基于神经网络与CFD相结合的仿蜻蜓串联扑翼气动性能优化研究

钱广; 朱建阳; 蔡芸; 汪超; 徐启炎; 侯宇

doi:10.13224/j.cnki.jasp.20220796

基于神经网络与CFD相结合的仿蜻蜓串联扑翼气动性能优化研究

doi: 10.13224/j.cnki.jasp.20220796

钱广^1,,
朱建阳^1,*, ,,
蔡芸¹,
汪超²,
徐启炎¹,
侯宇¹

1.
武汉科技大学机械自动化学院冶金装备及其控制省部共建教育部重点实验室，武汉 430081
2.
东莞理工学院机械工程学院，广东东莞 523808

基金项目: 国家自然科学基金（51975429，52005104）

详细信息

作者简介:
钱广（1998-），男，硕士生，主要从事扑翼飞行器样机研制

通讯作者:
朱建阳（1981-），男，教授、博士生导师，博士，主要从事扑翼飞行器机理和样机设计研究。E-mail：zhujy@wust.edu.cn

中图分类号: V211.3
计量
- 文章访问数: 629
- HTML浏览量: 215
- PDF量: 77
- 被引次数: 0
出版历程
- 收稿日期: 2022-10-17
- 网络出版日期: 2024-11-11

Research on the aerodynamic performance optimization of dragonfly-inspired tandem flapping wing based on neural network and CFD

QIAN Guang^1
,,
ZHU Jianyang^{1,*
, ,},
CAI Yun¹,
WANG Chao²,
XU Qiyan¹,
HOU Yu¹

1.
Key Laboratory of Metallurgical Equipment and Control Technology of Ministry of Education，School of Machinery and Automation，Wuhan University of Science and Technology，Wuhan 430081，China
2.
School of Mechanical Engineering，Dongguan University of Technology，Dongguan Guangdong 523808，China

摘要

摘要:
为了提升仿蜻蜓串联扑翼的气动性能，采用神经网络与CFD相结合的方法，系统分析了扭转角幅值，翼间距和前后翼相位差变化对仿蜻蜓串联扑翼升举效率的影响。研究结果表明：扭转角幅值，翼间距和前后翼相位差变化对仿蜻蜓串联扑翼的气动性能具有重要影响。就所研究的参数范围，通过神经网络优化后，最优和最差参数组合扑翼的升举效率相差90.33%。进一步通过对不同参数组合仿蜻蜓串联扑翼的流场分析，发现优化参数组合的串联扑翼前翼脱落的尾涡可以重新附着在后翼表面，减弱后翼上冲程时的涡旋强度，降低扑翼的能量消耗，从而使扑翼获得更好的气动性能。
- 串联扑翼 /
- 神经网络 /
- 升举效率 /
- CFD数值模拟 /
- 田口方法
Abstract:
In order to improve the aerodynamic performance of dragonfly-inspired tandem flapping wings, the influences of three parameters, i.e. pitching amplitude, wing spacing and phase difference between forewing and hindwing, on the lifting efficiency of dragonfly-inspired tandem flapping wing were systematically analyzed by the combination of the neural network and CFD. The results showed that pitching amplitude, wing spacing and phase difference between forewing and hindwing had important influence on the aerodynamic performance of dragonfly-inspired tandem flapping wings. In the analysis parameter range, compared with the flapping wing with worst parameter combination, the lifting efficiency of the flapping wing with the best parameter combinations increased 90.33% by using neural network optimization. Furthermore, through the analysis of flow field of dragonfly-inspired tandem flapping wing with different parameter combinations, it was found that the trailing vortices shedding from the forewing can be reattached to the surface of the hindwing for the tandem flapping wing with the optimal combination of parameter, which can weaken the vortex intensity during the upstroke of the hindwing and reduce the energy consumption of the flapping wing, so that the flapping wing can generate better aerodynamic performance.
- tandem flapping wing /
- neural network /
- lifting efficiency /
- CFD numerical simulation /
- taguchi method

