航空学报 > 2025, Vol. 46 Issue (10): 631478-631478   doi: 10.7527/S1000-6893.2025.31478

飞行器设计生成式模型专栏

基于径向基生成式对抗网络的多源数据融合方法

胡登峰, 向渝(), 张骏, 杨佳辰, 汪文勇   

  1. 电子科技大学 计算机科学与工程学院,成都 611731
  • 收稿日期:2024-11-01 修回日期:2025-01-13 接受日期:2025-04-15 出版日期:2024-04-29 发布日期:2025-04-25
  • 通讯作者: 向渝 E-mail:jcxiang@uestc.edu.cn
  • 基金资助:
    国家自然科学基金(62250067);中央高校基本科研业务费专项资金(ZYGX2024Z009)

Multi-source data fusion method based on radial basis function generative adversarial network

Dengfeng HU, Yu XIANG(), Jun ZHANG, Jiachen YANG, Wenyong WANG   

  1. School of Computer Science and Engineering,University of Electronic Science and Technology of China,Chengdu 611731,China
  • Received:2024-11-01 Revised:2025-01-13 Accepted:2025-04-15 Online:2024-04-29 Published:2025-04-25
  • Contact: Yu XIANG E-mail:jcxiang@uestc.edu.cn
  • Supported by:
    National Natural Science Foundation of China(62250067);The Fundamental Research Funds for the Central Universities(ZYGX2024Z009)

摘要:

飞行器气动数据可通过数值计算、风洞试验、飞行试验等途径获得。如何在低成本的基础上获得高保真度且不确定度较低的气动数据依然是一个挑战。在深度学习方法中,利用多层神经网络能够在更高维度上进行特征的表征与组合,从而有效地融合来自不同数据源的相关特征,提升模型的预测精度、可信度。然而,现有方法在融合不同数据集特征时,仍面临诸多技术难题。针对这一问题,提出了一种基于径向基生成式对抗网络(RBFGAN)的多源数据融合方法——RGAN-MSFM。该方法能够融合不同来源的气动数据集,挖掘多源气动数据特征之间的互补性、相关性。利用RBFGAN融合不同来源数据中的互补状态特征,生成新的融合数据集;构建功能网络和上下文网络,深入探索各数据源之间的相关性,从而实现多源数据特征的融合学习。基于包含计算流体力学(CFD)数据、风洞试验数据的数据集,以及包含CFD数据、飞行试验数据的数据集,设计了2组仿真实验。实验结果显示,采用该模型进行数据特征融合后,数据集的精度平均提高了30.2%、67.5%;与传统数据融合方法相比,不确定度分别降低了23.9%、27.6%。

关键词: 多源数据, 径向基函数, 生成式对抗网络, 状态融合, 数据融合

Abstract:

Aerodynamic data of aircraft can be obtained through methods such as numerical simulations, wind tunnel tests, and flight tests. However, obtaining high-fidelity aerodynamic data with low uncertainty at a low cost remains a challenging problem. In deep learning methods, multi-layer neural networks can represent and combine features in higher dimensions, effectively integrating relevant features from different data sources to improve the model’s prediction accuracy and reliability. However, the existing methods still face numerous technical challenges when fusing features from different datasets. To address this issue, we propose a multi-source data fusion method based on Radial Basis Function Generative Adversarial Network (RBFGAN)—RGAN-MSFM. This method can fuse aerodynamic datasets from different sources, mining the complementarity and correlation between multi-source aerodynamic data features. The proposed method first uses RBFGAN to fuse complementary state features from different data sources to generate a new fused dataset. Then, functional and contextual networks are constructed to explore the correlation among the data sources, enabling fusion learning of multi-source data features. We designed two sets of simulation experiments based on datasets containing CFD and wind tunnel test data, as well as CFD and flight test data. The experimental results show that after data feature fusion using this model, the accuracy of the datasets improved by an average of 30.2% and 67.5%; compared with traditional data fusion methods, the uncertainty was reduced by 23.9% and 27.6%, respectively.

Key words: multi-source data, radial basis function, generative adversarial networks, state fusion, data fusion

中图分类号: