The focus of this thesis is on the development of a fluid–solid interaction (FSI)
model, based on the idea of the immersed boundary method. The novelty of this
approach is the combination of a two–fluid approach to represent the solid phase
on a fluid finite–element mesh, with the conservative projection of data between
two unrelated meshes. While this is an important feature for two–way coupled FSI
models, this thesis analyses the outcome of this method based on one–way coupled
FSI problems, in which the solid phase has a prescribed velocity. The presented
FSI method is validated on several test cases with static solids as well as solids with
a prescribed velocity. For complex computational fluid dynamic (CFD) problems,
mesh adaptivity methods are used to reduce the computational effort while obtaining
the same accuracy compared to fixed meshes. In this work mesh adaptivity is also
used to increase the resolution of the fluid mesh near the solid boundary in order to
obtain an accurate representation of the solid’s shape on the fluid mesh. However,
spurious peaks in the pressure occur due to the projection of fields after adapting
the mesh. This causes peaks in the drag force and results in a potential problem
by decreasing the accuracy, especially for two–way coupled FSI problems. Since
the FSI method was developed with two–way coupled FSI problems in mind, the
occurrence of the spurious peaks was analysed and methods are shown to minimise
the peaks in the drag force. Finally, the developed FSI method is applied to rotating
vertical axis turbines and the results are compared to experimental results. This
again shows the difficulties of applying the method and assesses how it can be used
for turbine modelling, and furthermore used for analysing optimised turbine layouts.