Aerospace Perspectives Series: Machine Learning Enabled Real-Time Aerodynamic Analysis by Fusing Numerical and Experimental Data 9 December 2026 | 10–11 a.m. (ET) Online
Throughout the design process, engineers gather aerodynamic data from multiple sources, including numerical simulations and wind-tunnel testing.
In this webinar, we will explore how data fusion and machine-learning surrogate models can bring together CFD and sparse wind-tunnel measurements to build a more complete view of aerodynamic behavior in real time. By comparing measurements with model predictions while the test is still running, engineers can identify outlying data early, investigate potential test issues, and take corrective actions before the test is complete. This can help avoid costly re-testing, reduce schedule risk, and improve the overall efficiency of the test campaign. We’ll also look at how DLR Models can be visualized and interrogated in Tecplot 360, including exploring pressure distributions at locations where no physical measurement was taken and comparing predictions from models that evolve once new data is available.
This webinar is relevant to engineers and researchers working in simulation, experimental testing, model validation, and data-driven analysis.
What you’ll learn:
- How real-time data fusion can be used to get insights into aerodynamic behavior during testing.
- How ML models can help to identify measurement outliers while testing in real time.
- How CFD-trained surrogate models can be adapted once experimental data is available.
- How to combine ML models and visualization tools to interactively investigate aerodynamic behavior.
- How these workflows could evolve to support flight testing and more automated, AI-assisted aerodynamic analysis.
Speakers
Philipp Bekemeyer
Head of Group for Surrogates and Uncertainty Management at DLR
Institute of Aerodynamics and Flow Technology
Yves-Marie Lefebvre
Chief Technical Officer
Tecplot
Presented By
