基于LSTM和响应分解的冲击载荷识别方法研究

摘要
图/表
参考文献(20)
相关文章 (3)

全文: PDF (3777 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要同一量级的冲击载荷所产生的动响应要远大于静态响应，因此准确识别冲击载荷对于航空器结构件的动强度设计、校核与结构健康监测都具有重要意义。本文提出的方法主要针对一般线性结构的冲击载荷识别问题，从实测冲击响应应变信号出发，主要解决了冲击载荷与响应信号样本长度不一致这一突出矛盾。首先基于冲击响应信号分解方法来进行振动信号特征提取，然后基于长短期记忆神经网络（LSTM）对载荷和响应信号样本特征进行映射，从而实现冲击载荷识别。通过对挂架模型实测冲击载荷信号进行识别，结果表明四种工况下，该方法识别的冲击载荷的均方根相对误差小于0.6，相关系数大于0.94。结果初步表明，在理想的试验环境中，该方法具备一定的识别精度。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	黄大伟
	陈立昆
	高亚东

关键词 ：动力学逆问题, 冲击载荷识别, 响应分解, 振动信号特征提取, 长短期记忆神经网络（LSTM）

Abstract：The strain produced by impact load on a structural is much greater than the static load. It is of great significance to identify the impact load accurately. The method proposed aims at the prominent contradiction of inconsistent sample length between impact load and response signal. The method was based on linear model and the experiment was carried on a steel pylon structure. Firstly, the feature extraction of vibration signals was carried out based on the impact response signal decomposition theory. Then, the load and response signal sample characteristics are mapped to realize the impact load identification based on the long short term memory neural network (LSTM). The results show that the correlation coefficients between the true and predicted load are more than 94% and the RMSE (root-mean square error) are less than 0.6.

Key words： impact load identification response decomposition extracting signal characteristic long short term memory neural network(LSTM)

收稿日期: 2023-02-15 出版日期: 2023-02-15

引用本文:

黄大伟，陈立昆，高亚东. 基于LSTM和响应分解的冲击载荷识别方法研究[J]. 振动与冲击, 2024, 43(3): 69-76.
HUANG Dawei, CHEN Likun, GAO Yadong. Impact load identification method based on LSTM and response decomposition. JOURNAL OF VIBRATION AND SHOCK, 2024, 43(3): 69-76.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2024/V43/I3/69

