结合离线知识的时变结构模态参数在线辨识

岳振江; 刘莉; 余磊; 康杰

doi:10.7527/S1000-6893.2019.22931

航空学报 >

2019 , Vol. 40 >Issue 8: 222931 - 222931

DOI: https://doi.org/10.7527/S1000-6893.2019.22931

固体力学与飞行器总体设计

结合离线知识的时变结构模态参数在线辨识

岳振江 ,
刘莉 ,
余磊 ,
康杰

展开

1. 北京理工大学宇航学院, 北京 100081;
2. 北京理工大学飞行器动力学与控制教育部重点实验室, 北京 100081

收稿日期: 2019-01-22

修回日期: 2019-02-25

网络出版日期: 2019-04-17

收起

Online identification of time-varying structural modal parameters combined with offline knowledge

YUE Zhenjiang ,
LIU Li ,
YU Lei ,
KANG Jie

Expand

1. School of Aerospace Engineering, Beijing Institute of Technology, Beijing 100081, China;
2. Key Laboratory of Dynamics and Control of Flight Vehicle, Ministry of Education, Beijing Institute of Technology, Beijing 100081, China

Received date: 2019-01-22

Revised date: 2019-02-25

Online published: 2019-04-17

Fold

摘要

飞行器的结构模态参数在线获取对其高效、可靠运行具有重要意义。传统时变结构模态参数辨识方法存在辨识虚假结果较多，抵抗测量数据中的极端异常值能力差等问题，难以有效应用于在线过程。建立一种基于长短时记忆网络的时变结构模态参数在线辨识网络模型，通过数据集构建过程离线地引入先验信息，同时结合模型自身特性，有效提升制约在线辨识应用的可靠性。实验结果表明：在不同时变规律下，与传统辨识方法相比，在线辨识模型能有效缓解虚假结果问题，同时保证辨识结果的连续性；采用α稳定分布模型对脉冲噪声进行建模，验证了其在测量数据包含由于偶发因素产生的极端异常值时在线辨识鲁棒性。

关键词： 时变结构; 模态辨识; 在线; 深度学习; 虚假模态; 脉冲噪声

本文引用格式

岳振江 , 刘莉 , 余磊 , 康杰 . 结合离线知识的时变结构模态参数在线辨识[J]. 航空学报, 2019 , 40(8) : 222931 -222931 . DOI: 10.7527/S1000-6893.2019.22931

Abstract

The online acquisition of modal parameters of aircraft structures is of great significance for efficient and reliable operation of the aircraft. The traditional modal parameter identification methods for time-varying structures have problems such as more false results and the poor ability to resist extreme outliers in measured data, becoming difficult to effectively apply to online processes. To solve these problems, an online identification model of time-varying structural modal parameters based on long short-term memory networks is established. For a given time-varying structures, prior information is introduced offline through the data set construction process, and the characteristics of the model are utilized to effectively improve the reliability of the online identification application. The experimental results show that compared with the traditional identification method, the proposed online identification model can effectively alleviate the problem of false results and ensure the continuity of identification results. The α stable distribution model is used to model the impulse noise, verifying the robustness of the online identification model that contains extreme outliers in measured data due to accidental factors.

Key words： time varying structure; modal identification; online; deep learning; spurious modes; impulse noise

