应变电桥自适应赋权的飞机结构应变载荷模型-中国科协年会航空智能技术专栏

施英杰; 朱玉雯; 何发东; 刘斌超; 鲍蕊

doi:10.7527/S1000-6893.2026.33618

航空学报 >

0 1 - 0

DOI: https://doi.org/10.7527/S1000-6893.2026.33618

应变电桥自适应赋权的飞机结构应变载荷模型-中国科协年会航空智能技术专栏

施英杰 ,
朱玉雯 ,
何发东 ,
刘斌超 ,
鲍蕊

展开

1. 北京航空航天大学
2. 中国飞行试验研究院
3. 北京航空航天大学航空科学与工程学院

收稿日期: 2026-03-27

修回日期: 2026-07-30

网络出版日期: 2026-08-10

基金资助

强度与结构完整性全国重点实验室;杭州市北航国际创新研究院科研启动经费项目

收起

Strain-Based Loads Model for Aircraft Structures Using Self-Adaptive Weighting of Strain Bridges

SHI Ying-Jie ,
ZHU Yu-Wen ,
HE Fa-Dong ,
LIU Bin-Chao ,
BAO Rui

Expand

Received date: 2026-03-27

Revised date: 2026-07-30

Online published: 2026-08-10

Fold

摘要

建立准确可靠、高鲁棒性的飞机结构应变载荷模型，对于飞行载荷实测、飞机健康监控等航空结构领域先进发展方向具有重要意义。现有模型未考虑各应变电桥对预测任务的重要程度、以及此重要程度随载荷工况的变化，不仅导致模型的预测精度受限，还导致模型容易受到无效信息的干扰。对此，提出将多步注意力机制嵌入飞机结构应变载荷神经网络模型，实现对应变电桥重要程度的自适应识别与赋权，并采用双梁式机翼结构地面载荷校准试验数据进行验证。结果表明：相比于目前常用的多元线性回归方程组与全连接神经网络模型，本模型可将弯矩、剪力、扭矩预测均方根误差降低30%以上，提高了预测精度；实现了不同工况下应变电桥权重的自主分配、应变电桥数据有效性的自行识别、失效应变电桥数据的自动排除，对复杂场景下的飞机结构应变载荷模型构建具有稳定性与鲁棒性。

关键词： 应变载荷模型; 飞行载荷实测; 飞机结构; 多步注意力机制; 神经网络模型

本文引用格式

施英杰 , 朱玉雯 , 何发东 , 刘斌超 , 鲍蕊 . 应变电桥自适应赋权的飞机结构应变载荷模型-中国科协年会航空智能技术专栏[J]. 航空学报, 0 : 1 -0 . DOI: 10.7527/S1000-6893.2026.33618

Abstract

Establishing an accurate, reliable and robust strain-load model is of great significance to the advanced development of aircraft structures, such as in-flight load measurement and aircraft health monitoring. However, existing models neither weight the importance of each strain bridge onto the load prediction nor consider the varying importance among various load conditions, which leads to limited accuracy and robustness. To address this problem, this paper proposes to embed multi-step attention mechanism into neural network-based strain-load model. The multi-step attention mechanism realizes adaptive identification and weight assignment for each strain bridge, which is validated by the data of a dual-beam wing structure in a ground load calibration test. Compared with the multiple linear regression model and fully connected neural network model, the proposed model succeeds in reducing the root mean square error of bending moment, shear force and torque prediction by more than 30%, demonstrating a great improvement for the prediction accuracy. Moreover, it also enables autonomous weight allocation and validation judgement for all strain bridges under diverse load conditions, ensuring the robustness of strain-load models in complex scenarios.

Key words： Strain-based loads model; Flight load measurement; Aircraft structure; Multi-step attention mechanism; Neural network model?

