基于深度神经网络的客机总体设计参数敏感性分析

doi:10.7527/S1000-6893.2020.24353

固体力学与飞行器总体设计

本期目录 | 过刊浏览 | 高级检索

前一篇 | 后一篇

基于深度神经网络的客机总体设计参数敏感性分析

范周伟¹, 余雄庆¹, 王朝², 钟伯文²

1. 南京航空航天大学航空学院飞行器先进设计技术国防重点学科实验室, 南京 210016;
2. 中国商飞北京民用飞机技术研究中心, 北京 102211

收稿日期:2020-06-01 修回日期:2020-08-18 发布日期:2020-09-02
通讯作者: 余雄庆 E-mail:yxq@nuaa.edu.cn
基金资助:
中国商飞北京民用飞机技术研究中心民用飞机设计数字仿真技术北京市重点实验室开放课题

Sensitivity analysis of key design parameters of commercial aircraft using deep neural network

FAN Zhouwei¹, YU Xiongqing¹, WANG Chao², ZHONG Bowen²

1. Key Laboratory of Fundamental Science for National Defense-Advanced Design Technology of Flight Vehicle, College of Aerospace Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China;
2. Beijing Aeronautical Science & Technology Research Institute, COMAC, Beijing 102211, China

Received:2020-06-01 Revised:2020-08-18 Published:2020-09-02
Supported by:
Funded by Beijing Key Laboratory of Civil Aircraft Design and Simulation Technology, Beijing Aeronautical Science & Technology Research Institute, Commercial Aircraft Corporation of China

摘要/Abstract

摘要： 飞机总体主要设计参数敏感性分析揭示了总体主要设计参数对飞机特性指标的影响，有助于总体设计方案的决策。针对宽体客机总体主要设计参数敏感性问题，根据其总体主要设计参数和特性指标的特点，以及多学科间的耦合关系，建立了深度神经网络模型。该深度神经网络模型以客机总体主要设计参数为输入，对特性指标进行预测。在深度神经网络模型中，设置了多个输入层、多个输出层以及多个分块的隐藏层，从而模拟客机总体主要设计参数对特性指标的影响以及不同特性指标之间的相互作用。测试结果表明，与传统代理模型相比，深度神经网络模型对客机特性指标的预测精度更高，多参数适应性更好。利用该深度神经网络模型对客机总体主要设计参数进行敏感性分析。分析结果表明，机翼1/4弦线后掠角在30°~31.5°时，有利于减少最大起飞重量和起飞平衡场长；发动机海平面最大静推力和机翼面积对客机直接使用成本、最大起飞重量等特性指标的影响最为显著。

关键词: 飞机总体设计, 敏感性分析, 深度神经网络, 客机, 代理模型

Abstract: The sensitivity analysis of key design parameters of aircraft reveals the relationship between the key design parameters and aircraft characteristics, facilitating the decision making in aircraft preliminary design. Aiming at the key design parameter sensitivity of wide-body commercial aircraft, we establish a deep neural network model based on the features of key design parameters and aircraft characteristics and the coupling relationship among multiple disciplines, taking the key design parameters as input to predict the aircraft characteristics. In this model, multiple input layers, multiple output layers, and multiple blocks of hidden layers are set to simulate the effects of key design parameters on aircraft characteristics and the interactions among different aircraft characteristics. Comparisons with traditional surrogate models reveal that the deep neural network model has higher prediction accuracy and better adaptability to the aircraft characteristics. The proposed model is then used to analyze the sensitivity of the commercial aircraft primary parameters. The analysis results show that a lower maximum takeoff weight and a shorter takeoff balanced field length can be achieved when the wing sweep at 1/4 chord is between 30° to 31.5°. The maximum static thrust of engines at sea level and the wing area have the most significant influence on the direct operation cost, maximum takeoff weight, and other characteristics.

Key words: aircraft conceptual design, sensitivity analysis, deep neural network, commercial aircraft, surrogate model

中图分类号:

V221

范周伟, 余雄庆, 王朝, 钟伯文. 基于深度神经网络的客机总体设计参数敏感性分析[J]. 航空学报, 2021, 42(4): 524353-524353.

FAN Zhouwei, YU Xiongqing, WANG Chao, ZHONG Bowen. Sensitivity analysis of key design parameters of commercial aircraft using deep neural network[J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(4): 524353-524353.

