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; 汪文勇; 杨佳辰

doi:10.7527/S1000-6893.2025.31478

航空学报 >

0 1 - 0

DOI: https://doi.org/10.7527/S1000-6893.2025.31478

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 ,
汪文勇 ,
杨佳辰

展开

1. 电子科技大学
2. 电子科技大学计算机科学与工程学院

收稿日期: 2024-11-01

修回日期: 2025-04-18

网络出版日期: 2025-04-25

收起

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

HU Deng-Feng ,
XIANG Yu ,
ZHANG Jun ,
WANG Wen-Yong ,
YANG Jia-Chen

Expand

Received date: 2024-11-01

Revised date: 2025-04-18

Online published: 2025-04-25

Fold

摘要

飞行器气动数据可通过数值计算（CFD）、风洞试验和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，依然是一个挑战性问题。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和可信度。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，本文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与相关性。该方法首先利用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；然后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。我们基于包含CFD数据与风洞试验数据的数据集，以及包含CFD数据与飞行试验数据的数据集，设计了两组仿真实验。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%和67.5%；与传统数据融合方法相比，不确定度分别降低了23.9%和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

本文引用格式

胡登峰 , 向渝 , 张骏 , 汪文勇 , 杨佳辰 . 基于径向基生成式对抗网络的多源数据融合方法[J]. 航空学报, 0 : 1 -0 . DOI: 10.7527/S1000-6893.2025.31478

Abstract

Aerodynamic data of aircraft can be obtained through methods such as numerical simulations (CFD), wind tunnel tests, and flight tests. However, obtaining high-fidelity aerodynamic data with low uncertainty at a low cost remains a challeng-ing problem. In deep learning methods, multi-layer neural networks can represent and combine features in higher dimen-sions, effectively integrating relevant features from different data sources to improve the model's prediction accuracy and reliability. However, existing methods still face numerous technical challenges when fusing features from different datasets. To address this issue, this paper proposes 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 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 tradi-tional data fusion methods, the uncertainty was reduced by 23.9% and 27.6%, respectively.

Key words： multi-source; Radial Basis Function; Generative Adversarial Networks; state fusion; data fusion

