Prediction of Temperature Distribution on an Aircraft Hot-Air Anti-Icing Surface by ROM and Neural Networks

To address the inefficiencies and time-consuming nature of traditional hot-air anti-icing system designs, reduced-order models (ROMs) and machine learning techniques are introduced to predict anti-icing surface temperature distributions. Two models, AlexNet combined with Proper Orthogonal Decomposit...

Full description

Saved in:
Bibliographic Details
Main Authors: Ziying Chu, Ji Geng, Qian Yang, Xian Yi, Wei Dong
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Aerospace
Subjects:
Online Access:https://www.mdpi.com/2226-4310/11/11/930
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850267446735273984
author Ziying Chu
Ji Geng
Qian Yang
Xian Yi
Wei Dong
author_facet Ziying Chu
Ji Geng
Qian Yang
Xian Yi
Wei Dong
author_sort Ziying Chu
collection DOAJ
description To address the inefficiencies and time-consuming nature of traditional hot-air anti-icing system designs, reduced-order models (ROMs) and machine learning techniques are introduced to predict anti-icing surface temperature distributions. Two models, AlexNet combined with Proper Orthogonal Decomposition (POD-AlexNet) and multi-CNNs with GRU (MCG), are proposed by comparing several classic neural networks. Design variables of the hot-air anti-icing cavity are used as inputs of the two models, and the corresponding surface temperature distribution data serve as outputs, and then the performance of these models is evaluated on the test set. The POD-AlexNet model achieves a mean prediction accuracy of over 95%, while the MCG model reaches 96.97%. Furthermore, the proposed model demonstrates a prediction time of no more than 5.5 ms for individual temperature samples. The proposed models not only provide faster predictions of anti-icing surface temperature distributions than traditional numerical simulation methods but also ensure acceptable accuracy, which supports the design of aircraft hot-air anti-icing systems based on optimization methods such as genetic algorithms.
format Article
id doaj-art-38fbd3db97e845eeb5db844cbf28289c
institution OA Journals
issn 2226-4310
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Aerospace
spelling doaj-art-38fbd3db97e845eeb5db844cbf28289c2025-08-20T01:53:48ZengMDPI AGAerospace2226-43102024-11-01111193010.3390/aerospace11110930Prediction of Temperature Distribution on an Aircraft Hot-Air Anti-Icing Surface by ROM and Neural NetworksZiying Chu0Ji Geng1Qian Yang2Xian Yi3Wei Dong4School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaState Key Laboratory of Aerodynamics, Mianyang 621000, ChinaState Key Laboratory of Aerodynamics, Mianyang 621000, ChinaSchool of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, ChinaTo address the inefficiencies and time-consuming nature of traditional hot-air anti-icing system designs, reduced-order models (ROMs) and machine learning techniques are introduced to predict anti-icing surface temperature distributions. Two models, AlexNet combined with Proper Orthogonal Decomposition (POD-AlexNet) and multi-CNNs with GRU (MCG), are proposed by comparing several classic neural networks. Design variables of the hot-air anti-icing cavity are used as inputs of the two models, and the corresponding surface temperature distribution data serve as outputs, and then the performance of these models is evaluated on the test set. The POD-AlexNet model achieves a mean prediction accuracy of over 95%, while the MCG model reaches 96.97%. Furthermore, the proposed model demonstrates a prediction time of no more than 5.5 ms for individual temperature samples. The proposed models not only provide faster predictions of anti-icing surface temperature distributions than traditional numerical simulation methods but also ensure acceptable accuracy, which supports the design of aircraft hot-air anti-icing systems based on optimization methods such as genetic algorithms.https://www.mdpi.com/2226-4310/11/11/930hot-air anti-icingtemperature distribution predictionmachine learningneural networksreduced-order model
spellingShingle Ziying Chu
Ji Geng
Qian Yang
Xian Yi
Wei Dong
Prediction of Temperature Distribution on an Aircraft Hot-Air Anti-Icing Surface by ROM and Neural Networks
Aerospace
hot-air anti-icing
temperature distribution prediction
machine learning
neural networks
reduced-order model
title Prediction of Temperature Distribution on an Aircraft Hot-Air Anti-Icing Surface by ROM and Neural Networks
title_full Prediction of Temperature Distribution on an Aircraft Hot-Air Anti-Icing Surface by ROM and Neural Networks
title_fullStr Prediction of Temperature Distribution on an Aircraft Hot-Air Anti-Icing Surface by ROM and Neural Networks
title_full_unstemmed Prediction of Temperature Distribution on an Aircraft Hot-Air Anti-Icing Surface by ROM and Neural Networks
title_short Prediction of Temperature Distribution on an Aircraft Hot-Air Anti-Icing Surface by ROM and Neural Networks
title_sort prediction of temperature distribution on an aircraft hot air anti icing surface by rom and neural networks
topic hot-air anti-icing
temperature distribution prediction
machine learning
neural networks
reduced-order model
url https://www.mdpi.com/2226-4310/11/11/930
work_keys_str_mv AT ziyingchu predictionoftemperaturedistributiononanaircrafthotairantiicingsurfacebyromandneuralnetworks
AT jigeng predictionoftemperaturedistributiononanaircrafthotairantiicingsurfacebyromandneuralnetworks
AT qianyang predictionoftemperaturedistributiononanaircrafthotairantiicingsurfacebyromandneuralnetworks
AT xianyi predictionoftemperaturedistributiononanaircrafthotairantiicingsurfacebyromandneuralnetworks
AT weidong predictionoftemperaturedistributiononanaircrafthotairantiicingsurfacebyromandneuralnetworks