This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Open AccessArticle
Nonlinear Surrogate Model Design for Aerodynamic Dataset Generation Based on Artificial Neural Networks
by
Guillermo Suarez
Guillermo Suarez 1,*,
Emre Özkaya
Emre Özkaya 1,
Nicolas R. Gauger
Nicolas R. Gauger 1
,
Hans-Jörg Steiner
Hans-Jörg Steiner 2,
Michael Schäfer
Michael Schäfer 2 and
David Naumann
David Naumann 2
1
Chair for Scientific Computing, University of Kaiserslautern-Landau (RPTU), 67663 Kaiserslautern, Germany
2
Airbus Defense and Space (AD&S), 85077 Manching, Germany
*
Author to whom correspondence should be addressed.
Aerospace 2024, 11(8), 607; https://doi.org/10.3390/aerospace11080607 (registering DOI)
Submission received: 29 May 2024
/
Revised: 12 July 2024
/
Accepted: 19 July 2024
/
Published: 24 July 2024
Abstract
In this work we construct a surrogate model using artificial neural networks (ANN) to predict the steady-state behavior of an unmanned combat aircraft. We employ various strategies to improve the model’s accuracy, including the consideration of design tolerances, creating independent surrogate models for the different flow regimes and encoding non-numeric input features. We also explore alternative machine learning models, albeit they demonstrated a lower reliability than ANNs. Two scenarios are considered for the target variable: one focusing solely on predicting the pitching moment coefficient, and the other incorporating the roll moment coefficient as well. We investigate different methods for handling multiple targets, finding that constructing a single model with multiple outputs consistently outperforms developing separate models for each target variable. Overall, the ANN provides predictions that show excellent agreement with the experimental data, demonstrating its effectiveness and reliability in aerodynamic modeling.
Share and Cite
MDPI and ACS Style
Suarez , G.; Özkaya , E.; Gauger , N.R.; Steiner , H.-J.; Schäfer , M.; Naumann, D.
Nonlinear Surrogate Model Design for Aerodynamic Dataset Generation Based on Artificial Neural Networks. Aerospace 2024, 11, 607.
https://doi.org/10.3390/aerospace11080607
AMA Style
Suarez G, Özkaya E, Gauger NR, Steiner H-J, Schäfer M, Naumann D.
Nonlinear Surrogate Model Design for Aerodynamic Dataset Generation Based on Artificial Neural Networks. Aerospace. 2024; 11(8):607.
https://doi.org/10.3390/aerospace11080607
Chicago/Turabian Style
Suarez , Guillermo, Emre Özkaya , Nicolas R. Gauger , Hans-Jörg Steiner , Michael Schäfer , and David Naumann.
2024. "Nonlinear Surrogate Model Design for Aerodynamic Dataset Generation Based on Artificial Neural Networks" Aerospace 11, no. 8: 607.
https://doi.org/10.3390/aerospace11080607
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details
here.
Article Metrics
Article metric data becomes available approximately 24 hours after publication online.