Master defence César Bourgeois

Supervisor: Anand Nararajan, DTU Wind Energy

Examiner:Henrik Stensgaard Toft, Siemens

The defence is a part of the Online Wind Energy Master

About the Online Wind Energy Master

The part-time master programme is scheduled to take two to four years, making it possible for participants to study while working. The participants come from all over the world and in August, the first participants graduate.

On Friday August 30, 2019, there will be a reception in B118 for the eight graduates. 

Abstract: Quantification of Variability in Offshore Wind Turbine Jacket Fatigue Damage from Environmental Variations

Offshore wind turbine jacket structures have fatigue damage as a critical design limit state, which needs to be evaluated using aero-elastic simulations to satisfy an intended operational lifetime.

This project used an existing HAWC2 model of the DTU 10MW reference turbine mounted on a jacket foundation. A post-treatment method was implemented in Python to compute fatigue damage for all tubular connections of the jacket model.

Some simulation files were provided for all wind speeds to be considered, which allowed the determination of the fatigue life of all welded joints, using Palmgren-Miner’s rule. A variation of the directions distribution allowed to determine the most favourable orientation of the jacket as compared to the dominant wind. It has also been found that the use of S-N curves for the jacket in seawater without cathodic protection, although greatly reducing fatigue life as compared to the case with cathodic protection, does not significantly affect which nodes are the most critical.

Some sensitivity cases were then simulated for two wind speeds, as follows, with in each case multiple simulations with yaw error and different seeds to increase the reliability of results:

  • Sensitivity to wind direction within direction bin, with three sets equally spaced by 10° being compared;
  • Sensitivity to wave spectrum, with Pierson-Moskowitz being compared to JONSWAP for the central directions set.

The study of the variability of fatigue damage due to the yaw error and different seeds confirmed that multiple simulations are necessary to obtain reliable results. The direction sensitivity cases displayed a significant variability of fatigue results, showing that multiple wind directions within each direction bin should be simulated and involved in the fatigue life determination. Further
studies would be necessary to determine how many would be sufficient.

The spectrum sensitivity cases showed that the selection of the wave spectrum has a significant influence on the evaluation of fatigue life, and that the wave spectrum should therefore be carefully selected based on the metocean data. In the process, it was also found that Rainflow counting functions in Python and in Matlab give different cycle counts and ranges, resulting in a significantly different fatigue damage. Further studies would be necessary to determine if one of them is erroneous.


Tue 27 Aug 19
13:00 - 15:00



DTU Risø Campus
Frederiksborgvej 399, bld. 118, meeting room Poul la Cour
4000 Roskilde