参考文献

[1] 唐志共, 袁先旭, 钱炜祺, 等. 高速空气动力学三大手段数据融合研究进展[J]. 空气动力学学报, 2023, 41(8): 44?58. DOI: 10.7638/kqdlxxb-2023.0096. TANG Z G, YUAN X X, QIAN W Q, et al. Research progress on the fusion of data obtained by high-speed wind tunnels, CFD and model flight[J]. Acta Aerodynamica Sinica, 2023, 41(8): 44?58. DOI: 10.7638/kqdlxxb-2023.0096 [2] POLOCZEK M , WANG J , FRAZIER P I .Multi-information source optimization[C]//Neural Information Processing Systems.2017. [3] 邓晨, 陈功, 王文正, 等. 基于不确定度和气动模型的气动数据融合算法[J]. 空气动力学学报, 2022, 40(4): 117?123. DOI: 10.7638/kqdlxxb-2020.0151. DENG C, CHEN G, WANG W Z, et al. Aerodynamic data fusion algorithms based on aerodynamic model and uncertainly[J]. Acta Aerodynamica Sinica, 2022, 40(4): 117?123. DOI: 10.7638/kqdlxxb-2020.0151 [4] 唐志共, 朱林阳, 向星皓, 等. 智能空气动力学若干研究进展及展望[J]. 空气动力学学报, 2023, 41(7): 1?35. DOI: 10.7638/kqdlxxb-2023.0128. TANG Z G, ZHU L Y, XIANG X H, et al. Some research progress and prospect of Intelligent Aerodynamics[J]. Acta Aerodynamica Sinica, 2023, 41(7): 1?35. DOI: 10.7638/kqdlxxb-2023.0128 [5] X. LIANG, Y. QIAN, Q. GUO, H. CHENG and J. LIANG, "AF: An Association-Based Fusion Method for Multi-Modal Classification," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 12, pp. 9236-9254, 1 Dec. 2022, doi: 10.1109/TPAMI.2021.3125995. [6] ZHANG Z J, DURAISAMY K. Machine learning methods for data-driven turbulence modeling[C]//22nd AIAA computational fluid dynamics conference. 2015: 2460, doi: 10.2514/6.2015-2460. [7] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. nature, 2015, 521(7553): 436-444, doi: 10.1038/nature14539 [8] L. HU, J. ZHANG, Y. XIANG and W. WANG, "Neural Networks-Based Aerodynamic Data Modeling: A Comprehensive Review," in IEEE Access, vol. 8, pp. 90805-90823, 2020, doi: 10.1109/ACCESS.2020.2993562. [9] MARCATO A, BOCCARDO G, MARCHISIO D L. A computational workflow to study particle transport in porous media: coupling CFD and deep learn-ing[M]//Computer Aided Chemical Engineering. Else-vier, 2020, 48: 1759-1764. doi: 10.1016/b978-0-12-823377-1.50294-9 [10] PARISH E J, DURAISAMY K. A paradigm for data-driven predictive modeling using field inversion and machine learning[J]. Journal of computational physics, 2016, 305: 758-774. doi: 10.1016/j.jcp.2015.11.012 [11] J. LI, B. LI, J. XU, R. XIONG and W. GAO, "Fully Connected Network-Based Intra Prediction for Image Coding," in IEEE Transactions on Image Processing, vol. 27, no. 7, pp. 3236-3247, July 2018, doi: 10.1109/TIP.2018.2817044. [12] BELYAEV, M., BURNAEV, E., KAPUSHEV, E., ALESTRA, S., DORMIEUX, M., CAVAILLES, A., … FERREIRA, E. (2014). Building Data Fusion Surrogate Models for Spacecraft Aerodynamic Problems with Incomplete Factorial Design of Experiments. Advanced Materials Research, 1016, 405–412. doi:10.4028/www.scientific.net/amr.1016.405 [13] 王文正, 桂业伟, 何开锋, 袁先旭, 陈焰青. 基于数学模型的气动力数据融合研究[J]. 空气动力学学报, 2009, 27(5): 524-528. WANG Wen-zheng, GUI Ye-wei, HE Kai-feng, YUAN Xian-xu, CHEN Yan-qin. Aerodynamic data fusion technique exploration[J]. ACTA AERODYNAMICA SINICA, 2009, 27(5): 524-528. [14] 张骏, 张广博, 程艳青, 等. 一种气动大差异性数据多任务学习方法[J]. 空气动力学学报, 2022, 40(6): 64?72. DOI: 10.7638/kqdlxxb-2021.0222. ZHANG J, ZHANG G B, CHENG Y Q, et al. A multi-task learning method for large discrepant aerodynamic data[J]. Acta Aerodynamica Sinica, 2022, 40(6): 64?72. DOI: 10.7638/kqdlxxb-2021.0222 [15] 赵旋，彭绪浩，邓子辰，等. 基于多源数据融合的翼型表面压强精细化重构方法[J]. 实验流体力学，2022，36（3）：93-101. DOI: 10.11729/syltlx20210166. ZHAO X，PENG X H，DENG Z C，et al. Fine reconstruction method of airfoil surface pressure based on multi-source data fusion[J]. Journal of Experiments in Fluid Mechanics, 2022，36（3）：93-101. . doi: 10.11729/syltlx20210166 [16] 王旭, 宁晨伽, 王文正, 等. 面向飞行试验的多源气动数据智能融合方法[J]. 空气动力学学报, 2023, 41(2): 12?20. DOI: 10.7638/kqdlxxb-2021.0428. WANG X, NING C J, WANG W Z, et al. Intelligent fusion method of multi-source aerodynamic data for flight tests[J]. Acta Aerodynamica Sinica, 2023, 41(2): 12?20. DOI: 10.7638/kqdlxxb-2021.0428 [17] LI K., KOU J. and ZHANG W., “Deep Learning for Multifidelity Aerodynamic Distribution Modeling from Experimental and Simulation Data,” AIAA Journal, Vol. 60, No. 7, 2022, pp. 4413–4427. doi:10.2514/1.j061330 [18] HU L, XIANG Y, ZHANG J, et al. Aerodynamic data predictions based on multi-task learning[J]. Applied Soft Computing, 2022, 116: 108369. doi:10.1016/j.asoc.2021.108369 [19] GIROSI F , POGGIO T .Networks and the Best Approximation Property[J].Biological Cybernetics, 1990, 63(3):169-176.DOI:10.1007/BF00195855. [20] POGGIO T , GIROSI F .Networks for approximation and learning[J].Proc IEEE, 1990, 78(9):1481-1497.DOI:10.1109/5.58326. [21] L. HU, W. WANG, Y. XIANG and J. ZHANG, "Flow Field Reconstructions With GANs Based on Radial Basis Functions," in IEEE Transactions on Aerospace and Electronic Systems, vol. 58, no. 4, pp. 3460-3476, Aug. 2022, doi: 10.1109/TAES.2022.3152706 [22] 邓晨, 陈功, 王文正, 等. 基于不确定度和气动模型的气动数据融合算法[J]. 空气动力学学报, 2022, 40(4): 117?123. doi: 10.7638/kqdlxxb-2020.0151 DENG C, CHEN G, WANG W Z, et al. Aerodynamic data fusion algorithms based on aerodynamic model and uncertainly[J]. Acta Aerodynamica Sinica, 2022, 40(4): 117?123. doi: 10.7638/kqdlxxb-2020.0151

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献