基于自适应径向基网络的热防护结构可靠性评估

董朋虎; 陈强; 李彦斌; 张旭东; 马晗; 费庆国

doi:10.6052/j.issn.1000-4750.2022.07.0640

基于自适应径向基网络的热防护结构可靠性评估

董朋虎^{1, 2,},
陈强^{1, 2,},
李彦斌^{1, 2, ,},
张旭东^3,,
马晗^{1, 2,},
费庆国^{1, 2,}

1.
东南大学机械工程学院，南京 211189
2.
江苏省空天机械装备工程研究中心，南京 211189
3.
北京空天技术研究所，北京 100074

基金项目: 国家自然科学基金项目(52125209，52175220，52302514)；江苏省自然科学基金项目(BK20211558)；江苏省博士后科研资助计划项目(2021K230B)；江苏省科技厅重点研发产业前瞻项目(BE2022158)

详细信息

作者简介:
董朋虎(1998−)，男，江苏人，硕士，主要从事热防护结构可靠性研究(E-mail: 220200303@seu.edu.cn)

陈　强(1985−)，男，山东人，助理研究员，博士，主要从事飞行器结构设计方面的研究(E-mail: qiang_chen@seu.edu.cn)

张旭东(1989−)，男，河北人，高工，硕士，主要从事飞行器结构设计方面的研究(E-mail: tjuzxd@126.com)

马　晗(1998−)，男，山东人，博士生，主要从事热防护结构动力学方面的研究(E-mail: 220200322@seu.edu.cn)

费庆国(1977−)，男，江苏人，教授，博士，博导，主要从事飞行器结构设计方面的研究(E-mail: qgFei@seu.edu.cn)

通讯作者:
李彦斌(1986−)，男，山西人，副教授，博士，博导，主要从事飞行器结构设计方面的研究(E-mail: LYB@seu.edu.cn)

中图分类号: V414
计量
- 文章访问数: 214
- HTML全文浏览量: 71
- PDF下载量: 33
出版历程
- 收稿日期: 2022-07-17
- 修回日期: 2022-11-14
- 网络出版日期: 2022-12-10
- 刊出日期: 2024-09-02

RELIABILITY EVALUATION OF THERMAL PROTECTION STRUCTURE UPON ADAPTIVE RBF NEURAL NETWORK

DONG Peng-hu^{1, 2,},
CHEN Qiang^{1, 2,},
LI Yan-bin^{1, 2, ,},
ZHANG Xu-dong^3,,
MA Han^{1, 2,},
FEI Qing-guo^{1, 2,}

1.
School of Mechanical Engineering, Southeast University, Nanjing 211189, China
2.
Jiangsu Aerospace Machinery Equipment Engineering Research Center, Nanjing 211189, China
3.
Beijing Aerospace Technology Institute, Beijing 100074, China

摘要

摘要:
针对复杂载荷下热防护结构可靠性评估效率低、分析精度差等问题，该文提出一种基于自适应径向基神经网络的可靠性评估方法。通过引入非线性收敛因子，对传统灰狼优化算法进行改进；采用改进后的灰狼算法优化径向基神经网络的中心点个数和扩展常数，建立精确预示热防护结构应力响应的自适应径向基网络模型；开展热防护结构的仿真和试验研究。结果表明：通过引入非线性收敛因子，大幅提高了灰狼算法的优化性能；该文提出的自适应径向基网络可以在小样本条件下建立高精确的代理模型；基于该文方法获得的可靠性分析结果与蒙特卡罗仿真结果、试验结果具有较好的一致性。
- 热防护结构 /
- 灰狼优化算法 /
- 径向基神经网络 /
- 复杂环境 /
- 可靠性
Abstract:
An evaluation method based on adaptive radial basis function (RBF) neural network is proposed to solve the problems of low efficiency and poor analysis accuracy in the reliability assessment of thermal protection structures (TPS) under complex loads. The traditional grey wolf algorithm is improved by introducing a nonlinear convergence factor. The improved grey wolf algorithm is applied to optimize the number of central points and expansion constant of the radial basis function to establish an adaptive radial basis function neural network, which can accurately predict the stresses. The reliability of the thermal protection structure is numerically and experimentally investigated. It is concluded that the optimization performance of the grey wolf algorithm is significantly improved by introducing a nonlinear convergence factor. The proposed adaptive RBF model could quickly realize the data prediction through small samples while ensuring the accuracy. The reliability obtained by the method proposed matches well with that of Monte Carlo simulation and of experimental results.
- thermal protection structure /
- grey wolf optimization algorithm /
- radial basis function neural network /
- complex environment /
- reliability

HTML全文

图 1 收敛算法结果变化图

Figure 1. Results of convergence algorithm

下载: 全尺寸图片幻灯片

图 2 RBF神经网络的结构图

Figure 2. Structure of RBF neural network

下载: 全尺寸图片幻灯片

图 3 测试函数及收敛曲线

Figure 3. Test function and convergence curve

下载: 全尺寸图片幻灯片

图 4 不同模型的预测结果

Figure 4. Prediction results of different models

下载: 全尺寸图片幻灯片

图 5 RBF神经网络预测结果的相对误差

Figure 5. Relative error of prediction results of RBF neural network

下载: 全尺寸图片幻灯片

图 6 热防护结构试验件

Figure 6. Specimen of thermal protection structure

下载: 全尺寸图片幻灯片

图 7 热防护结构分析模型

Figure 7. Analysis model for thermal protection structure

下载: 全尺寸图片幻灯片

图 8 上面板静应力和动应力分布

Figure 8. Static and dynamic stress distribution in the upper panel

下载: 全尺寸图片幻灯片

图 9 胶层静应力和动应力分布

Figure 9. Static and dynamic stress distribution in the adhesive layer

下载: 全尺寸图片幻灯片

图 10 误差均方根的变化曲线

Figure 10. Variation curve of RMS error

下载: 全尺寸图片幻灯片

图 11 不同模型的预测误差图

Figure 11. Prediction errors of different models

下载: 全尺寸图片幻灯片

图 12 可靠性评估流程

Figure 12. Process of reliability assessment

下载: 全尺寸图片幻灯片

图 13 胶层失效判据的统计结果

Figure 13. Statistical results of the failure criterion

下载: 全尺寸图片幻灯片

图 14 蒙特卡罗模拟的可靠度收敛曲线

Figure 14. Convergence of reliability using Monte Carlo simulation

下载: 全尺寸图片幻灯片

图 15 多场试验装置的示意图

Figure 15. Schematic diagram of multi-field test equipment

下载: 全尺寸图片幻灯片

图 16 金属板的温度变化曲线

Figure 16. Temperature curve of metal plate

下载: 全尺寸图片幻灯片

图 17 热防护结构的最终状态

Figure 17. End state of thermal protection structures

下载: 全尺寸图片幻灯片

表 1 优化算法测试函数

Table 1 Test function for optimization algorithm

函数	维度	x_i范围	目标最值
${F_1}( x ) = \displaystyle\sum\limits_{i = 1}^n {x_i^2}$	30	[−100, 100]	0
${F_2}( x ) = [ {x_i^2 - 10\cos ( {2\pi {x_i}} ) + 10} ]$	30	[−5.12, 5.12]	0

下载: 导出CSV

表 2 不确定性参数设置

Table 2 Value of uncertain parameters

参数	取值范围	分布特性
静力载荷/N	[760.00, 840.00]	均匀分布
金属板温度/(℃)	[108.00, 132.00]
随机振动g_rms/g	[10.26, 11.34]
面板弹性模量/GPa	[4.75, 5.25]
芯层弹性模量/GPa	[0.09, 0.10]
胶层弹性模量/GPa	[0.49, 0.54]
金属板弹性模量/GPa	[114.00, 126.00]

下载: 导出CSV

表 3 面板与胶层的力学参数

Table 3 Mechanical parameters of panel and adhesive layer

方向	上面板强度/MPa	胶层强度/MPa
X方向压缩	15.4	−
Y方向压缩	10.3	−
Z方向压缩	10.3	−
X方向拉伸	20.1	4.7
Y方向拉伸	12.7	−
Z方向拉伸	12.7	−
XY方向	11.8	2.2
YZ方向	8.3	1.5

下载: 导出CSV

表 4 热防护结构的可靠性分析结果

Table 4 Reliability of thermal protection structures

分析对象	失效/(%)	可靠/(%)
上面板	0.057	99.943
胶层	0.105	99.895

下载: 导出CSV

表 5 试验载荷参数的设置

Table 5 Value of the load

参数	取值范围	分布特性
静力载荷/N	[300.00, 350.00]	均匀分布
金属板底部温度/(℃)	[320.00, 380.00]
随机振动g_rms/g	[10.26, 11.34]

下载: 导出CSV

参考文献(27)

[1]	邹学锋, 郭定文, 潘凯, 等. 综合载荷环境下高超声速飞行器结构多场联合强度试验技术[J]. 航空学报, 2018, 39(12): 240 − 250. ZOU Xuefeng, GUO Dingwen, PAN Kai, et al. Test technique for multi-load combined strength of hypersonic vehicle structure under complex loading environment [J]. Acta Aeronautica et Astronautica Sinica, 2018, 39(12): 240 − 250. (in Chinese)
[2]	ZHU J J, WANG X J, ZHANG H G, et al. Six sigma robust design optimization for thermal protection system of hypersonic vehicles based on successive response surface method [J]. Chinese Journal of Aeronautics, 2019, 32(9): 2095 − 2108. doi: 10.1016/j.cja.2019.04.009
[3]	陈硕, 赵玉良, 董胜. 导管架型靠船墩的系统可靠性分析[J]. 工程力学, 2020, 37(增刊 1): 340 − 346. doi: 10.6052/j.issn.1000-4750.2019.04.S040 CHEN Shuo, ZHAO Yuliang, DONG Sheng. System reliability analysis of jacket structured dolphin [J]. Engineering Mechanics, 2020, 37(Suppl 1): 340 − 346. (in Chinese) doi: 10.6052/j.issn.1000-4750.2019.04.S040
[4]	孙大奇, 朱颖, 刘晓光, 等. 基于非概率集合理论的屋面板风致疲劳寿命估计[J]. 工程力学, 2021, 38(4): 211 − 220. doi: 10.6052/j.issn.1000-4750.2020.07.0439 SUN Daqi, ZHU Ying, LIU Xiaoguang, et al. Fatigue life analysis of panels under wind load based on non-probabilistic set theory [J]. Engineering Mechanics, 2021, 38(4): 211 − 220. (in Chinese) doi: 10.6052/j.issn.1000-4750.2020.07.0439
[5]	WANG G Z, XIN H H, WU D, et al. Data-driven arbitrary polynomial chaos-based probabilistic load flow considering correlated uncertainties [J]. IEEE Transactions on Power Systems, 2019, 34(4): 3274 − 3276. doi: 10.1109/TPWRS.2019.2908089
[6]	彭祖军. 复合材料热结构模型的不确定性分析和实验修正方法[D]. 哈尔滨: 哈尔滨工业大学, 2018. PENG Zujun. Uncertainty analysis and experimental calibration methods of thermal structure model of composite materials [D]. Harbin: Harbin Institute of Technology, 2018. (in Chinese)
[7]	NAKAMURA T, FUJII K. Probabilistic transient thermal analysis of an atmospheric reentry vehicle structure [J]. Aerospace Science and Technology, 2006, 10(4): 346 − 354. doi: 10.1016/j.ast.2006.02.002
[8]	WRIGHT M J, BOSE D, CHEN Y K. Probabilistic modeling of aerothermal and thermal protection material response uncertainties [J]. AIAA Journal, 2007, 45(2): 399 − 410. doi: 10.2514/1.26018
[9]	邓诗圆, 辛健强, 张琨, 等. 基于重要抽样法的热防护系统热可靠性评估方法[J]. 重庆大学学报, 2019, 42(2): 62 − 70. DENG Shiyuan, XIN Jianqiang, ZHANG Kun, et al. Assessment of thermal reliability of thermal protection systems by importance sampling method [J]. Journal of Chongqing University, 2019, 42(2): 62 − 70. (in Chinese)
[10]	季熠, 李彦斌, 杭晓晨, 等. 基于响应面代理模型的雷达天线阵面风载分析[J]. 工程力学, 2019, 36(11): 222 − 229. doi: 10.6052/j.issn.1000-4750.2018.12.0680 JI Yi, LI Yanbin, HANG Xiaochen, et al. Wind load analysis of radar antenna based on response surface model [J]. Engineering Mechanics, 2019, 36(11): 222 − 229. (in Chinese) doi: 10.6052/j.issn.1000-4750.2018.12.0680
[11]	李正良, 王成, 王涛, 等. 基于主动学习Kriging模型的直立锁缝屋面系统抗风揭可靠度分析[J]. 工程力学, 2022, 39(10): 111 − 119. doi: 10.6052/j.issn.1000-4750.2021.06.0430 LI Zhengliang, WANG Cheng, WANG Tao, et al. Reliability analysis of wind-resistance of standing seam roof system based on active learning Kriging model [J]. Engineering Mechanics, 2022, 39(10): 111 − 119. (in Chinese) doi: 10.6052/j.issn.1000-4750.2021.06.0430
[12]	秦国华, 郭翊翔, 王华敏, 等. 飞机整体结构件的“加工变形-疲劳寿命”多目标结构优化方法[J]. 工程力学, 2021, 38(8): 222 − 236. doi: 10.6052/j.issn.1000-4750.2020.07.0517 QIN Guohua, GUO Yixiang, WANG Huamin, et al. Multiple objective structural optimization on “machining deformation-fatigue life” of aeronautical monolithic components [J]. Engineering Mechanics, 2021, 38(8): 222 − 236. (in Chinese) doi: 10.6052/j.issn.1000-4750.2020.07.0517
[13]	SU Y P, XU L H, GOODMAN E D. Hybrid surrogate-based constrained optimization with a new constraint-handling method [J]. IEEE Transactions on Cybernetics, 2022, 52(6): 5394 − 5407. doi: 10.1109/TCYB.2020.3031620
[14]	EL MRABTI I, EL HAKIMI A, TOUACHE A, et al. A comparative study of surrogate models for predicting process failures during the sheet metal forming process of advanced high-strength steel [J]. The International Journal of Advanced Manufacturing Technology, 2022, 121(1/2): 199 − 214.
[15]	李游, 李传习, 陈卓异, 等. 基于监测数据的钢箱梁U肋细节疲劳可靠性分析[J]. 工程力学, 2020, 37(2): 111 − 123. doi: 10.6052/j.issn.1000-4750.2019.03.0114 LI You, LI Chuanxi, CHEN Zhuoyi, et al. Fatigue reliability analysis of U-rib detail of steel box girder based on monitoring data [J]. Engineering Mechanics, 2020, 37(2): 111 − 123. (in Chinese) doi: 10.6052/j.issn.1000-4750.2019.03.0114
[16]	梁泽, 王玥瑶, 岳远紊, 等. 耦合遗传算法与RBF神经网络的PM_2.5浓度预测模型[J]. 中国环境科学, 2020, 40(2): 523 − 529. doi: 10.3969/j.issn.1000-6923.2020.02.007 LIANG Ze, WANG Yueyao, YUE Yuanwen, et al. A coupling model of genetic algorithm and RBF neural network for the prediction of PM_2.5 concentration [J]. China Environmental Science, 2020, 40(2): 523 − 529. (in Chinese) doi: 10.3969/j.issn.1000-6923.2020.02.007
[17]	董现, 王湛. 基于改进混沌粒子群的混合神经网络和蒙特卡洛法的结构随机灵敏度分析方法[J]. 工程力学, 2015, 32(12): 49 − 57. doi: 10.6052/j.issn.1000-4750.2014.05.0381 DONG Xian, WANG Zhan. Stochastic sensitivity analysis method based on hybrid neural network of improved chaotic particle swarm and Monte Carlo [J]. Engineering Mechanics, 2015, 32(12): 49 − 57. (in Chinese) doi: 10.6052/j.issn.1000-4750.2014.05.0381
[18]	BENCHERIF N, BENHALILIBA M. A Hashin criteria investigation to predict the interaction effect of defaults on the damage of composite pipe [J]. Oriental Journal of Chemistry, 2021, 37(2): 314 − 320. doi: 10.13005/ojc/370208
[19]	NEGI G, KUMAR A, PANT S, et al. GWO: A review and applications [J]. International Journal of System Assurance Engineering and Management, 2021, 12(1): 1 − 8.
[20]	CHIU C Y, SHIH P C, LI X C. A dynamic adjusting novel global harmony search for continuous optimization problems [J]. Symmetry, 2018, 10(8): 10080337. doi: 10.3390/sym10080337
[21]	郭振洲, 刘然, 拱长青, 等. 基于灰狼算法的改进研究[J]. 计算机应用研究, 2017, 34(12): 3603 − 3606, 3610. doi: 10.3969/j.issn.1001-3695.2017.12.019 GUO Zhenzhou, LIU Ran, GONG Changqing, et al. Study on improvement of gray wolf algorithm [J]. Application Research of Computers, 2017, 34(12): 3603 − 3606, 3610. (in Chinese) doi: 10.3969/j.issn.1001-3695.2017.12.019
[22]	MAYER K S, MÜLLER C, SOARES J A, et al. Deep phase-transmittance RBF neural network for beamforming with multiple users [J]. IEEE Wireless Communications Letters, 2022, 11(7): 1498 − 1502. doi: 10.1109/LWC.2022.3177162
[23]	CAPÓ M, PÉREZ A, LOZANO J A. An efficient split-merge re-start for the K-means algorithm [J]. IEEE Transactions on Knowledge and Data Engineering, 2020, 34(4): 1618 − 1627.
[24]	彭玉. 力/热环境下缝合式复合材料夹芯板失效仿真研究[D]. 南京: 东南大学, 2019. PENG Yu. Failure simulation of stitched composite sandwich panel under mechanical/thermal environment [D]. Nanjing: Southeast University, 2019. (in Chinese)
[25]	HUANG L, YAO W X. Experimental investigations on failure of silicon adhesive in thermal protection system [J]. Journal of Spacecraft and Rockets, 2019, 56(2): 615 − 618. doi: 10.2514/1.A34363
[26]	王永亮, 柳占立, 林三春, 等. 基于连续损伤的岩石渗流有限元分析[J]. 工程力学, 2016, 33(11): 29 − 37. doi: 10.6052/j.issn.1000-4750.2015.03.0229 WANG Yongliang, LIU Zhanli, LIN Sanchun, et al. Finite element analysis of seepage in rock based on continuum damage evolution [J]. Engineering Mechanics, 2016, 33(11): 29 − 37. (in Chinese) doi: 10.6052/j.issn.1000-4750.2015.03.0229
[27]	O'CONNOR P D T, KLEYNER A. Practical reliability engineering [M]. 5th ed. Hoboken: John Wiley & Sons, 2012: 125 − 126.

施引文献(7)

期刊类型引用(3)

1.	张小明，陈滔，周红梅，禹建功. 压电半导体板超声导波特性研究. 河南理工大学学报(自然科学版). 2023(01): 95-101 . 百度学术
2.	王鑫鑫，禹建功，张博，刘灿灿，王现辉. 一维六方准晶纳米板中Lamb波特性研究. 工程力学. 2023(10): 213-221 . 本站查看
3.	李思宇，张宇，刘宏业，范彦平，吕炎，刘增华. 黏弹性正交各向异性空心圆柱中纵向导波的传播. 复合材料学报. 2019(10): 2275-2285 . 百度学术