Contrast enhancement in thermographic flow visualization on wind turbines in operation

Abstract

Thermographic flow visualization is a non-contact, non-invasive approach for assessing the aerodynamic state of wind turbine rotor blades and, as a result, the overall efficiency of the wind turbine. The distinguishability between the laminar and turbulent flow regimes in operating wind turbines cannot be easily increased intentionally and is totally dependent on the energy input from the sun. To deal with low-contrast measurement conditions and improve the distinguishability between flow regimes, advanced image processing using the feature extraction method principal component analysis is used. The image processing is applied to an image series of thermographic flow visualizations of a steady flow situation in a free-field experiment on a wind turbine in operation with a low distinguishability between the laminar and turbulent flow regime. The resulting feature images, based on the temporal intensity fluctuations in the images, are evaluated with regard to the global distinguishability between the laminar and turbulent flow regime. By applying the principal component analysis, the contrast-to-noise ratio was increased by a factor of 2.5. Furthermore, the resulting flow visualizations enable a localization of the laminar-turbulent flow transition, that is not possible in the raw data due to missing features in the intensity profile that allow a clear separation.