|Damping estimation of offshore wind turbines using state-of-the art operational modal analysis techniques|
El-Kafafy, M.; Devriendt, C.; De Sitter, G.; De Troyer, T.; Guillaume, P. (2012). Damping estimation of offshore wind turbines using state-of-the art operational modal analysis techniques, in: Proceedings of ISMA2012-USD2012. pp. 2647-2662
In: (2012). Proceedings of ISMA2012-USD2012. KU Leuven: Belgium. , more
|Authors|| || Top |
In this paper, one existing modal analysis tool together with recently proposed modal parameter estimation approaches [1, 2] will be investigated with respect to their applicability to identify the damping values of an offshore wind turbine and compared with the traditional log-decrement approach. The experimental data has been obtained during a measurement campaign on an offshore wind turbine in the Belgian North Sea in the framework of the Flemish funded off shore wind infrastructure project. Real damping ratios are very difficult to predict by numerical tools and therefore measurements on existing offshore wind turbines are crucial to verify the design assumptions in estimating the lifetime of an offshore wind turbine. It will be shown that damping ratios can directly be obtained from vibrations of the tower under ambient excitation from wave and wind loading. The results will be compared with the approach that is used nowadays to determine the damping of offshore wind turbines: an overspeed emergency stop followed by a logarithmic decrement analysis. An emergency stop, however, reduces the remaining lifetime of the wind turbine. The frequency-domain OMA techniques, presented in this paper, do not require emergency stops, which is clearly an economic advantage and a more practical approach. The advanced modal analysis tools, which will be investigated, include the poly-reference Least Squares Complex Frequencydomain estimator (pLSCF) - commercially known as PolyMAX - estimator and two newly proposed modal estimation approaches [1, 2]. The newly proposed modal estimation approaches are a combination of the maximum likelihood estimator (MLE) and the pLSCF estimator. The advantage of these approaches is that they keep the benefits of the pLSCF estimator (e.g. very clear stabilization chart) while adding MLE features like improved estimates and proper handling of the measurement uncertainties.