Multiscale modeling of the atmospheric boundary layer for realistic wind turbine loading

The rise of the wind energy industry has significantly contributed to countering cli-
mate change. Recent advancements in wind turbine technology, particularly in per-
formance improvements, reliability, predictive maintenance, and cost efficiency, have
accelerated the energy transition. However, accurately downscaling atmospheric
conditions for realistic wind turbine loading remains a challenge for the industry.
This doctoral dissertation focuses on the downscaling of atmospheric conditions,
with a specific emphasis on extreme weather events (EWEs), to predict wind tur-
bine loading in the Belgian North Sea. Recent advancements in numerical weather
prediction (NWP) techniques have made it possible to downscale weather condi-
tions for operational use using models such as the Weather Research and Fore-
casting (WRF) model. Initially, a sensitivity study is conducted for multiple EWEs
over the Belgian North Sea, evaluated against SCADA data from a Belgian offshore
wind farm. It underscores the advantages of scale-aware physics parameterizations
in the WRF model. It is further extended to investigate the added benefit of air-
sea interactions through the Coupled Ocean-Atmosphere-Wave-Sediment Transport
(COAWST) model, highlighting marginal improvements in capturing wind behavior in
specific cases. Subsequently, a new scale-aware planetary boundary layer (PBL)
parameterization, VKI-3DTKE, is developed in-house and implemented in the WRF
model. This new implementation is tested in both idealized convective and neu-
tral boundary layer simulations, as well as in the eXperimental Planetary Boundary
Layer Instrumentation Assessment campaign and in extreme weather events. This
dissertation discusses the advantages and limitations of this scheme and proposes
potential avenues for further research. Finally, two new loosely coupled model frame-
works are developed to downscale weather conditions for realistic wind turbine load-
ing. These frameworks involve coupling the COAWST and WRF models with the
OpenFAST aeroelastic solver. An offshore wind turbine is scaled and implemented
in the OpenFAST model to represent its physical counterpart. This dissertation also
highlights the proposed model frameworks’ ease of use, capabilities, and limitations.