Data-driven Models for Wind Farm Performance Optimization
Sarah Barber, The Swiss Wind Energy R&D Network
-
CIS
IEEE Members: Free
Non-members: FreeLength: 00:31:56
Sarah Barber, The Swiss Wind Energy R&D Network
ABSTRACT: The power generated by a wind turbine is dependent on the atmospheric conditions, such as wind speed, air density, turbulence intensity and shear, as well as by the wakes of upstream wind turbines. Better understanding of these effects is important for optimising the total power output of a wind farm. In this talk, I will discuss how data-driven models are used in the wind energy industry to predict (i) the effects of atmospheric conditions on single wind turbine performance, and (ii) wakes within wind farms. The advantage of data-driven models is their efficiency and the incorporation of field data of actual wind farms. I will give examples of some of our work with Swiss industry partners, including a recent project applying graph neural networks (GNNs), which combine recursive neural networks and Markov Chains for the use on graph structures and allow for node-level applications, to predict wake interactions.