Machine Learning for Supersonic Wind Tunnel Modeling
Location
Havener Center, Carver/Turner Room, 9:30am-11:30am
Start Date
4-1-2026 10:30 AM
End Date
4-1-2026 11:00 AM
Presentation Date
April 1, 2026; 10:30am-11:00am
Description
This work investigates the development of a data-informed model of a supersonic wind tunnel to support future advancements in adaptive control. Using a combination of experimental measurements and numerical simulations, a predictive framework is constructed to capture key relationships between operating conditions and tunnel performance. Particular emphasis is placed on modeling transient behavior during startup and the resulting impacts on flow stability and mechanical loading. A machine learning approach is explored to characterize complex, nonlinear system dynamics that are difficult to represent using traditional methods alone. The resulting model is evaluated for its ability to reproduce observed tunnel behavior across a range of operating conditions. Results demonstrate that the proposed modeling strategy can capture key performance trends and transient responses with reasonable accuracy, providing a foundation for future implementation of real-time control strategies. This work establishes an important step toward the development of adaptive, data-driven control systems for supersonic wind tunnel operation.
Biography
Currently, I am a junior in Aerospace Engineering and an Honors Academy student. I have been doing research with the Missouri S&T Aerodynamics Research Laboratory for two years. Outside of that, I am involved on campus as the president of the Missouri S&T Inline Roller Hockey Team. I am from St. Louis, MO, and I enjoy activities such as hiking, cooking, video games, reading, and listening to music. Right now, in my spare time, I am working on revitalizing my grandpa's old Dodge Dakota to get it back running in perfect shape. Finally, in high school, I was the first-chair tuba player in my band. Thank you.
Meeting Name
2026 - Miners Solving for Tomorrow Research Conference
Department(s)
Mechanical and Aerospace Engineering
Document Type
Presentation
Document Version
Final Version
File Type
text
Language(s)
English
Rights
© 2026 The Authors, All rights reserved
Machine Learning for Supersonic Wind Tunnel Modeling
Havener Center, Carver/Turner Room, 9:30am-11:30am
This work investigates the development of a data-informed model of a supersonic wind tunnel to support future advancements in adaptive control. Using a combination of experimental measurements and numerical simulations, a predictive framework is constructed to capture key relationships between operating conditions and tunnel performance. Particular emphasis is placed on modeling transient behavior during startup and the resulting impacts on flow stability and mechanical loading. A machine learning approach is explored to characterize complex, nonlinear system dynamics that are difficult to represent using traditional methods alone. The resulting model is evaluated for its ability to reproduce observed tunnel behavior across a range of operating conditions. Results demonstrate that the proposed modeling strategy can capture key performance trends and transient responses with reasonable accuracy, providing a foundation for future implementation of real-time control strategies. This work establishes an important step toward the development of adaptive, data-driven control systems for supersonic wind tunnel operation.
Comments
Advisor: Davide Vigano, dvigano@mst.edu