Mooring force estimation for floating offshore wind turbines with augmented Kalman Filter: a step towards digital twin

During the recent research studies Digital Twin (DT) simulation models for Structural Health Monitoring (SHM)
based on data-driven mode have been developed, which can provide accurate simulation and prediction of mooring forces of Floating Offshore Wind Turbines (FOWTs). However, the performance of this kind modelling is highly affected by the quantity of real data training set and it is limited to some specific configuration and the recorded environmental conditions. More importantly, the DT cannot interpret the physical meaning of structural dynamic interactions.

Therefore, a new Physics-Based estimator is proposed in this work. The fully coupled FOWT simulations are carried out
in QBlade Ocean and the simulation results are used to prepare the Reduced-Order Model by system identification for different sea states. The proposed estimator is based on the Augmented Kalman Filter in which the unknown mooring force is augmented as a state. The prediction of state is adjusted with the measurable platform motion data. It demonstrates the ability of filtering the noise in measurements and capturing the dynamic behaviour of FOWT with acceptable low computational cost. This real-time state estimator also provides the foundation of developing the DT modelling framework of FOWT and enables us to scale-up FOWTs in the next stage.