Thermographic flow visualization enables a noninvasive detection of the laminar–turbulent flow transition and allows a measurement of the impact of surface erosion and contamination due to insects, rain, dust, or hail by quantifying the amount of laminar flow reduction. The state-of-the-art image processing is designed to localize the natural flow transition as occurring on an undisturbed blade surface by use of a one-dimensional gradient evaluation. However, the occurrence of premature flow transitions leads to a high measurement uncertainty of the localized transition line or to a completely missed flow transition detection. For this reason, regions with turbulent flow are incorrectly assigned to the laminar flow region, which leads to a systematic deviation in the subsequent quantification of the spatial distribution of the boundary layer flow regimes. Therefore, a novel image processing method for the localization of the laminar–turbulent flow transition is introduced, which provides a reduced measurement uncertainty for sections with premature flow transitions. By the use of a two-dimensional image evaluation, local maximal temperature gradients are identified in order to locate the flow transition with a reduced uncertainty compared to the state-of-the-art method. The transition position can be used to quantify the reduction of the laminar flow regime surface area due to occurrences of premature flow transitions in order to measure the influence of surface contamination on the boundary layer flow. The image processing is applied to the thermographic measurement on a wind turbine of the type GE 1.5 sl in operation. In 11 blade segments with occurring premature flow transitions and a high enough contrast of the developed turbulence wedge, the introduced evaluation was able to locate the flow transition line correctly. The laminar flow reduction based on the evaluated flow transition position located with a significantly reduced systematic deviation amounts to 22% for the given measurement and can be used to estimate the reduction of the aerodynamic lift. Therefore, the image processing method introduced allows a more accurate estimation of the effects of real environmental conditions on the efficiency of wind turbines in operation.
Read full abstract