Presentations

Abstract

Mesoscale simulations of an offshore wind farm located off the UK coast exhibited differences with SCADA data for a range of wind directions. The simulations revealed significant wavy patterns originating from land and advected towards the wind farm located 100km away. SAR images confirmed the existence of these wavy patterns, resulting in highs and lows of wind speed over hundreds of kilometres upwind of the wind farm for some wind directions. This phenomenon is referred to in the literature [1] as “trapped lee waves” which result from flow over topographies under stable atmospheric conditions. In the absence of SCADA data the authors of [1] wondered about their impact on offshore wind farms. The dataset discussed here could help improve industry knowledge in this regard. In this presentation we will explain the physical reason for trapped lee waves. We will compare SAR images with movies showing travelling waves obtained from mesoscale simulations and animations generated from SCADA data to analyse various situations: * For specific events confirmed by SAR images, SCADA data show peaks and troughs of wind speed travelling throughout the wind farm, but not necessarily affecting power output as long as troughs remain above rated wind speed. * For a specific wind direction, coinciding with the presence of another wind farm upwind of the array, simulations show a better agreement with SCADA. For these directions, in presence of trapped lee waves detected by SAR, SCADA do not show any impact on wind speed. It is presumed that the upwind array acts as a filter for trapped lee waves. We will conclude discussing the ability of flow models to properly model these waves, and the  potential impact of these waves on production.  [1] Ollier, S. J., & Watson, S. J. (2022). Modelling the impact of trapped lee waves on offshore wind farm power output. Wind Energy Science Discussions, 2022, 1-43.