Supervisorer: Senior Scientist Mark Kelly., DTU Vindenergi - Professor Jakob Mann, DTU Vindenergi - Senior Scientist Anand Natarajan, DTU Vindenergi
Eksaminatorer: Senior Scientist Morten Nielsen, DTU Vindenergi - Senior Scientist/Visiting Professor Ervin Bossanti, DNV-GL/Uni. Bristol, Professor Wim Bierbooms, T.U. Delft
Titel: Extreme Wind Speed Ramps — Probabilistic Characterization for Turbine Loads
The focus of the project is a statistical representation of extreme events/gusts, which will account for the variability observed in real-world gusts. The spatial scale of these events is relevant for modern wind turbines, as the events are observed simultaneously at multiple heights, with height dependent delay. The velocity and temporal scales of the events are characterized by rise time and velocity jump (amplitude). These extreme gusts will be incorporated in simulations for wind turbine response. The simulated loads of these events will be analysed and compared with the extreme operating gust (EOG) load cases of the IEC standards. One goal of the project is to improve the EOG of the standards.
Project objectives
The objectives of this PhD project are:
- Investigate real measured extreme gusts, that are observed at multiple heights
- Characterize these events by rise time and velocity jump (amplitude)
- Estimate the 50-year return period contour for a joint probability distribution of rise time and velocity jump
- Simulate these events and investigate the wind turbine responses
- Find out if the loads simulated with the characteristics of the measured real events are comparable with the EOG load cases
Methods
To achieve a statistical representation of gusts, a ten-year time series of high-resolution wind speed measurements is analysed. The data is from Høvsøre, a costal site in western Jutland. The wind speed measurements are made at heights representative for ‘state of the art’ wind turbines. Extreme events are identified in the data for further processing and characterization.
A joint probability distribution of rise time and velocity jump is used to estimate a 50-year return period contour of the parameters. This is done with iForm (inverse first order reliability method) analysis.
The statistical gust characterization will be used to investigate wind turbine response. The load simulation will be performed with HAWC2 (horizontal axis wind turbine simulation code 2nd generation) and include the gust events as input.