Low-dimensional models for aerofoil icing
Low-dimensional models for aerofoil icing
This work presents a framework for the development of low–dimensional models for application to icing around an aerofoil. The framework builds on: an adaptive sampling strategy to identify the critical icing conditions across the icing envelope for continuous intermittent icing; a classical proper orthogonal decomposition; and more modern neural network architectures. The variety in simulated ice profiles, ranging from smooth to rough and irregular shapes, motivated the use of an unsupervised classification of the icing envelope. This allowed deploying the proper orthogonal decomposition locally within each cluster, improving sensibly the prediction accuracy over the global model. On the other hand, the neural network architecture and the convolutional auto–encoder were found insensitive to the complexity in ice profiles. We found that in the worst case of icing conditions, ice causes a loss of maximum lift coefficient of up to 65%.
Aerospace Research Central
Massegur, David
d5bc71e8-f1b8-4c9f-9537-7ff63ad19426
Clifford, Declan
21e6f551-888d-4b54-8da1-22b8bcbe956d
Da Ronch, Andrea
a2f36b97-b881-44e9-8a78-dd76fdf82f1a
Lombardi, Riccardo
32bd02a9-ff01-470c-bb22-324298c360de
Panzeri, Marco
d2fa4ec7-645a-4058-a853-cf2e221a6921
Massegur, David
d5bc71e8-f1b8-4c9f-9537-7ff63ad19426
Clifford, Declan
21e6f551-888d-4b54-8da1-22b8bcbe956d
Da Ronch, Andrea
a2f36b97-b881-44e9-8a78-dd76fdf82f1a
Lombardi, Riccardo
32bd02a9-ff01-470c-bb22-324298c360de
Panzeri, Marco
d2fa4ec7-645a-4058-a853-cf2e221a6921
Massegur, David, Clifford, Declan, Da Ronch, Andrea, Lombardi, Riccardo and Panzeri, Marco
(2022)
Low-dimensional models for aerofoil icing.
In AIAA AVIATION 2022 Forum.
Aerospace Research Central..
(doi:10.2514/6.2022-3696).
Record type:
Conference or Workshop Item
(Paper)
Abstract
This work presents a framework for the development of low–dimensional models for application to icing around an aerofoil. The framework builds on: an adaptive sampling strategy to identify the critical icing conditions across the icing envelope for continuous intermittent icing; a classical proper orthogonal decomposition; and more modern neural network architectures. The variety in simulated ice profiles, ranging from smooth to rough and irregular shapes, motivated the use of an unsupervised classification of the icing envelope. This allowed deploying the proper orthogonal decomposition locally within each cluster, improving sensibly the prediction accuracy over the global model. On the other hand, the neural network architecture and the convolutional auto–encoder were found insensitive to the complexity in ice profiles. We found that in the worst case of icing conditions, ice causes a loss of maximum lift coefficient of up to 65%.
This record has no associated files available for download.
More information
e-pub ahead of print date: 20 June 2022
Venue - Dates:
2022 AIAA Aviation and Aeronautics Forum and Exposition (AIAA AVIATION Forum), , Chicago, United States, 2022-06-27 - 2022-07-01
Identifiers
Local EPrints ID: 484028
URI: http://eprints.soton.ac.uk/id/eprint/484028
PURE UUID: 5002b60a-227c-4b13-9afd-e0082c9bd7da
Catalogue record
Date deposited: 09 Nov 2023 17:34
Last modified: 18 Mar 2024 03:59
Export record
Altmetrics
Contributors
Author:
David Massegur
Author:
Declan Clifford
Author:
Riccardo Lombardi
Author:
Marco Panzeri
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