Numerical Analysis of a Floating Wind Turbine - Global Load Effects in the Tower Structure

Abstract

The OO Star 10 MW is a semi-submersible floater that aims to support a 10 MW wind turbine. It consists of three outer columns with a shaft in the centre. The floater is designed to be built from concrete, and a steel tower is mounted on top of the concrete shaft providing a sufficient height for the wind turbine. Bottom part of the steel tower is sensitive to fatigue damage, even though steel thickness is close to both a practical and economical limit. A study of the effect on fatigue damage when increasing concrete shaft length is therefore performed.

Three cases with variable concrete shaft length are tested. The cases are named Case 0, Case 20 and Case 40. The number at each case represents increased concrete shaft length in metre compared to a base case. Small geometry modifications are done on Case 20 and Case 40 to maintain sufficient stability. Also the steel towers are slightly stiffer at comparable vertical positions for Case 20 and Case 40 compared to Case 0. Wind turbine is taken as the 10 MW DTU reference turbine, and hub height is kept constant for all three cases.

The problem is solved in time domain by use of the software SIMA. SIMA uses hydrodynamic forces calculated in the software Wadam and a turbulent wind field created by the software TurbSim. Calculations are performed for severe environmental conditions, and stress is calculated at several vertical positions on the steel tower. Stress concentration and safety factors are taken in accordance with regulations. Fatigue damage is estimated by use of a S-N curve.

A sensitivity study showed that first bending mode frequency for the steel tower is depending on the water plane stiffness, and that it is increased when the tower is mounted on a floater compared to being fixed. For current designs, this is found to be critical. All three cases have steel tower eigenfrequency close to the blade passing frequency (3p).

Contributions to fatigue is found to be wind, wave and 3p forces, where the latter one gives the largest contribution. Preliminary results show that an increased concrete shaft will improve fatigue life time, but due to the large 3p effects, none of the three designs have a sufficient life time. The fatigue life time is found to be 1.2 years for Case 0, 1.32 years for Case 20 and 1.39 years for Case 40. Fatigue life time for Case 40 is estimated to increase to 12.5 years if 3p effects are minimized. In addition to improving fatigue life time, an increase of concrete shaft with 40 metre will reduce the material cost by 5 mNOK.

Fatigue life time must be improved, and it is therefore recommended to modify the steel tower such that eigenfrequency is outside the 3p frequency range. Also using a steel tower with variable bending stiffness at different angular positions is found to be an option.

Additional design aspects were tested for Case 0 to be used in a screening process. Results indicate that mooring line tension, horizontal offset, pitch motion, acceleration at hub height and air gap is within requirements. Freeboard at outer column is found to be critical low, and the column risks being fully submerged during critical environmental conditions.