参考文献

[1] 张帅, 余雄庆. 中短程客机总体参数敏感性分析[J]. 航空学报, 2013, 34(4):809-816. ZHANG S, YU X Q. Sensitivity analysis of primary parameters in preliminary design of a short/medium-haul airliner[J]. Acta Aeronautica et Astronautica Sinica, 2013, 34(4):809-816(in Chinese).
[2] QUEIPO N V, HAFTKA R T, SHYY W, et al. Surrogate-based analysis and optimization[J]. Progress in Aerospace Sciences, 2005, 41(1):1-28.
[3] FORRESTER A, SOBESTER A, KEANE A. Engineering design via surrogate modelling:a practical guide[M]. Chichester:John Wiley & Sons, 2008:33-59.
[4] 张陈力子, 宋晓玉, 祝雯生, 等. 利用代理模型的客机总体参数敏感性分析[J]. 空军工程大学学报:自然科学版, 2016, 17(3):18-22. ZHANG C L Z, SONG X Y, ZHU W S, et al. Analysis of parameters sensitivity for airliner conceptual design using surrogate-model[J]. Journal of Air Force Engineering University:Natural Science Edition, 2016, 17(3):18-22(in Chinese).
[5] 张伟, 高正红, 周琳, 等. 基于代理模型全局优化的自适应参数化方法研究[J]. 航空学报, 2020, 41(10):123815. ZHANG W, GAO Z H, ZHOU L, et al. Efficient surrogate-based aerodynamic shape optimization with adaptive design space expansion[J]. Acta Aeronautica et Astronautica Sinica, 2020, 41(10):123815(in Chinese).
[6] BOUHLEL M A, BARTOLI N, OTSMANE A, et al. An improved approach for estimating the hyperparameters of the Kriging model for high-dimensional problems through the partial least squares method[J]. Mathematical Problems in Engineering, 2016, 2016:1-11.
[7] 武亮, 左向梅, 邱勇. 基于SVR多学科设计优化代理模型技术研究[J]. 飞行力学, 2020, 38(2):23-28. WU L, ZUO X M, QIU Y. Study of surrogate model technology in multidisciplinary design optimization based on SVR[J]. Flight Dynamics, 2020, 38(2):23-28(in Chinese).
[8] GOODFELLOW I, BENGIO Y, COURVILLE A. Deep learning[M]. Cambridge:The MIT Press, 2016:3-10.
[9] LING J, KURZAWSKI A, TEMPLETON J. Reynolds averaged turbulence modelling using deep neural networks with embedded invariance[J]. Journal of Fluid Mechanics, 2016, 807:155-166.
[10] DENG L, WANG Y, LIU Y, et al. A CNN-based vortex identification method[J]. Journal of Visualization, 2019, 22(1):65-78.
[11] 陈海, 钱炜祺, 何磊. 基于深度学习的翼型气动系数预测[J]. 空气动力学学报, 2018, 36(2):294-299. CHEN H, QIAN W Q, HE L. Aerodynamic coefficient prediction of airfoils based on deep learning[J]. Acta Aerodynamic Sinica, 2018, 36(2):294-299(in Chinese).
[12] 周旺旺, 姚佩阳, 张杰勇, 等. 基于深度神经网络的空中目标作战意图识别[J]. 航空学报, 2018, 39(11):322468. ZHOU W W, YAO P Y, ZHANG J Y, et al. Combat intention recognition for aerial targets based on deep neural network[J]. Acta Aeronautica et Astronautica Sinica, 2018, 39(11):322468(in Chinese).
[13] 王宇, 余雄庆. 考虑不确定性的飞机总体参数优化方法[J]. 航空学报, 2009, 30(10):1883-1888. WANG Y, YU X Q. Optimization method for aircraft conceptual design under uncertainty[J]. Acta Aeronautica et Astronautica Sinica, 2009, 30(10):1883-1888(in Chinese).
[14] CHAI X, YU X, WANG Y. Multipoint optimization on fuel efficiency in conceptual design of wide-body air-craft[J]. Chinese Journal of Aeronautics, 2018, 31(1):99-106.
[15] GOOGLE. TensorFlow:An end-to-end open source machine learning platform[EB/OL]. (2019-8-1)[2019-8-10]. https://www.tensorflow.org.
[16] NAIR V, HINTON G E. Rectified linear units improve restricted boltzmann machines[C]//Proceedings of the 27th International Conference on International Conference on Machine (ICML-10),2010:807-814.
[17] NEWEY W K. Adaptive estimation of regression models via moment restrictions[J]. Journal of Econometrics, 1988, 38(3):301-339.
[18] IOFFE S, SZEGEDY C. Batch normalization:accelerating deep network training by reducing internal covariate shift[DB/OL]. arXiv preprint:1502.03167, 2015.
[19] MICROSOFT. Neural Network Intelligence[EB/OL]. (2019-08-05)[2019-09-01]. https://github.com/microsoft/nni.
[20] BERGSTRA J, BARDENET R, BENGIO Y, et al. Algorithms for hyper-parameter optimization[C]//Advances in Neural Information Processing Systems, 2011:2546-2554.
[21] HINTON G E, SRIVASTAVA N, KRIZHEVSKY A, et al. Improving neural networks by preventing co-adaptation of feature detectors[DB/OL]. arXiv preprint:1207.0580, 2012.
[22] BOUHLEL M A, HWANG J T, BARTOLI N, et al. A Python surrogate modeling framework with derivatives[J]. Advances in Engineering Software, 2019, 135:102662.
[23] 闻新, 李新, 张兴旺, 等. 应用MATLAB实现神经网络[M]. 北京:国防工业出版社, 2015:96-137. WEN X, LI X, ZHANG X W, et al. Application of MATLAB neural network[M]. Beijing:National Defense Industry Press, 2015:96-137(in Chinese).
[24] 杨哲, 李曙林, 周莉, 等. 飞机作战生存力设计参数灵敏度分析[J]. 北京航空航天大学学报, 2013, 39(8):1096-1101. YANG Z, LI S L, ZHOU L, et al. Design parameters sensitivity analysis of aircraft survivability[J]. Journal of Beijing University of Aeronautics and Astronautics, 2013, 39(8):1096-1101(in Chinese).
[25] WILKINSON L. Revising the Pareto chart[J]. The American Statistician, 2006, 60(4):332-334.