[1] 何发东,舒成辉. 贝叶斯正则化BP网络在机翼载荷分析中的应用[J]. 飞行力学, 2009, 27(4): 85-88. HE Fa-dong1,SHU Cheng-hui2(1.Aircraft Flight Test Technology Institute,Chinese Flight Test Establishment,Xi'an 710089,China;2.Agency for Chief Engineer,Chinese Flight Test Establishment,Xi'an 710089,China). Application of BP Neural Networks Based on Bayesian Regularization to Aircraft Wing Loads Analysis[J]. Flight Dynamics, 2009, 27(4): 85-88. [2] 杨智春,贾有. 动载荷的识别方法[J]. 力学进展, 2015, 45(00): 29-54. YANG Zhichun;JIA You;Institute of Structural Dynamics and Control,School of Aeronautics,Northwestern Polytechnical University;. The identification of dynamic loads[J]. Advances in Mechanics, 2015, 45(00): 29-54. [3] 张方,朱德懋. 基于神经网络模型的动载荷识别[J]. 振动工程学报, 1997(2): 40-46. Zhang Fang Zhu Demao (Vibration Engineering Research Institute, Nanjing University of Aeronautics and Astronautics Nanjing,210016). The Dynamic Load Identification Research Based on Neural Network Model[J]. JOURNAL OF VIBRATION ENGINEERING, 1997(2): 40-46. [4] X. Cao,Y. Sugiyama,Y. Mitsui. Application of artificial neural networks to load identification[J]. Computers & Structures, 1998, 69(1): 63-78. [5] Edi Sofyan,Pavel Trivailo. SOLVING AERODYNAMIC LOAD INVERSE PROBLEMS USING A HYBRID FEM -ARTIFICIAL INTELLIGENCE[J]. [6] 窦春红,林近山,寇兴磊. 基于BP神经网络的海洋平台振动载荷识别[J]. 石油矿场机械, 2007(7): 11-15. DOU Chun-hong1,LIN Jin-shan2, KOU Xing-lei3(1. School of Information and Control Engineering, Weifang College, Weifang 261061, China;2. School of Mechanical and Electronic Engineering, Weifang College, Weifang 261061, China;3. Shandong Shouguang Juneng Electric Power Co., Ltd., Shouguang 262700, China ). Vibration Load Identification of Offshore Platform Based on BP Neural Network[J]. Oil Field Equipment, 2007(7): 11-15. [7] 沙瑞华. 基于神经网络的水电机组动载识别研究[D]. 大连理工大学, 2005. Ruihua Sha. Dynamic Load Identification for Turbine Generator Set Based on Neural Network[D]. Chinese Doctoral Dissertations & Master's Theses Full-text Database (Master), 2005. [8] 董会丽. 基于RBF神经网络和遗传算法的复合材料层合板、壳载荷识别[D]. 南京航空航天大学, 2007. Huili Dong. Load Identification for a Composite Laminated Shell Using Radial Base Function Neural Network and Genetic Algorithm[D]. Chinese Master's Theses Full-text Database, 2007. [9] ALLEN M J, DIBLEY R P.Modeling Aircraft Wing Loads From Flight Data Using Neural Networks[J]. SAE Transactions, 2003, Vol.112: 512-520. [10] MARANO G C, QUARANTA G, MONTI G. Modified Genetic Algorithm for the Dynamic Identification of Structural Systems Using Incomplete Measurements[J]. Computer‐Aided Civil and Infrastructure Engineering, 2011(2). [11] Samson B. Cooper,Dario DiMaio. Static load estimation using artificial neural network: Application on a wing rib[J]. Advances in Engineering Software, 2018, 125: 113-125. [12] 杨特,杨智春,梁舒雅,等. 平稳随机载荷的信号时频特征提取与深度神经网络识别[J]. 航空学报: 1-12. YANG Te;YANG Zhichun;Liang Shuya;KANG Zaifei;JIA You;. Feature extraction and identification of stationary random dynamic load using deep neural network[J]. Acta Aeronautica et Astronautica Sinica: 1-12. [13] 夏鹏,杨特,徐江,等. 利用时延神经网络的动载荷倒序识别[J]. 航空学报, 2021, 42(7): 1-9. XIA Peng;YANG Te;XU Jiang;WANG Le;YANG Zhichun;. A reversed time sequence dynamic load identification method using time delay neural network[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(7): 1-9. [14] 夏鹏,杨特,王乐,等. 平稳随机载荷的灵敏度分析识别方法[J]. 振动与冲击, 2022, 41(23): 300-306. XIA Peng;YANG Te;WANG Le;YANG Zhichun; Identification of stationary random dynamic load based on sensitivity analysis[J]. Journal of Vibration and Shock, 2022, 41(23): 300-306. [15] Liu, J., Meng, X., Jiang, C., Han, X., and Zhang, D. (2016) Time-domain Galerkin method for dynamic load identification. International Journal for Numerical Methods in Engineering, 2016,105(8): 620– 640. [16] LIU J, LI K. Sparse identification of time-space coupled distributed dynamic load, Mechanical Systems and Signal Processing.2021,148:107177. [17] 陈树海,郭安丰,吴邵庆,等. 基于BP神经网络的星箭界面动载荷识别[J]. 振动与冲击, 2023, 42(5): 279-286, 304. CHEN Shuhai;GUO Anfeng;WU Shaoqing;FEI Qingguo. Dynamic load identification of satellite-rocket interface based on BP neural network[J]. Journal of Vibration and Shock, 2023, 42(5): 279-286, 304. DOI:10.13465/j.cnki.jvs.2023.05.034 [18] 张志宏,张宏,陈有,等. 基于遗传神经网络的履带行驶系统载荷识别方法[J]. 振动与冲击, 2022, 41(3): 54-61, 89. ZHANG Zhihong;ZHANG Hong;CHEN You;LI Zhi;LI Guohua;FU Zheng. Load identification method of track driving system based on genetic neural network[J]. Journal of Vibration and Shock, 2022, 41(3): 54-61, 89. DOI:10.13465/j.cnki.jvs.2022.03.007 [19] 曾俊玮,季元进,任利惠,等. 卡尔曼滤波器与神经网络串行的轮胎载荷识别模型[J]. 振动与冲击, 2023, 42(11). ZENG Junwei; JI Yuanjin; REN Lihui. A serial tire load identification model based on Kalman filter and neural network [J]. Journal of Vibration and Shock, 2023, 42(11): 262-270, 294.DOI:10.13465/j.cnki.jvs.2023.11.031 [20] 蔡元奇. 时域内动态载荷识别理论及实施技术研究[D]. 武汉大学, 2004. Yuanqi Cai. Theory and Applying Technique Study on Identification Method for Dynamic Loads in Time Domain[D]. Chinese Doctoral Dissertations & Master's Theses Full-text Database (Doctor), 2004.