TOWARD ENVIRONMENTAL AND STRUCTURAL DIGITAL TWIN OF OFFSHORE WIND TURBINE

In the wind energy industry, a digital twin (DT) is very useful for managing the operation of a wind turbine and predicting structural health conditions in real-time as well as projections in the near future. A real-time surrogate model is a very crucial part in building a DT. Towards that end, we employ a Reduced-Order Modelling (ROM) approach to construct a surrogate model for the environment-structure system of a bottom-fixed offshore wind turbine (OWT). The entire environment-structure system is broken down into major sub-systems of wind, wave, and structure. Based on the high-fidelity Computational Fluid Dynamics (CFD) data, the wind ROM model can quickly provide the loadings on the components above water level in the parameter space spanned by the incoming wind speed, the rotational speed of the rotor, the pitch and tilt angles of the turbine. The hydro loadings on the underwater components of the OWT are solved by the empirical Morison formula. The OWT structural ROM model is component-based consisting of blades, hub, nacelle, tower, and monopile. The structural ROM model takes the loading data feeds from the wind and wave models to predict the structure responses of the OWT system, including stress. Since the DT is constructed via the component-based, it can also be used to play out "what if" scenarios when there are component level changes. For instance, the model can predict the system response of OWT when certain parts undergo structure failures. Benefiting from the cost-efficient ROM models, the DT is over two orders of magnitude faster than high-fidelity simulations while maintaining good accuracy.