Speaker
Description
Critical structural components such as the blades, parts of the mooring systems and the tower of Floating Offshore Wind Turbines (FOWTs) suffer from high loadings during normal operation. The monitoring of the components’ structural health is vital as the malfunction of these components may completely stop the FOWT operation and demand their costly transfer on land for repair. The floating behaviour along with the varying environmental conditions affect so significantly the dynamics as to cover the effects of potential early damages, making thus the task of a remote vibration-based Structural Health Monitoring system highly challenging. The study of diagnosis (detection, localization, magnitude estimation) of damages in these FOWT components is still limited, especially about the tower. Vibration measurements (acceleration and displacement) are employed in the relative studies due to well-known advantages such as the big variety of low-cost sensors and data acquisition devices and others.
This study focuses on the damage detection and magnitude estimation in the critical connection between the tower and the base of a new FOWT type designed during the ARCWIND project under varying operating conditions (OCs). The study constitutes the sequel of a previous work where a Structural Health Monitoring (SHM) framework for damage diagnosis to the tendons of the considered FOWT under varying sea and wind conditions has been developed. The specific SHM framework is extended in this study in order to exploit the information obtained from vibration measurements from two points on the FOWT via the multivariate version of a powerful data-driven method. This employs stochastic functional models for the representation of the structural dynamics with their parameters being functions of the varying OCs and damage magnitude.
Vibration acceleration signals of properly selected directions from two points on the wind turbine under varying sea and wind conditions are obtained via an ANSYS-AQWA model coupled with FAST from NREL corresponding to a 10MW FOWT. The simulations include signals from the healthy and damaged FOWT exploring damage scenarios corresponding to buckling in the critical area where the tower is connected to the upper tank of the FOWT platform. The buckling at the critical area is simulated via stiffness reduction of various percentages. The employed method’s performance is for the first time examined using models with different functional subspaces as these are selected via a Genetic Algorithm, Particle Swarm Optimization and Bayesian Optimization. Damage diagnosis is attempted for 224 test cases including the healthy and damaged FOWT with the obtained results indicating flawless damage detection and remarkable magnitude estimation with mean error of almost 3% in stiffness reduction.
