A comparison of the influence of using empirical or mathematically pre-defined wave energy spectra for tower base bending fatigue calculations

Abstract

In the structural design of Floating Offshore Wind Turbines (FOWT), fatigue plays a critical role in determining the final design of the system. The fatigue loads are the result of combined aerodynamic and hydrodynamic forces acting on the elastic structure. The industry standard approach for assessing the fatigue loads involves grouping the environmental conditions into bins. These conditions include wind speed and direction, wave height, period and direction and the sea state spectral shape. In shallow seas with limited fetch the JONSWAP spectrum, or a JONSWAP-derived spectrum, is normally fitted to the site conditions and used, which also includes a peak enhancement factor (GAMMA) in a range defined by the significant wave height and peak period. However, this adjustment is sensitive to the parameter fitting process, while the vital Peak Enhancement Factor (gamma) parameter is commonly chosen as an arbitrary empirical value in the given range. In this paper, we examine how the calculation of bending fatigue of the tower base of the IEA 15MW open source turbine supported by the UMaine VolturnUS semi-submersible is influenced by either the use of empirical spectra (measured or simulated for the specific site) against pre-described site-fitted formulas for spectral shape, and the use of different spectra per hourly sea state against a single spectrum per data bin. The results indicate an influence of both the used spectral shape as well as the use of spectra for each sea state instead of a single spectrum per bin of data.