Ld
L.P.F. de Graaf
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
2 records found
1
This thesis presents the implementation, verification, and assessment of a partitioned CFD-CA coupling framework between the high-fidelity flow solver ENSOLV and the comprehensive rotorcraft analysis environment FLIGHTLAB, based on the delta-airloads methodology. The coupling was implemented in FLIGHTLAB through a dedicated AERODELTA component, which applies an aerodynamic correction, derived from the difference between the CFD and baseline lifting-line airloads, during the trim process, while the updated rotor motion and deformation are fed back into the subsequent CFD solution. The numerical interface was first verified to confirm consistent transfer of reference-frame information, blade motion and deformation, and aerodynamic loads and moments between the two solvers. The framework was then assessed using a test matrix of rigid and flexible Mach-scaled BO105 rotor configurations, covering hover, forward flight, and descending flight, with both steady and unsteady CFD.
For the rigid rotor, the coupled simulations converged stably across all flight conditions, and the predicted mean airloads and approximate blade-vortex interaction (BVI) locations compared reasonably well against HART-II experimental data, although the coarse CFD mesh limited the accuracy of the detailed sectional load fluctuations associated with BVI. In contrast, the flexible rotor simulations revealed a critical limitation: inconsistencies in the predicted aerodynamic pitching moment propagated into the elastic torsional response, producing an unstable aeroelastic feedback loop that caused chaotic blade flapping and feathering and, in several cases, prevented convergence of the trim solution.
These results indicate that the ENSOLV-FLIGHTLAB coupling is a verified and reliable framework for rigid rotor load prediction, but that its current implementation is not yet suitable for flexible aeroelastic rotor analysis. The dominant limitations are the coupling between the aerodynamic pitching moment and the structural torsional response, and insufficient resolution and preservation of the rotor wake. Recommendations to address these limitations are provided to guide future development of the framework. ...
For the rigid rotor, the coupled simulations converged stably across all flight conditions, and the predicted mean airloads and approximate blade-vortex interaction (BVI) locations compared reasonably well against HART-II experimental data, although the coarse CFD mesh limited the accuracy of the detailed sectional load fluctuations associated with BVI. In contrast, the flexible rotor simulations revealed a critical limitation: inconsistencies in the predicted aerodynamic pitching moment propagated into the elastic torsional response, producing an unstable aeroelastic feedback loop that caused chaotic blade flapping and feathering and, in several cases, prevented convergence of the trim solution.
These results indicate that the ENSOLV-FLIGHTLAB coupling is a verified and reliable framework for rigid rotor load prediction, but that its current implementation is not yet suitable for flexible aeroelastic rotor analysis. The dominant limitations are the coupling between the aerodynamic pitching moment and the structural torsional response, and insufficient resolution and preservation of the rotor wake. Recommendations to address these limitations are provided to guide future development of the framework. ...
This thesis presents the implementation, verification, and assessment of a partitioned CFD-CA coupling framework between the high-fidelity flow solver ENSOLV and the comprehensive rotorcraft analysis environment FLIGHTLAB, based on the delta-airloads methodology. The coupling was implemented in FLIGHTLAB through a dedicated AERODELTA component, which applies an aerodynamic correction, derived from the difference between the CFD and baseline lifting-line airloads, during the trim process, while the updated rotor motion and deformation are fed back into the subsequent CFD solution. The numerical interface was first verified to confirm consistent transfer of reference-frame information, blade motion and deformation, and aerodynamic loads and moments between the two solvers. The framework was then assessed using a test matrix of rigid and flexible Mach-scaled BO105 rotor configurations, covering hover, forward flight, and descending flight, with both steady and unsteady CFD.
For the rigid rotor, the coupled simulations converged stably across all flight conditions, and the predicted mean airloads and approximate blade-vortex interaction (BVI) locations compared reasonably well against HART-II experimental data, although the coarse CFD mesh limited the accuracy of the detailed sectional load fluctuations associated with BVI. In contrast, the flexible rotor simulations revealed a critical limitation: inconsistencies in the predicted aerodynamic pitching moment propagated into the elastic torsional response, producing an unstable aeroelastic feedback loop that caused chaotic blade flapping and feathering and, in several cases, prevented convergence of the trim solution.
These results indicate that the ENSOLV-FLIGHTLAB coupling is a verified and reliable framework for rigid rotor load prediction, but that its current implementation is not yet suitable for flexible aeroelastic rotor analysis. The dominant limitations are the coupling between the aerodynamic pitching moment and the structural torsional response, and insufficient resolution and preservation of the rotor wake. Recommendations to address these limitations are provided to guide future development of the framework.
For the rigid rotor, the coupled simulations converged stably across all flight conditions, and the predicted mean airloads and approximate blade-vortex interaction (BVI) locations compared reasonably well against HART-II experimental data, although the coarse CFD mesh limited the accuracy of the detailed sectional load fluctuations associated with BVI. In contrast, the flexible rotor simulations revealed a critical limitation: inconsistencies in the predicted aerodynamic pitching moment propagated into the elastic torsional response, producing an unstable aeroelastic feedback loop that caused chaotic blade flapping and feathering and, in several cases, prevented convergence of the trim solution.
These results indicate that the ENSOLV-FLIGHTLAB coupling is a verified and reliable framework for rigid rotor load prediction, but that its current implementation is not yet suitable for flexible aeroelastic rotor analysis. The dominant limitations are the coupling between the aerodynamic pitching moment and the structural torsional response, and insufficient resolution and preservation of the rotor wake. Recommendations to address these limitations are provided to guide future development of the framework.
Bachelor thesis
(2021)
-
L.P.F. de Graaf, S.D.S. Keemink, P.J. Koopdonk, B.J. Kroese, N.W.M. de Krom, M.S. Kukkola, Lars La Heij, M.N. Lengkeek, J.J. L'Ortije, C.J. Middelhoek, C.A. Dransfeld, M.T.H. Brown, S. Hamaza