Automated Detection of Premature Flow Transitions on Wind Turbine Blades Using Model-Based Algorithms

Defects on rotor blade leading edges of wind turbines can lead to premature laminar–turbulent transitions, whereby the turbulent boundary layer flow forms turbulence wedges. The increased area around is interest here, as it have a negative effect energy production turbine. Infrared thermography an established method visualize transition from laminar flow, but contrast-to-noise ratio (CNR) wedges often too low allow reliable wedge detection with existing image processing techniques. To facilitate detection, model-based algorithm presented that uses prior knowledge about wedge-like shape transition. A verification simulated thermograms and validation measured operating turbine are performed. As result, proposed able detect determine their down CNR 2. For in recorded thermogram 0.2, at least 80% detected. Thus, proven be powerful tool for blades in-service determining resulting areas additional regions measurement error.