Aeroacoustic Liner Impedance Metamodel from Simulation and Experimental Data Using Probabilistic Learning - Université Gustave Eiffel Accéder directement au contenu
Article Dans Une Revue AIAA Journal Année : 2023

Aeroacoustic Liner Impedance Metamodel from Simulation and Experimental Data Using Probabilistic Learning

Résumé

In order to be able to calculate a parameterized aeroacoustic liner impedance, a robust statistical metamodel is constructed as a function of the frequency and of the control parameters that are the percentage of open area and the sound pressure level. This construction is based on the use of simulated data generated with a computationally expensive aeroacoustic model, which translates to a very small training dataset. This means that the learning process has to be used and the probabilistic learning on manifolds algorithm is chosen. Although the aeroacoustic simulation is conducted on a large aeroacoustic computational model, some approximations are introduced, generating model errors that are taken into account by a probability model in the constructed training dataset. This probability model is calibrated using dimensionless experiments available from the open literature. Despite the fact that only a small amount of data is available, a novel statistical metamodel is successfully developed for which the predictions are consistent. This statistical framework allows for exhibiting a confidence region of the parameterized aeroacoustic liner impedance, which gives an information about the level of uncertainties as a function of the frequency and the control parameters.
Fichier principal
Vignette du fichier
publi-2023-AIAA-sinha-soize-desceliers-cunha.pdf (1.94 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04182265 , version 1 (19-08-2023)

Identifiants

Citer

Amritesh Sinha, Christian Soize, Christophe Desceliers, Guilherme Coelho-Cunha. Aeroacoustic Liner Impedance Metamodel from Simulation and Experimental Data Using Probabilistic Learning. AIAA Journal, 2023, 61 (11), pp.4926-4934. ⟨10.2514/1.J062991⟩. ⟨hal-04182265⟩
36 Consultations
25 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More