Intelligent fusion method of multi-source aerodynamic data for flight tests
-
摘要: 风洞试验和飞行试验是飞行器研制过程中进行气动性能分析与优化设计的重要手段，然而，在高超声速飞行条件下，真实气体效应、黏性干扰效应和尺度效应的复杂变化给气动数据精准预测带来巨大挑战。为了提升天地气动数据一致性，针对某外形飞行试验数据开展了典型对象的天地气动数据融合方法研究。结合数据挖掘的随机森林方法，本文提出了一种面向飞行试验的数据融合框架，通过引入地面风洞试验气动数据，实现了对复杂输入参数的特征分析与特征排序，进一步对不同飞行时刻下飞行试验的气动数据开展了交叉验证。结果表明随机森林的机器学习框架对风洞-飞行试验数据关联具有较好的预测与外推能力，可以有效提升气动数据预测精度，相关研究为复杂环境下气动数据多源融合提供了思路。Abstract: Wind tunnel and flight tests are two of the most important methods for aerodynamic analyses and optimization design during the development of aircraft. However, under hypersonic flight conditions, the real gas effects, viscous interference effects, and multi-scale flow fields pose huge challenges to aerodynamic prediction. To improve the consistency of aerodynamic data between wind-tunnel and flight tests, an aerodynamic data fusion framework based on the random forest method for data mining is proposed and applied in the aerodynamic data fusion of a hypersonic aircraft. Feature analyses and ranking of aerodynamic data obtained by ground wind tunnel tests are conducted first. Then aerodynamic data in a flight envelop are cross-validated. Results show that the machine learning framework based on the random forest has good prediction and extrapolation capabilities for the correlation of aerodynamic data obtained by wind-tunnel and flight tests, and can effectively improve the prediction accuracy of aerodynamic data. The method provides a promising solution to the multi-source fusion of aerodynamic data in complex environments.
-
Key words:
- hypersonic /
- wind tunnel test /
- flight test /
- random forest /
- data fusion /
- parameter characteristic analysis /
- cross validation
-
图 3 飞行过程迎角与侧滑角变化（50～320 s）
Figure 3. Variations of angle-of-attack and angle-of-sideslip during the flight (50～320 s)
图 4 飞行参数辨识得到气动力系数响应历程（50～320 s）
Figure 4. Time histories of aerodynamic coefficients during the flight (50～320 s)
图 5 风洞试验与飞行试验轴向力系数对比
Figure 5. Comparison of axial force between wind tunnel test and flight test
图 6 风洞试验与飞行试验法向力系数对比
Figure 6. Comparison of vertical force between wind tunnel test and flight test
图 12 气动数据融合模型的预测结果（50～100 s）
Figure 12. Predictions of the data fusion model at 50～100 s
图 13 气动数据融合模型的预测结果（150 ~ 200 s）
Figure 13. Predictions of the data fusion model at 150 ~ 200 s
图 14 气动数据融合模型的预测结果（130～230 s）
Figure 14. Predictions of the data fusion model at 130～230 s
图 15 气动数据融合模型的预测结果（180～280 s）
Figure 15. Predictions of the data fusion model at 180～280 s
表 1 9个状态参数特征占比
Table 1. Ranking of nine characteristic parameters
特征 占比/% a 0.14 Ma 2.10 ρ 0.12 α 20.00 β 2.20 d1 1.00 d2 2.10 d3 0.52 d4 1.90 
下载: 导出CSV
表 2 数据融合框架输入参数特征占比
Table 2. Feature ranking of the input data for the data fusion framework
特征 占比/% a 0.09 Ma 1.01 ρ 0.08 α 13.30 β 1.11 d1 0.82 d2 1.54 d3 0.31 d4 1.77 CN 17.50 CA 0.95
下载: 导出CSV
表 3 训练预测数据划分（训练预测比4∶1）
