Wind turbine wake detection and characterisation utilising blade loads and SCADA data: a generalised approach

Abstract. Large offshore wind farms face operational challenges due to turbine wakes, which can reduce energy yield and increase structural fatigue. These problems may be mitigated through wind farm flow control techniques, which require reliable wake detection (recognising the presence of a clear wake) and characterisation (parametric description of a wake’s properties) as prerequisites. This paper presents a novel three-stage framework for generalised wake detection and characterisation. First, a regression model utilises blade loads and SCADA data to estimate the wind speed distribution across the rotor plane. Second, a Convolutional Neural Network (CNN) undertakes pattern recognition analysis to perform the wake detection, classifying rotor-plane wind estimates as "fully-impinged", “left-impinged”, “right-impinged” or “not impinged.” Third, where wake impingement is detected, 2D Gaussian fitting is undertaken to provide a parametric wake characterisation, providing outputs of the wake centre location and wake lateral width. The framework is tested and assessed using a virtual wind farm in the North Sea and a wide range of wind conditions (mean ambient wind speeds from 5–15 m/s, turbulence intensities from 3–9 %, full range of wind directions). Results show high accuracy of wind field estimation, with the mean RMSE over all test cases being 0.351 m/s, or 5.23 % when normalised by mean ambient wind speed. A wake detection sensitivity study confirms accurate performance across a majority of wind conditions, with minor issues observed only for more extreme conditions or those at the limits of the utilised training data. The final wake characterisation stage is shown to flexibly adapt to changing wind conditions, successfully tracking the wake’s position even in more demanding partial-impingement cases. The proposed framework therefore demonstrates strong potential as a generalised approach to wake detection and characterisation.