15 - Guest lectures
GARTEUR RESEARCH GROUP AG60 EXPERIENCE AND ACTIVITIES ON MACHINE LEARNING AND DATA-DRIVEN APPROACHES FOR AERODYNAMICS AND UNCERTAINTY QUANTIFICATION
E. Andrés Pérez¹, D. Quagliarella, CIRA, Italy; D. Hines Chaves², P. Bekemeyer², T. Taddei, Sapienza University of Rome, Italy; J. Peter, Onera, France; O. Amoignon, FOI, Sweden; A. Alaia³, A. Scardigli³, R. Castellanos, UC3M, Spain; J. Nieto-Centenero¹; ¹INTA, Spain ;²DLR, Germany ;³OPTIMAD/ESTECO, Italy
This article provides an overview of the activities and results achieved in the GARTEUR AD/AG60 project, titled ”Machine Learning and Data Mining Techniques for Aerodynamic Analysis and Uncertainty Quantification.” The project partners developed and compared several machine learning models for surrogate modeling and uncertainty quantification and applied them to the AIRBUS XRF1 aircraft configuration of industrial interest. They assessed the feasibility of advanced Machine Learning methods for predicting aerodynamic behaviors in industrially relevant configurations. The article details the models created and the different approaches adopted, along with a comprehensive analysis and discussion of the results, highlighting their implications for advancing aerodynamic predictions in practical industrial scenarios.
DOI: 10.71945/icas2026_0019