HTML

图 1 扑翼运动坐标系

Figure 1. Moving coordinate system of flapping wing

下载: 全尺寸图片幻灯片

图 2 仿蜻蜓翅翼模型

Figure 2. Dragonfly-inspired wing model

下载: 全尺寸图片幻灯片

图 3 计算域和边界条件示意图

Figure 3. Schematic of computational domain and boundary conditions

下载: 全尺寸图片幻灯片

图 4 不同网格密度下的扑翼升力和阻力系数

Figure 4. Evolution of lift and drag coefficient of double flapping wings for different grid densities

下载: 全尺寸图片幻灯片

图 5 不同时间步长下的扑翼升力和阻力系数

Figure 5. Evolution of lift and drag coefficient of double flapping wings for different time steps

下载: 全尺寸图片幻灯片

图 6 本文数值方法计算的升力和推力与文献数据的对比

Figure 6. Comparison of the lift and thrust of the wing obtained by the literature data and present numerical method

下载: 全尺寸图片幻灯片

图 7 BP神经网络结构

Figure 7. Structure of BP neural network

下载: 全尺寸图片幻灯片

图 8 各参数水平的平均信噪比

Figure 8. Average signal-to-noise ratio of each parameter level

下载: 全尺寸图片幻灯片

图 9 各参数对升举效率的影响程度

Figure 9. Influence range of each parameter on lifting efficiency

下载: 全尺寸图片幻灯片

图 10 训练过程中神经网络的误差变化

Figure 10. Error variation during neural network training

下载: 全尺寸图片幻灯片

图 11 两组试验的升力系数

Figure 11. Lift coefficient of two groups of trials

下载: 全尺寸图片幻灯片

图 12 两组试验的阻力系数

Figure 12. Drag coefficient of two groups of trials

下载: 全尺寸图片幻灯片

图 13 两组试验的能耗系数

Figure 13. Energy consumption coefficient of two groups of trials

下载: 全尺寸图片幻灯片

图 14 试验3和试验27扑动和扭转运动产生的力矩

Figure 14. Moment produced by flapping and pitching motion in trial 3 and trial 27

下载: 全尺寸图片幻灯片

图 15 试验3和试验27在t=0.75T, z=1.0c, 1.7c, 2.3c处的涡量、压力等值线图

Figure 15. Contour plots of vorticity and pressure for trial 3 and trial 27 at t=0.75T, z=1.0c, 1.7c, 2.3c

下载: 全尺寸图片幻灯片

图 16 试验3和试验27在不同时刻的流线图

Figure 16. Streamline chart of the trial 3 and trial 27 at different times

下载: 全尺寸图片幻灯片

图 17 试验3和试验27在一个运动周期内的三维涡流结构

Figure 17. Three-dimensional vortex structures in the flow for the trial 3 and trial 27 in one motion period

下载: 全尺寸图片幻灯片

表 1 扑翼几何参数

Table 1. Geometric parameters of flapping wing

参数	数值
b/m	0.2
c/m	0.06
b_r/m	0.014
S/m²	0.010673

下载: 导出CSV

表 2 网格密度细节

Table 2. Details of grid density

网格组号	扑翼表面第1层网格高度	内球域网格数/10⁵	总网格数/ 10⁵
Grid 1	b/33	1.19	2.47
Grid 2	b/44	2.09	4.37
Grid 3	b/54	3.07	6.45

下载: 导出CSV

表 3 因素水平表

Table 3. Factors and levels

因素	水平
因素	1	2	3	4	5
φ/（°）	0	45	90	135	180
l	0.5c	1.0c	1.5c	2.0c	2.5c
β_m	5	10	15	20	25

下载: 导出CSV

表 4 田口试验正交表

Table 4. Orthogonal table of Taguchi test

试验序号	φ/（°）	l	β_m/（°）
1	0	0.5	5
2	45	1.0	5
3	90	1.5	5
4	135	2.0	5
5	180	2.5	5
6	45	0.5	10
7	90	1.0	10
8	135	1.5	10
9	180	2.0	10
10	0	2.5	10
11	90	0.5	15
12	135	1.0	15
13	180	1.5	15
14	0	2.0	15
15	45	2.5	15
16	135	0.5	20
17	180	1.0	20
18	0	1.5	20
19	45	2.0	20
20	90	2.5	20
21	180	0.5	25
22	0	1.0	25
23	45	1.5	25
24	90	2.0	25
25	135	2.5	25

下载: 导出CSV

表 5 试验结果

Table 5. Test results

试验序号	$ \overline {{C_{\mathrm{l}}}} $	$ - \overline {{C_{\mathrm{d}}}} $	$\overline {{C_{\mathrm{p}}}} $	η_l/%	（S/N）/dB
1	0.4246	0.4046	5.0860	8.35	18.43
2	0.3765	0.4087	5.0700	7.43	17.42
3	0.2892	0.3146	4.0503	7.14	17.07
4	0.2888	0.2622	3.4968	8.26	18.34
5	0.3575	0.3257	4.1704	8.57	18.66
6	0.4195	0.7743	4.8484	8.65	18.74
7	0.3377	0.6711	4.2677	7.91	17.97
8	0.3237	0.4982	3.2905	9.84	19.86
9	0.3271	0.5101	3.3418	9.79	19.81
10	0.3101	0.6155	3.9363	7.88	17.93
11	0.3977	1.0058	4.2796	9.29	19.36
12	0.3151	0.7655	3.3842	9.31	19.38
13	0.3048	0.6311	2.8240	10.79	20.66
14	0.3480	0.9431	4.0389	8.62	18.71
15	0.3160	0.7340	3.2314	9.78	19.80
16	0.3513	1.0565	3.4840	10.08	20.07
17	0.3105	0.7727	2.6365	11.78	21.42
18	0.3948	1.2062	3.8859	10.16	20.14
19	0.3229	0.9798	3.2528	9.93	19.94
20	0.3028	0.8068	2.7241	11.11	20.92
21	0.3204	0.9509	2.6204	12.23	21.75
22	0.4107	1.3274	3.4496	11.91	21.52
23	0.3650	1.1998	3.2072	11.38	21.12
24	0.3120	0.9111	2.5082	12.44	21.90
25	0.3106	0.9154	2.4569	12.64	22.04

下载: 导出CSV

表 6 寻优试验结果

Table 6. Test of optimization results

编号	β_m	l	φ	仿真值/%	预测值/%	误差/%
26	25	1.5	180	13.36	13.35	0.06
27	25	2.0	180	13.59	13.45	1.03
28	25	2.5	180	12.95	13.24	2.22

下载: 导出CSV

参考文献(20)

