Steady-state flowfield prediction in transonic regime via deep-learning framework
Steady-state flowfield prediction in transonic regime via deep-learning framework
This article focuses on the development of a deep–learning framework for predicting distributed quantities around aircraft flying in the transonic regime. These quantities play a crucial role in determining aerodynamic loads and conducting aeroelastic analysis. Angle of attack and Mach number are chosen as the two independent parameters for the reduced–order models. A comparative assessment is conducted between the proposed deep-learning framework and the Proper Orthogonal Decomposition approach to identify strengths and weaknesses of each method. The accuracy of the data–driven machine–learning method in modelling transonic, steady–state aerodynamics is assessed against three benchmark cases of three-dimensional wings. Despite the challenges of the analyzed scenarios, promising results are obtained for each test case, showing the effectiveness of the model implemented. Furthermore, the article demonstrates the application of the method for aeroelastic analysis and uncertainty quantification. This quantifies robustness and versatility of the implemented model.
Immordino, Gabriele
ed9626cc-aa2b-40be-b376-0868967e5e65
Da Ronch, Andrea
a2f36b97-b881-44e9-8a78-dd76fdf82f1a
Righi, Marcello
83c3435f-2abf-45f3-9aef-1f1971043a82
Immordino, Gabriele
ed9626cc-aa2b-40be-b376-0868967e5e65
Da Ronch, Andrea
a2f36b97-b881-44e9-8a78-dd76fdf82f1a
Righi, Marcello
83c3435f-2abf-45f3-9aef-1f1971043a82
Immordino, Gabriele, Da Ronch, Andrea and Righi, Marcello
(2023)
Steady-state flowfield prediction in transonic regime via deep-learning framework.
AIAA Journal.
(In Press)
Abstract
This article focuses on the development of a deep–learning framework for predicting distributed quantities around aircraft flying in the transonic regime. These quantities play a crucial role in determining aerodynamic loads and conducting aeroelastic analysis. Angle of attack and Mach number are chosen as the two independent parameters for the reduced–order models. A comparative assessment is conducted between the proposed deep-learning framework and the Proper Orthogonal Decomposition approach to identify strengths and weaknesses of each method. The accuracy of the data–driven machine–learning method in modelling transonic, steady–state aerodynamics is assessed against three benchmark cases of three-dimensional wings. Despite the challenges of the analyzed scenarios, promising results are obtained for each test case, showing the effectiveness of the model implemented. Furthermore, the article demonstrates the application of the method for aeroelastic analysis and uncertainty quantification. This quantifies robustness and versatility of the implemented model.
Text
AIAA_Journal_Steady_Flowfield_Prediction_in_Transonic_Regime_Via_Deep_Learning_Framework
- Accepted Manuscript
More information
Submitted date: 18 October 2023
Accepted/In Press date: 28 December 2023
Identifiers
Local EPrints ID: 484752
URI: http://eprints.soton.ac.uk/id/eprint/484752
ISSN: 0001-1452
PURE UUID: e448635f-55af-456a-a15c-340f132d90f8
Catalogue record
Date deposited: 21 Nov 2023 17:33
Last modified: 18 Mar 2024 05:02
Export record
Contributors
Author:
Gabriele Immordino
Author:
Marcello Righi
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics