基于RBF神经网络的旋翼桨叶气动外形优化

王清; 陆波; 王亮权

doi:10.13224/j.cnki.jasp.20240182

基于RBF神经网络的旋翼桨叶气动外形优化

doi: 10.13224/j.cnki.jasp.20240182

1.
兰州理工大学能源与动力工程学院，兰州 730050
2.
中国空气动力研究与发展中心旋翼空气动力学重点实验室，四川绵阳 621000

基金项目: 中国空气动力研究与发展中心旋翼空气动力学重点实验室研究开放课题（RAL202201）；兰州理工大学红柳优秀青年人才支持计划（2022）

详细信息

作者简介:
王清（1986-），男，副教授，博士，研究方向为旋翼空气动力学。E-mail：wangqing_lut@foxmail.com

中图分类号: V211.3
计量
- 文章访问数: 416
- HTML浏览量: 340
- PDF量: 30
- 被引次数: 0
出版历程
- 收稿日期: 2024-03-27
- 网络出版日期: 2025-10-13

Rotor blade aerodynamic shape optimization based on RBF neural network

1.
College of Energy and Power Engineering，Lanzhou University of Technology，Lanzhou 730050，China
2.
Key Laboratory of Rotor Aerodynamics，China Aerodynamics Research and Development Center，Mianyang Sichuan 621000，China

摘要

摘要:
良好的直升机旋翼桨叶气动外形能有效提高旋翼的气动性能，但复杂桨叶外形具有多自由度、高度非线性的特点，传统的梯度优化算法容易陷入局部最优陷阱。针对这类问题，结合RBF神经网络方法、遗传算法和CFD方法，建立了适用于直升机悬停状态下的旋翼桨叶气动外形优化设计方法，可在较小计算量的同时获得全局优化结果。在此基础上，针对Helishape 7A旋翼桨叶外形开展了气动外形优化设计研究。优化桨叶具有前-后掠组合的形式，数值计算结果表明在相同的拉力系数下转矩系数减小了3.81%，因此优化旋翼的悬停效率得到了有效地提升，最大悬停效率提高了3.99%，表明采用优化桨叶的旋翼具有更好的悬停性能，能够有效提升直升机的起飞载重。
- 神经网络 /
- 直升机 /
- 旋翼 /
- 气动外形优化 /
- 计算流体力学
Abstract:
A good aerodynamic profile of a helicopter rotor blade can effectively improve its aerodynamic performance, but complex blade profiles are characterized by many degrees of freedom and high nonlinearity, so the traditional gradient optimization algorithm is vunerable to fall into the local optimal trap. To solve these problems, an optimal design method of rotor blade aerodynamic profile was established by combining RBF neural network method, genetic algorithm and computation fluid dynamics (CFD) method, which can obtain global optimization results in a small amount of computation. On this basis, the aerodynamic shape optimization design of Helishape 7A rotor blade was studied. The optimized blade had the form of front-sweep combination. The numerical calculation results showed that the torque coefficient was reduced by 3.81% under the same tension coefficient, so the hover efficiency of the optimized rotor was effectively improved, and the maximum hover efficiency was increased by 3.99%, indicating that the rotor with optimized blade had better hover performance and can effectively improve the takeoff load of the helicopter.
- neural network /
- helicopter /
- rotor /
- aerodynamic shape optimization /
- computation fluid dynamics

HTML

图 1 RBF神经网络拓扑结构

Figure 1. Topology of RBF neural network

下载: 全尺寸图片幻灯片

图 2 基于RBF神经网络的优化设计框架

Figure 2. Optimization design framework based on RBF neural network

下载: 全尺寸图片幻灯片

图 3 运动嵌套网格系统

Figure 3. Moving embedded mesh system

下载: 全尺寸图片幻灯片

图 4 Helishape 7A旋翼桨叶外形

Figure 4. Rotor blade profile of Helishape 7A

下载: 全尺寸图片幻灯片

图 5 Helishape 7A桨叶不同剖面位置处的压强系数对比

Figure 5. Comparison of C_p at different profile positions of Helishape 7A blades

下载: 全尺寸图片幻灯片

图 6 RBF神经网络函数拟合

Figure 6. Function fitting of RBF neural network

下载: 全尺寸图片幻灯片

图 7 桨叶前、后缘设计空间

Figure 7. Design space in leading and trailing edge

下载: 全尺寸图片幻灯片

图 8 RBF神经网络回归分析

Figure 8. Regression analysis of RBF neural network

下载: 全尺寸图片幻灯片

图 9 旋翼桨叶优化结果

Figure 9. Rotor blade optimization results

下载: 全尺寸图片幻灯片

图 10 优化旋翼及基准旋翼气动特性对比

Figure 10. Comparison of aerodynamic characteristics between the optimized and baseline rotor

下载: 全尺寸图片幻灯片

图 11 不同桨叶压强分布对比（C_T=0.00789）

Figure 11. Comparison of pressure distribution of different blades （C_T=0.00789）

下载: 全尺寸图片幻灯片

图 12 优化桨叶及基准桨叶压强系数对比（C_T=0.00789）

Figure 12. Comparison of C_p between the optimized and baseline blade （C_T=0.00789）

下载: 全尺寸图片幻灯片

图 13 不同旋翼桨叶的法向力系数分布对比（C_T=0.00789）

Figure 13. Comparison of C_n distribution of different rotor blades （C_T=0.00789）

下载: 全尺寸图片幻灯片

图 14 不同旋翼桨尖涡的对比

Figure 14. Comparison of blade tip vortex between different rotors

下载: 全尺寸图片幻灯片

表 1 遗传算法参数设置

Table 1. Parameter setting of GA

参数	数值
交叉概率	0.9
变异概率	0.05
种群规模	100000
进化代数	100

下载: 导出CSV

表 2 设计变量取值范围

Table 2. Range of design variable

设计变量	展向位置	上边界	下边界
P₁	0.9R	0.25c	−0.25c
P₂	0.95R	0.5c	−0.5c
P₃	1.0R	0	−0.75c
P₄	0.9R	0.25c	−0.25c
P₅	0.95R	0.25c	−0.5c
P₆	1.0R	0	−0.75c
n₁	0.75R	2.0°	−4.0°
n₂	0.9R	2.0°	−5.0°
n₃	1.0R	3.0°	−6.0°

下载: 导出CSV

表 3 优化桨叶与基准桨叶气动载荷特性对比

Table 3. Comparison of aerodynamic load characteristics between optimized blade and baseline blade

桨叶类型	F_m	C_T/10⁻³	C_Q/10⁻⁴
优化桨叶	0.751	7.87	6.57
基准桨叶	0.726	7.89	6.83

下载: 导出CSV

参考文献(26)

[1]	LEISHMAN J G. Principles of helicopter aerodynamics[M]. 2nd. New York: Cambridge University Press, 2006.
[2]	LORBER P F, STAUTER R C, LANDGREBE A J. A comprehensive hover test of the airloads and airflow of an extensively instrumented model helicopter rotor[R]. Boston, US: Proceedings of the 45th Annual Forum of the American Helicopter Society, 1989.
[3]	BROCKLEHURST A, BARAKOS G N. A review of helicopter rotor blade tip shapes[J]. Progress in Aerospace Sciences, 2013, 56: 35-74. doi: 10.1016/j.paerosci.2012.06.003
[4]	PERRY F J. Aerodynamics of the world speed record[R]. Louis, US: Proceedings of the 43rd Annual Forum of the American Helicopter Society, 1987.
[5]	WALSH J L, BINGHAM G J, RILEY M F. Optimization methods applied to the aerodynamic design of helicopter rotor blades[J]. Journal of the American Helicopter Society, 1987, 32(4): 39-44. doi: 10.4050/JAHS.32.39
[6]	QUACKENBUSH T R, WACHSPRESS D A, KAUFMAN A E. Optimization of rotor performance in hover using a free wake analysis[J]. Journal of Aircraft, 1991, 28(3): 200-207. doi: 10.2514/3.46013
[7]	LE PAPE A, BEAUMIER P. Numerical optimization of helicopter rotor aerodynamic performance in hover[J]. Aerospace Science and Technology, 2005, 9(3): 191-201. doi: 10.1016/j.ast.2004.09.004
[8]	LEE S W, KWON O J. Aerodynamic shape optimization of hovering rotor blades in transonic flow using unstructured meshes[J]. AIAA Journal, 2006, 44(8): 1816-1825. doi: 10.2514/1.15385
[9]	孙伟, 张呈林. 直升机桨叶气动外形多目标优化设计[J]. 航空动力学报, 2011, 26(7): 1608-1614. SUN Wei, ZHANG Chenglin. Multi-objective optimization for aerodynamic shape of helicopter blade[J]. Journal of Aerospace Power, 2011, 26(7): 1608-1614. (in Chinese SUN Wei, ZHANG Chenglin. Multi-objective optimization for aerodynamic shape of helicopter blade[J]. Journal of Aerospace Power, 2011, 26(7): 1608-1614. (in Chinese)
[10]	ZHAO Qijun, XU Guohua. A study on aerodynamic and acoustic characteristics of advanced tip-shape rotors[J]. Journal of the American Helicopter Society, 2007, 52(3): 201-213. doi: 10.4050/JAHS.52.201
[11]	RAUCH P, GERVAIS M, GRANGA P, et al. Blue edge^TM: the design, development and testing of a new blade concept[R]. Virginia Beach, US: Proceedings of the 67th Annual Forum of the American Helicopter Society, 2011.
[12]	宋居正, 李耕耘. 基于遗传算法的横向减阻沟槽优化及机理[J]. 航空动力学报, 2024, 39(5): 20220387. SONG Juzheng, LI Gengyun. Shape optimization and mechanism of transverse groove for drag reduction based on genetic algorithm[J]. Journal of Aerospace Power, 2024, 39(5): 20220387. (in Chinese SONG Juzheng, LI Gengyun. Shape optimization and mechanism of transverse groove for drag reduction based on genetic algorithm[J]. Journal of Aerospace Power, 2024, 39(5): 20220387. (in Chinese)
[13]	王宗辉, 杨云军, 赵弘睿, 等. 多飞行状态倾转旋翼气动优化设计[J]. 航空学报, 2024, 45(9): 529024. WANG Zonghui, YANG Yunjun, ZHAO Hongrui, et al. Aerodynamic optimization design of tiltrotor under multiple flight conditions[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(9): 529024. (in Chinese WANG Zonghui, YANG Yunjun, ZHAO Hongrui, et al. Aerodynamic optimization design of tiltrotor under multiple flight conditions[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(9): 529024. (in Chinese)
[14]	HAN Zhonghua, ZHANG Yu, SONG Chenxing, et al. Weighted gradient-enhanced Kriging for high-dimensional surrogate modeling and design optimization[J]. AIAA Journal, 2017, 55(12): 4330-4346. doi: 10.2514/1.J055842
[15]	许瑞飞, 段卓毅, 钱瑞战. 基于Kriging代理模型的高升力构型优化设计[J]. 航空科学技术, 2023, 34(3): 58-63. XU Ruifei, DUAN Zhuoyi, QIAN Ruizhan. Optimization design of high-lift configuration using Kriging model[J]. Aeronautical Science & Technology, 2023, 34(3): 58-63. (in Chinese XU Ruifei, DUAN Zhuoyi, QIAN Ruizhan. Optimization design of high-lift configuration using Kriging model[J]. Aeronautical Science & Technology, 2023, 34(3): 58-63. (in Chinese)
[16]	陈晨铭, 郭雪岩, 常林森. 基于BP神经网络代理模型的翼型优化及领域自适应研究[J]. 动力工程学报, 2022, 42(7): 657-663. CHEN Chenming, GUO Xueyan, CHANG Linsen. Research on airfoil optimization and domain adaptation based on BP neural network surrogate model[J]. Journal of Chinese Society of Power Engineering, 2022, 42(7): 657-663. (in Chinese CHEN Chenming, GUO Xueyan, CHANG Linsen. Research on airfoil optimization and domain adaptation based on BP neural network surrogate model[J]. Journal of Chinese Society of Power Engineering, 2022, 42(7): 657-663. (in Chinese)
[17]	李超群, 唐硕, 李易, 等. 基于神经网络的减阻沟槽壁面形状优化[J]. 航空动力学报, 2022, 37(3): 639-648. LI Chaoqun, TANG Shuo, LI Yi, et al. Sub-optimization of riblet shape based on neural networks[J]. Journal of Aerospace Power, 2022, 37(3): 639-648. (in Chinese LI Chaoqun, TANG Shuo, LI Yi, et al. Sub-optimization of riblet shape based on neural networks[J]. Journal of Aerospace Power, 2022, 37(3): 639-648. (in Chinese)
[18]	陈海昕, 邓凯文, 李润泽. 机器学习技术在气动优化中的应用[J]. 航空学报, 2019, 40(1): 522480. CHEN Haixin, DENG Kaiwen, LI Runze. Utilization of machine learning technology in aerodynamic optimization[J]. Acta Aeronautica et Astronautica Sinica, 2019, 40(1): 522480. (in Chinese CHEN Haixin, DENG Kaiwen, LI Runze. Utilization of machine learning technology in aerodynamic optimization[J]. Acta Aeronautica et Astronautica Sinica, 2019, 40(1): 522480. (in Chinese)
[19]	柳家齐, 陈荣钱, 楼锦华, 等. 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. LIU Jiaqi, CHEN Rongqian, LOU Jinhua, et al. Aerodynamic shape optimization of high-speed helicopter rotor airfoil based on deep learning[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(9): 529828. (in Chinese LIU Jiaqi, CHEN Rongqian, LOU Jinhua, et al. Aerodynamic shape optimization of high-speed helicopter rotor airfoil based on deep learning[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(9): 529828. (in Chinese)
[20]	宣金婷, 丁志伟, 赵洪, 等. 火星六旋翼无人机旋翼气动外形稳健设计与优化[J]. 南京航空航天大学学报, 2024, 56(2): 264-272. XUAN Jinting, DING Zhiwei, ZHAO Hong, et al. Robust design and optimization of rotor aerodynamic shape of the Mars hexacopter[J]. Journal of Nanjing University of Aeronautics & Astronautics, 2024, 56(2): 264-272. (in Chinese XUAN Jinting, DING Zhiwei, ZHAO Hong, et al. Robust design and optimization of rotor aerodynamic shape of the Mars hexacopter[J]. Journal of Nanjing University of Aeronautics & Astronautics, 2024, 56(2): 264-272. (in Chinese)
[21]	TANG Dewei, TANG Bo, SHEN Wenqing, et al. On genetic algorithm and artificial neural network combined optimization for a Mars rotorcraft blade[J]. Acta Astronautica, 2023, 203: 78-87. doi: 10.1016/j.actaastro.2022.11.032
[22]	ROE P L. Approximate Riemann solvers, parameter vectors, and difference schemes[J]. Journal of Computational Physics, 1997, 135: 250-258. doi: 10.1006/jcph.1997.5705
[23]	HARTEN A, HYMAN J M. Self adjusting grid methods for one-dimensional hyperbolic conservation laws[J]. Journal of Computational Physics, 1983, 50(2): 235-269. doi: 10.1016/0021-9991(83)90066-9
[24]	MENTER F. Two-equation eddy-viscosity turbulence models for engineering applications[J]. AIAA Journal, 1994, 32(8): 1598-1605. doi: 10.2514/3.12149
[25]	STOLL P, GERLINGER P, BRUEGGEMANN D, et al. Domain decomposition for an implicit LU-SGS scheme using overlapping grids[R]. AIAA Paper 1997-1896, 1997.
[26]	POMIN H, WAGNER S. Navier-stokes analysis of helicopter rotor aerodynamics in hover and forward flight[J]. Journal of Aircraft, 2002, 39(5): 813-821. doi: 10.2514/2.3001