Table 3. Data for training and prediction for case 1 and 2
训练状态 预测状态 算例1 100～320 s 50～100 s 算例2 50～150 s 与 200～320 s 150～200 s
下载: 导出CSV
表 4 训练预测数据划分（训练预测比1.7∶1）
Table 4. Data for training and prediction for case 3 and 4
训练状态 预测状态 算例3 50～130 s 与 230～320 s 130～230 s 算例4 50～180 s 与 280～320 s 180～280 s
下载: 导出CSV
表 5 不同预测算例均方根误差对比
Table 5. Mean-square errors of predictions
预测算例 地面风洞数据 模型预测数据 $ \sigma_{C_N}$ $\sigma_{C_A} $ $\sigma_{C_N} $ $\sigma_{C_A} $ 1 0.0517 0.0262 0.0179 0.0242 2 0.1407 0.0983 0.0361 0.0108 3 0.1265 0.0291 0.0539 0.0189 4 0.0927 0.0202 0.0411 0.0117
下载: 导出CSV
-
[1] 安复兴, 李磊, 苏伟, 等. 高超声速飞行器气动设计中的若干关键问题[J]. 中国科学 (物理学 力学 天文学), 2021, 51(10): 2-21.AN F X, LI L, SU W, et al. Key issues in hypersonic vehicle aerodynamic design[J]. SCIENTIA SINICA Physica, Mechanica & Astronomica, 2021, 51(10): 2-21. (in Chinese) [2] 陈坚强, 张益荣, 张毅锋, 等. 高超声速气动力数据天地相关性研究综述[J]. 空气动力学学报, 2014, 32(5): 587-599. doi: 10.7638/kqdlxxb-2014.0092CHEN J Q, ZHANG Y R, ZHANG Y F, et al. Review of correlation analysis of aerodynamic data between flight and ground prediction for hypersonic vehicle[J]. Acta Aerodynamica Sinica, 2014, 32(5): 587-599. (in Chinese) doi: 10.7638/kqdlxxb-2014.0092 [3] 罗长童, 胡宗民, 刘云峰, 等. 高超声速风洞气动力/热试验数据天地相关性研究进展[J]. 实验流体力学, 2020, 34(3): 78-89. doi: 10.11729/syltlx20200006LUO C T, HU Z M, LIU Y F, et al. Research progress on ground-to-flight correlation of aerodynamic force and heating data from hypersonic wind tunnels[J]. Journal of Experiments in Fluid Mechanics, 2020, 34(3): 78-89. (in Chinese) doi: 10.11729/syltlx20200006 [4] BYKERK T, VERSTRAETE D, STEELANT J. Low speed longitudinal aerodynamic, static stability and performance analysis of a hypersonic waverider[J]. Aerospace Science and Technology, 2020, 96: 105531. DOI: 10.1016/j.ast.2019.105531 [5] 王刚. 高超声速粘性干扰效应数据关联方法研究[D]. 绵阳: 中国空气动力研究与发展中心. [6] 毛枚良, 陈亮中, 万钊, 等. 高超声速流动多物理效应对美国航天飞机气动力影响研究的回顾[J]. 空气动力学学报, 2017, 35(1): 1-12.MAO M L, CHEN L Z, WAN Z, et al. Review on influence of the multi-physical effects in hypersonic flow on aerodynamic characteristics of the United States space shuttle[J]. Acta Aerodynamica Sinica, 2017, 35(1): 1-12. (in Chinese) [7] 叶友达. 高超声速空气动力学研究进展与趋势[J]. 科学通报, 2015, 60(12): 1095-1103. doi: 10.1360/N972014-01180YE Y D. Advances and prospects in hypersonic aerodynamics[J]. Chinese Science Bulletin, 2015, 60(12): 1095-1103. (in Chinese) doi: 10.1360/N972014-01180 [8] NAVON I M. Data Assimilation for Numerical Weather Prediction: A ReviewData Assimilation for Atmospheric, Oceanic and Hydrologic Applications, 2009. DOI: 10.1007/978-3-540-71056-1_2 [9] 何开锋, 钱炜祺, 汪清, 等. 数据融合技术在空气动力学研究中的应用[J]. 空气动力学学报, 2014, 32(6): 777-782.HE K F, QIAN W Q, WANG Q, et al. Application of data fusion technique in aerodynamics studies[J]. Acta Aerodynamica Sinica, 2014, 32(6): 777-782. (in Chinese) [10] MOHAMMADI-AMIN M, ENTEZARI M M, ALIKHANI A. An efficient surrogate-based framework for aerodynamic database development of manned reentry vehicles[J]. Advances in Space Research, 2018, 62(5): 997-1014. DOI: 10.1016/j.asr.2018.06.022 [11] 邓晨, 陈功, 王文正, 等. 基于飞行试验和风洞试验数据的融合算法研究[J/OL]. 空气动力学学报: 1-6 [2021-12-06].DENG C, CHEN G, WANG W Z, et al. Research on the data fusion algorithm based on flight test data and wind tunnel test data[J/OL]. Acta Aerodynamica Sinica, 2022 [2021-12-06] (in Chinese). [12] 韩忠华. Kriging模型及代理优化算法研究进展[J]. 航空学报, 2016, 37(11): 3197-3225.HAN Z H. Kriging surrogate model and its application to design optimization: a review of recent progress[J]. Acta Aeronautica et Astronautica Sinica, 2016, 37(11): 3197-3225. (in Chinese) [13] GHOREYSHI M, BADCOCK K J, WOODGATE M A. Accelerating the numerical generation of aerodynamic models for flight simulation[J]. Journal of Aircraft, 2009, 46(3): 972-980. DOI: 10.2514/1.39626 [14] DA RONCH A, GHOREYSHI M, BADCOCK K J. On the generation of flight dynamics aerodynamic tables by computational fluid dynamics[J]. Progress in Aerospace Sciences, 2011, 47(8): 597-620. DOI: 10.1016/j.paerosci.2011.09.001 [15] YONDO R, ANDRÉS E, VALERO E. A review on design of experiments and surrogate models in aircraft real-time and many-query aerodynamic analyses[J]. Progress in Aerospace Sciences, 2018, 96: 23-61. DOI: 10.1016/j.paerosci.2017.11.003 [16] 李金晟, 庄凌, 宋加洪, 等. 适用于工程数据的飞行器气动特性修正框架[J/OL]. 航空学报: 1-14 [2021-12-06]LI J S, ZHUANG L, SONG J H, et al. Aircraft aerodynamic characteristics correction framework suitable for engineering data[J/OL]. Acta Aeronautica et Astronautica Sinica: 1-14 [2021-12-06] (in Chinese). http://kns.cnki.net/kcms/detail/11.1929.V.20210318.1516.008.html [17] 龚安龙, 刘周, 王晓璐, 等. 高超声速黏性干扰相关性参数的修正[J]. 航空学报, 2013, 34(7): 1582-1587.GONG A L, LIU Z, WANG X L, et al. Correction of hypersonic viscous interaction correlation parameter[J]. Acta Aeronautica et Astronautica Sinica, 2013, 34(7): 1582-1587. (in Chinese) [18] 王文正, 桂业伟, 何开锋, 等. 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. doi: 10.3969/j.issn.0258-1825.2009.05.004WANG W Z, GUI Y W, HE K F, et al. Aerodynamic data fusion technique exploration[J]. Acta Aerodynamica Sinica, 2009, 27(5): 524-528. (in Chinese) doi: 10.3969/j.issn.0258-1825.2009.05.004 [19] 沙心国, 刘荣健, 苏鹏辉, 等. 某组合体模型气动热数据相关性研究[J]. 气体物理, 2016, 1(4): 43-50.SHA X G, LIU R J, SU P H, et al. Correlation of aerodynamic heating data of a combination model[J]. Physics of Gases, 2016, 1(4): 43-50. (in Chinese) [20] LUO C T, HU Z M, ZHANG S L, et al. Adaptive space transformation: an invariant based method for predicting aerodynamic coefficients of hypersonic vehicles[J]. Engineering Applications of Artificial Intelligence, 2015, 46: 93-103. DOI: 10.1016/j.engappai.2015.09.001 [21] BREIMAN L. Random forests[J]. Machine learning, 2001, 45(1): 5-32. doi: 10.1023/A:1010933404324 -
点击查看大图
计量
- 文章访问数: 241
- HTML全文浏览量: 78
- PDF下载量: 32
- 被引次数: 0
