Artificial Intelligence and High-Performance Computing in the Context of Particle-Laden Turbulent Flow and Wind Energy*

Leading edge erosion on wind turbine blades is a significant issue that adversely affects the efficiency of wind turbines. Various airborne objects, such as ice, dust, insects, raindrops, and snowflakes, contribute to erosion, with the outer areas of the turbine blades being particularly susceptible due to higher speeds.

To investigate the behavior of particles causing erosion, this study investigates the fundamentals of turbulent particle-laden flow using experiments. Specifically, the stagnation area of the leading edge is measured, to observe particle behavior and erosion effects. A Lagrangian particle tracking technique is employed to gather data pertaining to the inertial particle dynamics and tracer particles, separately.

The study sheds light into the relationship between turbulence intensity, particle sizes, and deformation rates on leading edge erosion. An innovative approach involving deep learning-based models and high-performance computing to predict and model leading-edge erosion using the acquired dataset is proposed. The resulting predictive model can potentially be used to optimize blade surface materials and mitigate erosion effectively.