In the previous chapter, the the data driven methodologies introduced in Chapter 3 were applied to data from a high fidelity simulation in SOWFA where Wake Redirection Control was tested. More specifically, the Input Output Dynamic Mode Decomposition was used, where the streamwise velocity component u figured in the snapshot data matrices. A simulation was initially performed to obtain data to identify models. The wake dynamics were explored with this data set and several models calculated. A second simulation was then performed. The data set obtained in the latter was then used to validate the already computed models based on the pre-defined evaluation criteria. A best performing model was then chosen, and this was investigated in detail. Its ability to predict the turbine power output and the wake’s behaviour was showcased. The dynamical properties of the model were then inspected. Some of the high dimensional vectors of the state space matrix \(\mathbf {A}\) were visually represented in an attempt to have a better grasp of the dynamics at play. In an attempt to better capture the existing non-linear dynamics in a simplified linear model, non-linear transformations of the data used in the snapshot matrices were performed.