参考文献

[1] 闫中午, 宗宁, 任文广, 等. 飞行载荷发展综述[J]. 航空工程进展, 2020, 11(6): 873-886.
YAN Z W, ZONG N, REN W G, et al. Overview of flight loads development[J]. Advances in Aeronautical Science and Engineering, 2020, 11(6): 873-886. (in Chinese).
[2] 谢克诚, 周轩, 董雷霆, 等. 基于长短期记忆网络与小波变换的直升机飞参-载荷预测[J]. 航空科学技术, 2024, 35(11): 51-57.
XIE K C, ZHOU X, DONG L T, et al. Flight-parameter-based load prediction of helicopter using LSTM network and wavelet transform[J]. Aeronautical Science & Technology, 2024, 35(11): 51-57(in Chi-nese).
[3] 张海涛, 余建虎, 李志蕊, 等. T型尾翼布局的垂尾载荷测量技术[J]. 航空学报, 2019, 40(3): 122074.
ZHANG H T, YU J H, LI Z R, et al. Measuring tech-nology for vertical fin load of T-shaped empennage layout[J]. Acta Aeronautica et Astronautica Sinica, 2019, 40(3): 122074(in Chinese).
[4] 曹景涛. 飞机全动式鸭翼载荷飞行测量技术[J]. 航空学报, 2015, 36(4): 1135-1141.
CAO J T. Aircraft all movable canard load flight meas-urement technology[J]. Acta Aeronautica et Astro-nautica Sinica, 2015, 36(4): 1135-1141(in Chinese).
[5] 兑红娜, 刘栋梁, 张志贤, 等. 基于应变测量的结构载荷分布反演方法[J]. 航空学报, 2021, 42(5): 524337.
DUI H N, LIU D L, ZHANG Z X, et al. Distributed load recovery approach based on strain measure-ments[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(5): 524337(in Chinese).
[6] 李俊, 范华飞, 何发东, 等. 飞机折叠翼结构载荷飞行测量模型构建及应用[J]. 航空学报, 2024, 45(23): 230370.
LI J, FAN H F, HE F D, et al. Construction of flight measurement model for folding wing structural load of aircraft[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(23): 230370(in Chinese).
[7] 沈雯阳, 江文松, 罗哉, 等. 基于高阶加权最小二乘的机翼多源载荷校准方法[J]. 计量学报, 2026, 47(1): 16-25.
SHEN W Y, JIANG W S, LUO Z, et al. Wing multi-source load calibration method based on high-order weighted least square[J]. Acta Metrologica Sinica. 2026, 47(1): 16-25(in Chinese).
[8] 侯乔乔, 张清勇. 一种改进的载荷标定试验数据回归算法[J]. 工程与试验, 2018, 58(04): 22-24.
HOU Q Q, ZHANG Q Y. An Improved Regression Algorithm for Load Calibration Test Data[J]. Engineer-ing and Test, 2018, 58(04): 22-24(in Chinese).
[9] 张赐宝, 滕申科, 唐吉运. 基于BP神经网络的三角机翼飞行载荷模型研究[J]. 工程与试验, 2014, 54(3): 5-8.
ZHANG C B, TENG S K, TANG J Y. Study on flight load model for delta wing based on BP neural net-work[J]. Engineering & Test, 2014, 54(3): 5-8(in Chi-nese).
[10] 尚琳, 何发东, 杨华保, 等. 基于神经网络的垂尾飞行载荷模型研究[J]. 强度与环境, 2011, 38(5): 49-53.
SHANG L, HE F D, YANG H B, et al. Research on flight loads model of vertical tail based on neural net-works[J]. Structure & Environment Engineering, 2011, 38(5): 49-53(in Chinese).
[11] 刘佳玮, 于明鑫, 祝连庆, 等. 基于深度学习的机翼蒙皮载荷计算方法[J]. 电子测量与仪器学报, 2022, 36(4): 1-8.
LIU J W, YU M X, ZHU L Q, et al. Calculation meth-od of wing skin load based on deep learning[J]. Journal of Electronic Measurement and Instrumentation, 2022, 36(4): 1-8(in Chinese).
[12] 金鑫, 殷建业, 王健志, 等. 基于深度学习的飞行载荷测试与反演方法研究[J]. 航空工程进展, 2020, 11(06): 887-893.
JIN X, YIN J Y, WANG J Z, et al. Research on deep-learning-based flight load test and estimation meth-od[J]. Advances in Aeronautical Science and Engineer-ing, 2020, 11(06): 887-893(in Chinese).
[13] COOPER S, DIMAIO D H, ZHAO H. Static load esti-mation using artificial neural network: Application on a wing rib[J]. Advances in Engineering Software, 2018, 125: 113-125.
[14] 蒋祖国. 飞机结构载荷/环境谱[M]. 北京: 电子工业出版社, 2012: 254-260.
JIANG Z G. Aircraft structural load/environment spec-trum[M]. Beijing: Electronic Industry Press, 2012: 260-267(in Chinese).
[15] 闫楚良. 飞机载荷谱实测技术与编制原理[M]. 北京: 航空工业出版社, 2011: 58-66.
YAN C L. Aircraft load spectrum measurement tech-nique and compilation principle[M]. Beijing: Aviation Industry Press, 2011: 96-102(in Chinese).
[16] YANG F, YUAN P J, SHEN L, et al. Visual awareness sharpens and accelerates attentional sampling through enhancing inhibitory neural modulation in the atten-tion network[J]. Nat Commun, 2025, 16: 10058
[17] LI J N, WONG Y, ZHAO Q, et al. Attention transfer from web images for video recognition[C]//Proceedings of the 25th ACM international conference on multi-media. New York: ACM, 2017: 1-9.
[18] CHEN C F R, FAN Q F, PANDA R. CrossViT: cross-attention multi-scale vision transformer for image classification[C]//2021 IEEE/CVF International Con-ference on Computer Vision. Piscataway: IEEE Press, 2021: 347-356.
[19] SUN Y T, LI Z Y, ZHANG Y K, et al. Efficient atten-tion mechanisms for large language models: A sur-vey[DB/OL]. arXiv preprint: 2507.19595, 2025.
[20] ZHOU H P, DENG B, SUN K L, et al. UTE CrackNet： Transformer-guided and edge feature ex-traction U-shaped road crack image segmentation[J]. The Visual Computer, 2025, 41(4): 2271-2283.
[21] 曲桂娴, 刘冬阳, 杨旭, 等. 基于传感器时序信息增强的剩余寿命预测方法[J]. 航空学报, 2025, 46(17): 231634.
QU G X, LIU D Y, YANG X, et al. Remaining useful life prediction method based on temporal information enhancement of sensors[J]. Acta Aeronautica et Astro-nautica Sinica, 2025, 46(17): 231634(in Chinese).
[22] BUTEREZ D, JANET J P, OGLIC D, et al. An end-to-end attention-based approach for learning on graphs[J]. Nat Commun, 2025, 16: 5244
[23] ARIK S O, PFISTER T. TabNet: attentive interpretable tabular learning[C]// Proceedings of the 35th AAAI Conference on Artificial Intelligence, Menlo Park: AAAI Press, 2021, 35(8): 6679-6687
[24] 吴文辉, 何家峰, 蔡高琰, 等. 基于TabNet-LSTNet的多特征短期负荷预测[J]. 重庆大学学报, 2024, 9: 129-140.
WU W H, HE J F, CAI G Y, et al. Multi-featured short-term load forecasting based on TabNet-LSTNet[J]. Journal of Chongqing University, 2024, 9: 129-140(in Chinese).
[25] SOMVANSHI S, DAS S, JAVED S A, et al. A survey on deep tabular learning[DB/OL]. arXiv preprint: 2410.12034, 2024.
[26] SUDARSAHN V, SEIDER W D. Advancing machine learning in industry 4.0: benchmark framework for ra-re-event prediction in chemical processes[DB/OL]. arXiv preprint: 2409.00485, 2024.

Options

摘要页面

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献