Impact of aerodynamic modeling on seakeeping performance of a floating horizontal axis wind turbine. Issue 8 (3rd April 2019)

• Record Type: Journal Article
• Authors: Leroy, Vincent; Gilloteaux, Jean‐Christophe; Lynch, Mattias; Babarit, Aurélien; Ferrant, Pierre
• Is Part Of: Wind energy. Volume 22:Issue 8(2019)
• Page Start: 1019
• Page End: 1033
• Publication Date: 2019-04-03
• Subjects: aerodynamic modeling -- coupled simulation -- dynamic response analysis -- floating wind turbines; Wind power -- Periodicals; 621.312136
• DOI: 10.1002/we.2337
• Languages: English
• ISSNs: 1095-4244

Abstract

Abstract: Over the last decade, several coupled simulation tools have been developed in order to design and optimize floating wind turbines (FWTs). In most of these tools, the aerodynamic modeling is based on quasi‐steady aerodynamic models such as the blade element momentum (BEM). It may not be accurate enough for FWTs as the motion of the platform induces highly unsteady phenomena around the rotor. To address this issue, a new design tool has been developed coupling a seakeeping solver with an unsteady aerodynamic solver based on the free vortex wake (FVW) theory. This tool is here compared with the reference code FAST, which is based on the BEM theory in order to characterize the impact of the aerodynamic model on the seakeeping of a floating horizontal axis wind turbine (HAWT). Aerodynamic solvers are compared for the case of the free floating NREL 5MW HAWT supported by the OC3Hywind SPAR. Differences obtained between the models have been analyzed through a study of the aerodynamic loads acting on the same turbine in imposed harmonic surge and pitch motions. This provides a better understanding of the intrinsic differences between the quasi‐steady and unsteady aerodynamic solvers. The study shows that differences can be observed between the three aerodynamic solvers, especially at high tip speed ratio (TSR) for which unsteady aerodynamic phenomena and complex wake dynamics occur. Observed discrepancies in the predictions of the FWT dynamic response can raise issues when designing such a system with a state‐of‐the‐art design tool.