Issue |
JNWPU
Volume 42, Number 2, April 2024
|
|
---|---|---|
Page(s) | 328 - 334 | |
DOI | https://doi.org/10.1051/jnwpu/20244220328 | |
Published online | 30 May 2024 |
Multi-fidelity aerodynamic data analysis by using composite neural network
基于复合神经网络的多源气动数据建模
School of Mathematics and Statistics, Xi'an Jiaotong University, Xi'an 710049, China
Received:
2
March
2023
Applying deep learning to aerodynamic data modeling has important practical significance. In this paper, the composite neural network is applied to the aerodynamics, making full use of the different characteristics of high and low-fidelity aerodynamic data. Multi-fidelity analysis technique is also used to analyze the correlation between the two types of data so as to establish the composite neural network. The experimental results show that the learning of multi-fidelity aerodynamic data based on the composite neural network model can better capture the mapping relationship between the aerodynamic input and the output data. And after comparing with the single neural network, it is verified that the present model has excellent performance in the regression modeling of aerodynamic data.
摘要
将深度学习方法应用至气动数据建模, 能够解决传统建模方法效率低、代价高的问题, 具有重要的现实意义。基于复合神经网络模型对多源气动数据进行学习, 利用低精度数据辅助高精度数据进行预测。与不同网络模型进行对比, 验证了文中提出的复合神经网络在气动数据建模中表现优良, 且泛化能力较好。
Key words: aerodynamic data modeling / deep neural network / composite neural network
关键字 : 气动数据建模 / 深度神经网络 / 复合神经网络
© 2024 Journal of Northwestern Polytechnical University. All rights reserved.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.