Fault Detection of the Mooring system in Floating Offshore Wind Turbines based on the Wave-excited Linear Model

Yichao Liu, Alessandro Fontanella, Ping Wu, Riccardo M.G. Ferrari, Jan Willem Van Wingerden

Research output: Contribution to journalConference articleScientificpeer-review

2 Citations (Scopus)
17 Downloads (Pure)


Floating Offshore Wind Turbines (FOWTs) are more prone to suffer from faults and failures than bottom-fixed counterparts due to the severe wind and wave loads typical of deep water sites. In particular, mooring line faults may lead to unacceptably high operation and maintenance costs due to the limited accessibility of FOWTs. Detecting the mooring line faults is therefore critical, but the application of Fault Detection (FD) techniques has not been investigated yet. In this paper, an FD scheme based on a wave-excited linear model is developed to detect in a reliable way critical mooring line faults occurring at the fairlead and anchor ends. To reach the goal, a linear model of the FOWT is obtained by approximating the wave radiation and incident wave forces. Based on this model, an observer is built to predict the rigid rotor and platform dynamics. The FD scheme is thus implemented by comparing the Mahalanobis Distance of the observer prediction error against a probabilistic detection threshold. Numerical simulations in some selected fault scenarios show that the wave-excited linear model can predict the FOWT dynamics with good accuracy. Based on this, the FD scheme capabilities are demonstrated, showing that it is able to effectively detect two critical mooring line faults.

Original languageEnglish
Article number022049
Number of pages10
JournalJournal of Physics: Conference Series
Issue number2
Publication statusPublished - 2020
EventScience of Making Torque from Wind 2020, TORQUE 2020 - Online, Virtual, Online, Netherlands
Duration: 28 Sep 20202 Oct 2020


Dive into the research topics of 'Fault Detection of the Mooring system in Floating Offshore Wind Turbines based on the Wave-excited Linear Model'. Together they form a unique fingerprint.

Cite this