Research Papers: Ocean Renewable Energy

Frequency Versus Time Domain Fatigue Analysis of a Semisubmersible Wind Turbine Tower

[+] Author and Article Information
Marit I. Kvittem

Centre for Ships and Ocean Structures;
Norwegian Research Centre for
Offshore Wind Technology,
Trondheim N-7491, Norway
e-mail: marit.irene.kvittem@ntnu.no

Torgeir Moan

Centre for Ships and Ocean Structures;
Norwegian Research Centre for
Offshore Wind Technology;
Centre for Autonomous Marine
Operations and Systems,
Trondheim N-7491, Norway
e-mail: torgeir.moan@ntnu.no

1Corresponding author.

Contributed by the Ocean, Offshore, and Arctic Engineering Division of ASME for publication in the JOURNAL OF OFFSHORE MECHANICS AND ARCTIC ENGINEERING. Manuscript received March 23, 2014; final manuscript received August 10, 2014; published online September 25, 2014. Assoc. Editor: Ron Riggs.

J. Offshore Mech. Arct. Eng 137(1), 011901 (Sep 25, 2014) (11 pages) Paper No: OMAE-14-1034; doi: 10.1115/1.4028340 History: Received March 23, 2014; Revised August 10, 2014

The current paper deals with a study of a semisubmersible wind turbine (WT), where short-term tower base bending moments and tower fatigue damage were estimated by a frequency domain (FD) method. Both a rigid structure assumption and a generalized degree-of-freedom (DOF) model for including the first flexible mode of the turbine tower were investigated. First, response to wind and wave loads was considered separately, then superposition was used to find the response to combined wind and wave loading. The bending moments and fatigue damage obtained by these methods were compared to results from a fully coupled, nonlinear time domain (TD) analysis. In this study a three column, catenary moored semisubmersible with the NREL 5 MW turbine mounted on one of the columns was modeled. The model was inspired by the WindFloat concept. The TD simulation tool used was Simo-Riflex-AeroDyn from Marintek and CeSOS. The FD method gave a good representation of the tower base bending moment histories for wave-only analyses, for the moderate sea states considered in these analyses. With the assumption that the structure is completely rigid, bending moments were underestimated, but including excitation of the elastic tower and blades, improved the results. The wind-induced low-frequency bending moments were not captured very well, which presumably comes from a combination of nonlinear effects being lost in the linearization of the thrust force and that the aerodynamic damping model was derived for a fixed turbine. Nevertheless, standard deviations of the bending moments were still reasonable. The FD model captured the combined wind and wave analyses quite well when a generalized coordinates model for wind excitation of the first bending mode of the turbine was included. The FD fatigue damage predictions were underestimated by 0–60%, corresponding to discrepancies in standard deviations of stress in the order of 0–20%.

Copyright © 2015 by ASME
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Fig. 1

Motion response spectra for surge and sway under 10 m/s turbulent wind, with and without loads from irregular waves with Hs = 5 m and Tp = 14 s. (a) Modeshape of the first fore-aft bending mode of the turbine and tower (ψ). (b) Fixed (left) and floating (right) turbine in a generalized single-degree-of-freedom (SDOF) system.

Grahic Jump Location
Fig. 2

Generalized SDOF system for the turbine

Grahic Jump Location
Fig. 3

Frequency-response function from generalized properties of the first fore-aft bending mode of the turbine. The shown function includes 2% structural damping and no aerodynamic damping. (a) Semisubmersible WT. (b) Top view: The coordinate system z = 0 is in the mean water line. Wind and long crested waves are assumed to follow the x-axis.

Grahic Jump Location
Fig. 4

Semisubmersible WT

Grahic Jump Location
Fig. 5

Motions, accelerations, and bending moments for wave-only cases (mean value over ten samples). Errors are calculated as FD estimate relative to TD simulations, i.e., negative errors means that the FD estimate is smaller than the TD result. Results are shown for two FD methods: FD1 and FD2.

Grahic Jump Location
Fig. 6

Bending moment spectra for wave loading. FD estimates with (FD1) and without (FD2) dynamic amplification due to flexibility of the tower and blades are compared to TD simulations.

Grahic Jump Location
Fig. 7

Wind-only cases. Errors in FD estimate compared to simulations (mean value over ten samples). Results for three FD methods: FD1-FD3.

Grahic Jump Location
Fig. 8

High-frequency part of bending moment spectra for wind loading. FD with dynamic amplification due to flexibility of the turbine (FD3) is compared to TD simulations.

Grahic Jump Location
Fig. 9

Errors in short-term fatigue damage for wave-only TD simulations compared to the three FD methods FD1-FD2

Grahic Jump Location
Fig. 10

Errors in short-term fatigue damage for wind-only TD simulations compared to the three FD methods FD1-FD3

Grahic Jump Location
Fig. 11

Combined wind and waves. Errors in superimposed FD estimates compared to TD simulations (mean value over ten samples). Results are shown for FD1–FD3.

Grahic Jump Location
Fig. 12

Errors in short-term fatigue damage from TD simulations compared to superimposed wind and wave analyses using FD1-FD3

Grahic Jump Location
Fig. 13

Wind-frequency part of tower base bending moment spectra for FD and nonlinear simulations for full and reduced aerodynamic damping




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In