADE-VORTEX INTERACTION USING COUPLED CFD/CSD AND THREE LEVELS OF AERODYNAMIC MODELING

dc.contributor.advisorBaeder, James Den_US
dc.contributor.authorAmiraux, Mathieuen_US
dc.contributor.departmentAerospace Engineeringen_US
dc.contributor.publisherDigital Repository at the University of Marylanden_US
dc.contributor.publisherUniversity of Maryland (College Park, Md.)en_US
dc.date.accessioned2014-10-17T05:31:01Z
dc.date.available2014-10-17T05:31:01Z
dc.date.issued2014en_US
dc.description.abstractRotorcraft Blade-Vortex Interaction (BVI) remains one of the most challenging flow phenomenon to simulate numerically. Over the past decade, the HART-II rotor test and its extensive experimental dataset has been a major database for validation of CFD codes. Its strong BVI signature, with high levels of intrusive noise and vibrations, makes it a difficult test for computational methods. The main challenge is to accurately capture and preserve the vortices which interact with the rotor, while predicting correct blade deformations and loading. This doctoral dissertation presents the application of a coupled CFD/CSD methodology to the problem of helicopter BVI and compares three levels of fidelity for aerodynamic modeling: a hybrid lifting-line/free-wake (wake coupling) method, with modified compressible unsteady model; a hybrid URANS/free-wake method; and a URANS-based wake capturing method, using multiple overset meshes to capture the entire flow field. To further increase numerical correlation, three helicopter fuselage models are implemented in the framework. The first is a high resolution 3D GPU panel code; the second is an immersed boundary based method, with 3D elliptic grid adaption; the last one uses a body-fitted, curvilinear fuselage mesh. The main contribution of this work is the implementation and systematic comparison of multiple numerical methods to perform BVI modeling. The trade-offs between solution accuracy and computational cost are highlighted for the different approaches. Various improvements have been made to each code to enhance physical fidelity, while advanced technologies, such as GPU computing, have been employed to increase efficiency. The resulting numerical setup covers all aspects of the simulation creating a truly multi-fidelity and multi-physics framework. Overall, the wake capturing approach showed the best BVI phasing correlation and good blade deflection predictions, with slightly under-predicted aerodynamic loading magnitudes. However, it proved to be much more expensive than the other two methods. Wake coupling with RANS solver had very good loading magnitude predictions, and therefore good acoustic intensities, with acceptable computational cost. The lifting-line based technique often had over-predicted aerodynamic levels, due to the degree of empiricism of the model, but its very short run-times, thanks to GPU technology, makes it a very attractive approach.en_US
dc.identifierhttps://doi.org/10.13016/M2P318
dc.identifier.urihttp://hdl.handle.net/1903/15930
dc.language.isoenen_US
dc.subject.pqcontrolledAerospace engineeringen_US
dc.subject.pqcontrolledAcousticsen_US
dc.subject.pquncontrolledacousticsen_US
dc.subject.pquncontrolledaerodynamicsen_US
dc.subject.pquncontrolledblade vortex interactionen_US
dc.subject.pquncontrolledCFDen_US
dc.subject.pquncontrolledhelicopteren_US
dc.titleNUMERICAL SIMULATION AND VALIDATION OF HELICOPTER BLADE-VORTEX INTERACTION USING COUPLED CFD/CSD AND THREE LEVELS OF AERODYNAMIC MODELINGen_US
dc.typeDissertationen_US

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Amiraux_umd_0117E_15330.pdf
Size:
39.01 MB
Format:
Adobe Portable Document Format