One of the greatest challenges facing modern wind turbines is leading edge erosion (LEE). Even small defects or surface roughness on the blade leading edge (LE) can reduce its aerodynamic efficiency, leading to lower energy production and higher maintenance costs. As turbines become larger and operate at higher tip speeds, accurately predicting these effects becomes increasingly important.

A recent study presented at TORQUE 2026 brings together researchers from leading research organisations and industry to compare how different aerodynamic simulation tools predict the impact of LEE on wind turbine performance. The benchmark forms part of IEA Wind Task 46, an international collaboration dedicated to improving the understanding of blade erosion and its consequences.

Why does benchmarking matter?

Wind turbine manufacturers and operators rely heavily on aerodynamic models to estimate energy production, optimise blade designs and plan maintenance strategies. However, not all computational models capture the aerodynamic penalties of erosion equally well.

If different simulation tools predict differing aerodynamic penalties for the same blade damage, blade designers need to understand where these differences come from and how reliable the predictions are. Benchmark studies like this one provide an objective comparison using common test cases and experimental measurements, helping researchers identify strengths, limitations and areas for improvement.

What did the researchers do?

The team evaluated a range of state-of-the-art aerodynamic tools, including CFD-RANS solvers and viscous–inviscid interaction methods, by comparing their predictions against wind tunnel measurements.

The benchmark studied the FFA-W3-211 aerofoil, an open-access profile representative of those used within the industry, with seven different surface conditions:

  • Clean aerofoil with natural transition.
  • Clean aerofoil with forced turbulent flow.
  • Two roughness cases using P40 and P400 sandpaper (P40: severe roughness; P400: extremely mild).
  • Three realistic leading edge erosion profiles extracted from high-resolution 3D scans of damaged blades.

Unlike simplified roughness models, the erosion cases reproduced the actual geometry of damaged blade surfaces, allowing researchers to assess how well numerical tools capture real erosion effects. Simulations were performed for several Reynolds numbers and compared with measurements obtained in Poul la Cour DTU Wind’s Poul la Cour wind tunnel.

Figure 1. Benchmark erosion geometries used in the study. The figure compares mild and severe leading edge erosion profiles extracted from high-resolution scans of real damaged blade surfaces. These realistic geometries were used to evaluate how accurately different aerodynamic models reproduce erosion effects.

Main findings

The comparison showed that most aerodynamic models provide very similar predictions when airflow remains attached to the blade, meaning they can reliably estimate performance under normal operating conditions.

Figure 2. Comparison of aerodynamic performance predictions for different erosion severities. Despite using different numerical approaches, the simulation tools show good agreement under severe erosion conditions, while mild erosion highlights differences in how models predict the transition from laminar to turbulent flow.

However, larger differences appear near stall conditions, where airflow begins separating from the blade surface. Predicting this behaviour remains one of the most challenging aspects for current aerodynamic models.

The study also found that:

  • Moderate surface roughness can generally be predicted well by existing models.
  • Very rough surfaces are more difficult to simulate because current roughness models mainly account for additional friction, while real erosion also increases pressure drag through local flow separation.
  • For realistic erosion geometries, the transition from laminar to turbulent flow becomes the dominant mechanism affecting aerodynamic performance.
  • The ability to predict this transition depends strongly on mesh resolution: higher-resolution computational grids capture erosion features more accurately and therefore predict earlier transition and more realistic aerodynamic losses.

Interestingly, once erosion becomes sufficiently severe and the flow is fully turbulent, the different simulation tools produce remarkably similar performance predictions despite using different numerical approaches.

Why is this important for AIRE?

One of the objectives of the AIRE Project is to improve the prediction of wind turbine performance under realistic environmental conditions, including the effects of rain-induced blade erosion.

This benchmark demonstrates that today’s aerodynamic tools already provide reliable predictions over much of a turbine’s normal operating range, while also highlighting where further improvements are needed. In particular, better representation of transition mechanisms and higher-resolution modelling of erosion damage will help reduce uncertainty in performance predictions and energy yield assessments.

These advances contribute directly to developing more accurate digital models for capturing the effects blade degradation, supporting better maintenance planning, improved blade designs and more reliable operation of future wind farms.

Wrap up

Understanding how erosion changes the aerodynamic behaviour of wind turbine blades is essential for reducing maintenance costs and improving energy production throughout a turbine’s lifetime.

By systematically comparing the world’s leading aerodynamic simulation tools against high-quality experimental measurements, this benchmark provides valuable guidance for both researchers and industry. The results not only improve confidence in existing modelling approaches but also identify the key challenges that future aerodynamic models must overcome to better represent real blade damage.

 

For the full publication, visit the Journal of Physics: Conference Series (TORQUE 2026).

LINK: https://iopscience.iop.org/article/10.1088/1742-6596/3224/4/042037

Author: Laia Mencía
Editor: Alex Meyer Forsting
September, 2026