E-mail：hkxb@buaa.edu.cn

关于我们

期刊社服务

专业学科

封面文章

友情链接

主管单位：中国科学技术协会主办单位：中国航空学会北京航空航天大学

基于深度神经网络的客机总体设计参数敏感性分析

Sensitivity analysis of key design parameters of commercial aircraft using deep neural network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	崔壮壮, 原昕, 赵国庆, 井思梦, 招启军. 共轴刚性旋翼高速直升机前飞性能操纵策略影响[J]. 航空学报, 2024, 45(9): 529256-529256.
[2]	柳家齐, 陈荣钱, 楼锦华, 韩旭, 吴昊, 尤延铖. 基于深度学习的高速直升机旋翼翼型气动优化设计[J]. 航空学报, 2024, 45(9): 529828-529828.
[3]	陈树生, 冯聪, 张兆康, 赵轲, 张新洋, 高正红. 基于全局/梯度优化方法的宽速域乘波-机翼布局气动设计[J]. 航空学报, 2024, 45(6): 629596-629596.
[4]	李玉涵, 杨宝玉, 吴亦农, 张强, 唐晓. 卫星光机载荷热模型参数高效修正方法研究进展[J]. 航空学报, 2024, 45(6): 628814-628814.
[5]	周贵荣, 徐见源, 马少博, 宗军耀, 沈金清, 朱海杰. 大型客机航电系统综合集成关键技术综述[J]. 航空学报, 2024, 45(5): 529956-529956.
[6]	姜乐, 陈以彪, 李炎军, 李贵林, 刘涛. 航空发动机增压式离心通风器流动与分离特性[J]. 航空学报, 2024, 45(2): 128675-128675.
[7]	张恺玲, 李思怡, 段毅, 阎超. 进气道流动中SST湍流模型参数的不确定度量化[J]. 航空学报, 2023, 44(S2): 729429-729429.
[8]	孙晓哲, 侯东, 杨建忠. 双余度机电作动器力纷争机理及敏感性[J]. 航空学报, 2023, 44(S1): 727661-727661.
[9]	李忠智, 马金毅, 艾剑良, 董一群. 拟VGG16网络的航空传感器故障检测分类[J]. 航空学报, 2023, 44(S1): 727615-727615.
[10]	赵欢, 高正红, 夏露. 基于新型多可信度代理模型的多目标优化方法[J]. 航空学报, 2023, 44(6): 126962-126962.
[11]	王志凯, 陈盛, 范玮. 神经网络宽度对燃烧室排放预测的影响[J]. 航空学报, 2023, 44(5): 126816-126816.
[12]	赵欢, 高正红, 夏露. 基于新型高维代理模型的气动外形设计方法[J]. 航空学报, 2023, 44(5): 126924-126924.
[13]	郑志阳, 张阳, 张钊, 吴宝海, 张莹. 基于GA⁃SVR的薄壁叶片辅助支撑布局优化方法[J]. 航空学报, 2023, 44(4): 426805-426805.
[14]	于汀, 李璐祎, 刘昱杉, 常泽明. 观测不确定性下的高效贝叶斯更新方法及其在机翼结构中的应用[J]. 航空学报, 2023, 44(24): 228592-228592.
[15]	顾奕然, 黄江涛, 陈树生, 刘德园, 高正红. 基于逆向增广Burgers方程的声爆反演技术[J]. 航空学报, 2023, 44(2): 626258-626258）.