Wind turbine model-test method for achieving similarity of both model- and full-scale thrusts and torques
Hao, H., Liao, K., Ma, Q. ORCID: 0000-0001-5579-6454 , Zheng, X., Sun, H. & Khayyer, A. (2023). Wind turbine model-test method for achieving similarity of both model- and full-scale thrusts and torques. Applied Ocean Research, 130, article number 103444. doi: 10.1016/j.apor.2022.103444
Abstract
For model tests of a floating offshore wind turbine (FOWT) system, a great challenge is how to model the interaction between the wind turbine and floating platform with correctly-scaled aerodynamic and hydrodynamic loads because of a well-known contradiction between Reynolds and Froude scaling. Several approaches have been proposed in the literature to tackle the challenge but none of them can correctly and simultaneously model the scaled thrust and torque, and so the interaction between turbine and platform. This paper will present a new model-test method for achieving similarity of both thrust and torque. This is achieved by redesigning the model blades with keeping the blade twist angle same as that of the full-scale turbine and by adjusting the pitch angle of blades and rotational speed of wind turbine in model tests. Numerical simulations and wind tunnel model tests are carried out to validate the present method. Both numerical results and experimental data show that the present method can realize the similarity of thrust and torque simultaneously, and thus, making it possible to study full interaction between turbine and floating platform by physical testing in wind-wave basins.
Publication Type: | Article |
---|---|
Additional Information: | © 2022. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/ |
Publisher Keywords: | Wind turbine model test, Scaled thrust and torque, Reynolds and Froude similarity, FOWT |
Subjects: | T Technology > TA Engineering (General). Civil engineering (General) |
Departments: | School of Science & Technology > Engineering |
Available under License Creative Commons Attribution Non-commercial No Derivatives.
Download (1MB) | Preview
Export
Downloads
Downloads per month over past year