Floating offshore wind turbines in deep waters offer significant advantages to onshore and near-shore wind turbines. However, due to the motion of floating platforms in response to wind and wave loading, the aerodynamics are substantially more complex. Traditional aerodynamic models and design codes do not adequately account for the floating platform dynamics to assess its effect on turbine loads and performance. Turbines must therefore be over designed due to loading uncertainty and are not fully optimized for their operating conditions. Previous research at the University of Massachusetts, Amherst developed the Wake Induced Dynamics Simulator, or WInDS, a free vortex wake model of wind turbines that explicitly includes the velocity components from platform motion. WInDS rigorously accounts for the unsteady interactions between the wind turbine rotor and its wake, however, as a potential flow model, the unsteady viscous response in the blade boundary layer is neglected. To address this concern, this thesis presents the development of a Leishman-Beddoes dynamic stall model integrated into WInDS. The stand-alone dynamic stall model was validated against two-dimensional unsteady data from the OSU pitch oscillation experiments and the coupled WInDS model was validated against three-dimensional data from NREL’s UAE Phase VI campaign. WInDS with dynamic stall shows substantial improvements in load predictions for both steady and unsteady conditions over the base version of WInDS. Furthermore, use of WInDS with the dynamic stall model should provide the necessary aerodynamic model fidelity for future research and design work on floating offshore wind turbines.
| Identifer | oai:union.ndltd.org:UMASS/oai:scholarworks.umass.edu:masters_theses_2-1116 |
| Date | 07 November 2014 |
| Creators | Gaertner, Evan M |
| Publisher | ScholarWorks@UMass Amherst |
| Source Sets | University of Massachusetts, Amherst |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Masters Theses |
Page generated in 0.0024 seconds