[1]	谢鹏,姜洪利,周超英. 一种仿生扑翼飞行器的设计及动力学分析[J]. 航空动力学报,2018,33(3): 703-710. XIE Peng,JIANG Hongli,ZHOU Chaoying. Design and dynamic analysis of a flapping wing air vehicle[J]. Journal of Aerospace Power,2018,33(3): 703-710. (in Chinese XIE Peng, JIANG Hongli, ZHOU Chaoying. Design and dynamic analysis of a flapping wing air vehicle[J]. Journal of Aerospace Power, 2018, 33(3): 703-710. (in Chinese)
[2]	TRUONG Q T,ARGYOGANENDRO B W,PARK H C. Design and demonstration of insect mimicking foldable artificial wing using four-bar linkage systems[J]. Journal of Bionic Engineering,2014,11(3): 449-458. doi: 10.1016/S1672-6529(14)60057-3
[3]	CHEN Chen,ZHANG Tianyu. A review of design and fabrication of the bionic flapping wing micro air vehicles[J]. Micromachines,2019,10(2): 144. doi: 10.3390/mi10020144
[4]	HAIDER N,SHAHZAD A,MUMTAZ QADRI M N,et al. Recent progress in flapping wings for micro aerial vehicle applications[J]. Proceedings of the Institution of Mechanical Engineers,Part C: Journal of Mechanical Engineering Science,2021,235(2): 245-264. doi: 10.1177/0954406220917426
[5]	VIIERU D,TANG Jian,LIAN Yongsheng,et al. Flapping and flexible wing aerodynamics of low Reynolds number flight vehicles[R]Reno,US: 44th AIAA Aerospace Sciences Meeting and Exhibit,2006.
[6]	XU Wenfu,PAN Erzhen,LIU Juntao,et al. Flight control of a large-scale flapping-wing flying robotic bird: system development and flight experiment[J]. Chinese Journal of Aeronautics,2022,35(2): 235-249. doi: 10.1016/j.cja.2021.03.009
[7]	ALEXANDER D E. Unusual phase relationships between the forewings and hindwings in flying dragonflies[J]. Journal of Experimental Biology,1984,109(1): 379-383. doi: 10.1242/jeb.109.1.379
[8]	LI Qiushi,ZHENG Mengzong,PAN Tianyu,et al. Experimental and numerical investigation on dragonfly wing and body motion during voluntary take-off[J]. Scientific Reports,2018,8: 1011. doi: 10.1038/s41598-018-19237-w
[9]	AZUMA A,WATANABE T. Flight performance of a dragonfly[J]. Journal of Experimental Biology,1988,137(1): 221-252. doi: 10.1242/jeb.137.1.221
[10]	WANG Z J,RUSSELL D. Effect of forewing and hindwing interactions on aerodynamic forces and power in hovering dragonfly flight[J]. Physical Review Letters,2007,99(14): 148101. doi: 10.1103/PhysRevLett.99.148101
[11]	BIE Dawei,LI Daochun. Numerical analysis of the wing-wake interaction of tandem flapping wings in forward flight[J]. Aerospace Science and Technology,2022,121: 107389. doi: 10.1016/j.ast.2022.107389
[12]	BROERING T M,LIAN Yongsheng. The effect of phase angle and wing spacing on tandem flapping wings[J]. Acta Mechanica Sinica,2012,28(6): 1557-1571. doi: 10.1007/s10409-012-0210-8
[13]	GONG Wuqi,JIA Bobo,XI Guang. Experimental study on instantaneous thrust and lift of two plunging wings in tandem[J]. Experiments in Fluids,2015,57(1): 8.
[14]	NAGAI H,FUJITA K,MUROZONO M. Experimental study on forewing-hindwing phasing in hovering and forward flapping flight[J]. AIAA Journal,2019,57(9): 3779-3790. doi: 10.2514/1.J058335
[15]	WARKENTIN J,DELAURIER J. Experimental aerodynamic study of tandem flapping membrane wings[J]. Journal of Aircraft,2007,44(5): 1653-1661. doi: 10.2514/1.28160
[16]	HO S,NASSEF H,PORNSINSIRIRAK N,et al. Unsteady aerodynamics and flow control for flapping wing flyers[J]. Progress in Aerospace Sciences,2003,39(8): 635-681. doi: 10.1016/j.paerosci.2003.04.001
[17]	TONG L I,SU C T. Optimizing multi-response problems in the taguchi method by fuzzy multiple attribute decision making[J]. Quality and Reliability Engineering International,1997,13(1): 25-34. doi: 10.1002/(SICI)1099-1638(199701)13:1<25::AID-QRE59>3.0.CO;2-B
[18]	ZHU Jianyang,ZHU Mingkang,ZHANG Tao,et al. Improvement of the power extraction performance of a semi-active flapping airfoil by employing two-sided symmetric slot airfoil[J]. Energy,2021,227: 120458. doi: 10.1016/j.energy.2021.120458
[19]	HORNIK K,STINCHCOMBE M,WHITE H. Multilayer feedforward networks are universal approximators[J]. Neural Networks,1989,2(5): 359-366. doi: 10.1016/0893-6080(89)90020-8
[20]	张立明. 人工神经网络的模型及其应用[M]. 上海: 复旦大学出版社,1993. ZHANG Liming. Models and applications of artificial neural networks[M]. Shanghai: Fudan Press,1993. (in Chinese ZHANG Liming. Models and applications of artificial neural networks[M]. Shanghai: Fudan Press, 1993. (in Chinese)