Sammendrag
The most common type of Floating Offshore Wind Turbine (FOWT) installed in Norway, is the spar FOWT in which a delta mooring system (DMS) is used. The mooring system of a FOWT is an essential part for its station-keeping, whose loss can lead to the collapse of the FOWT and the endangerment of the human safety. Thus, early detection of damages in the mooring system is vital. In this study, damage detection in the DMS of a spar FOWT under varying environmental conditions (ECs) is investigated through a comparison of the Multiple Model-AutoRegressive (MM-AR) method, the Multiple Model-Power Spectral Density (MM-PSD) method and the Functional Model Based Method (FMBM). The MM-AR and the MM-PSD methods are based on multiple PSD based or AR models and the FMBM on a single Functional Model (FM) for the description of the healthy FOWT’s dynamics under varying ECs. The results show that successful and precise damage detection in a spar FOWT’s DMS can be achieved through the employed statistical methods as the MM-AR method and the FMBM detect all the examined 7 healthy and 16 damage cases whereas the MM-PSD method misses only one damage case. The latter shows that the parametric AR models and FM describe more precisely the FOWT dynamics under varying ECs in comparison to the non-parametric PSDs.
Vis fullstendig beskrivelse
