Using the framework of ambit processes, Jürgen Schmiegel and Ole E. Barndorff-Nielsen have built a stochastic model for the velocity and energy dissipation processes of fully developed turbulence. The modelling framework is very general and requires estimation of parameters to model a concrete situation, for example atmospheric boundary layer flows observed in wind farms and elsewhere. We demonstrate simple ways of estimating these parameters and compare simulated velocities with measured velocities from various sites, and we show how the model is able to reproduce the stylized features of fully developed turbulence. The current model describes the timewise development of the streamwise velocity component at a single point in space. The ambition is to generalize the model to a full spatio-temporal stochastic vector field. One important application is to improve the modelling of the interaction between wind and wind turbines where the current industry standard is to use just a Gaussian vector field which captures nothing beyond the second order structure functions.
Keywords: Turbulence; Ambit Processes; Wind Energy
Biography: Emil Hedevang is working as an Industrial PhD student in Applied Mathematics at Siemens Wind Power A/S in collaboration with the Department of Mathematical Sciences at the University of Aarhus. The subject is stochastic modelling of turbulence with applications to wind farm flows, in particular modelling of free, fully developed turbulence and its interaction with wind turbines. He holds a Master's degree in Mathematics and is interested in typography and all that can be computed.