参考文献

[1] 王昌瑞, 康仁科,鲍岩,等. 飞机蒙皮镜像铣加工稳定性分析[J]. 航空学报, 2018, 39(11):422109. WANG C R, KANG R K, BAO Y, et al. Stability analysis of aircraft skin mirror milling process[J]. Acta Aeronautica et Astronautica Sinica, 2018, 39(11):422109(in Chinese).
[2] 唐晓峰, 何振威,常洪振,等. 轴承支撑的舵面热模态试验及支撑刚度辨识[J]. 航空学报, 2019, 40(6):222617. TANG X F, HE Z W, CHANG H Z, et al. Thermo-modal test on an axle bearing supported rudder and identification of its supporting stiffness[J]. Acta Aeronautica et Astronautica Sinica, 2019, 40(6):222617(in Chinese).
[3] 梁力, 杨智春,欧阳炎,等. 垂尾抖振主动控制的压电作动器布局优化[J]. 航空学报, 2016, 37(10):3035-3043. LIANG L, YANG Z C, OUYANG Y, et al. Optimization of piezoelectric actuator configuration on a vertical tail for buffeting control[J]. Acta Aeronautica et Astronautica Sinica, 2016, 37(10):3035-3043(in Chinese).
[4] 陈果, 罗云,郑其辉,等. 复杂空间载流管道系统流固耦合动力学模型及其验证[J]. 航空学报, 2013, 34(3):597-609. CHEN G, LUO Y, ZHENG Q H, et al. Fluid-structure coupling dynamic model of complex spatial fluid-conveying pipe system and its verification[J]. Acta Aeronautica et Astronautica Sinica, 2013, 34(3):597-609(in Chinese).
[5] 余建新, 卫剑征,谭惠丰. 飞艇骨架结构动态损伤识别方法[J]. 航空学报, 2016, 37(11):3385-3394. YU J X, WEI J Z, TAN H F. Dynamic damage detection meth-ods for airship framework structure[J]. Acta Aeronautica et Astronautica Sinica, 2016, 37(11):3385-3394(in Chinese).
[6] GARIBALDI L, FASSOIS S. MSSP special issue on the identification of time varying structures and sys-tems[J]. Mechanical Systems and Signal Processing, 2014, 47(1-2):1-2.
[7] HOUTZAGER I, VAN WINGERDEN J W, VERHAEGEN M. Recursive predictor-based subspace identification with application to the real-time closed-loop tracking of flutter[J]. IEEE Transactions on Control Systems Technology, 2012, 20(4):934-949.
[8] 倪智宇, 刘金国,吴志刚. 一种改进的航天器时变模态参数递推辨识方法[J]. 宇航学报, 2018, 39(10):1097-1106. NI Z Y, LIU J G, WU Z G. An improved recursive identification method for time-varying modal parameters of spacecraft[J]. Journal of Astronautics, 2018, 39(10):1097-1106(in Chinese).
[9] BASSEVILLE M, BENVENISTE A, GOURSAT M, et al. In-flight vibration monitoring of aeronautical struc-tures-Subspace-based online automated identification versus detection[J]. IEEE Control Systems Magazine, 2007, 27(5):27-42.
[10] AVENDA O, VALENCIA L D, FASSOIS S D. Gener-alized stochastic constraint TARMA models for in-operation identification of wind turbine non-stationary dynamics[J]. Key Engineering Materials, 2013, 569-570:587-594.
[11] MA Z S, LIU L, ZHOU S D, et al. Parametric output-only identification of time-varying structures using a ker-nel recursive extended least squares TARMA approach[J]. Mechanical Systems and Signal Processing, 2018, 98(Supplement C):684-701.
[12] LIU G R, LAM K Y, HAN X. Determination of elastic constants of anisotropic laminated plates using elastic waves and a progressive neural network[J]. Journal of Sound and Vibration, 2002, 252(2):239-259.
[13] XU B, WU Z S, CHEN G D, et al. Direct identification of structural parameters from dynamic responses with neural networks[J]. Engineering Applications of Artificial Intelligence, 2004, 17(8):931-943.
[14] FACCHINI L, BETTI M, BIAGINI P. Neural network based modal identification of structural systems through output-only measurement[J]. Computers & Structures, 2014, 138:183-194.
[15] FEI M R, ZHANG J, HU H S, et al. A novel linear re-current neural network for multivariable system identifi-cation[J]. Transactions of the Institute of Measurement and Control, 2006, 28(3):229-242.
[16] TUTUNJI T A. Approximating transfer functions using neural network weights[C]//Proceedings of 4th International IEEE/EMBS Conference on Neural Engineering. Piscataway, NJ:IEEE Press, 2009:641-644.
[17] 许斌, 龚安苏,贺佳,等. 基于神经网络模型的结构参数提取新方法[J]. 工程力学, 2011, 28(4):35-41. XU B, GONG A S, HE J, et al. A novel structural pa-rameter extraction methodology with neural network based nonparametric model[J]. Engineering Mechanics, 2011, 28(4):35-41(in Chinese).
[18] 于开平, 庞世伟,赵婕. 时变线性/非线性结构参数识别及系统辨识方法研究进展[J]. 科学通报, 2009, 54(20):3147-3156. YU K P, PANG S W, ZHAO J. Advances in method of time-varying linear/nonlinear structural system identification and parameter estimate[J]. Chinese Sci Bull (Chinese Vertion), 2009, 54(20):3147-3156(in Chinese).
[19] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553):436-444.
[20] WANG X F, DONG X M, KONG X W, et al. Drogue detection for autonomous aerial refueling based on convolutional neural networks[J]. Chinese Journal of Aeronautics, 2017, 30(1):380-390.
[21] GAO L L, GUO Z, ZHANG H W, et al. Video caption-ing with attention-based LSTM and semantic consistency[J]. IEEE Transactions on Multimedia, 2017, 19(9):2045-2055.
[22] MCCANN B, BRADBURY J, XIONG C, et al. Learned in translation:Contextualized word vectors[C]//Proceedings of Neural Information Processing Systems 2017. San Diego, CA:NIPS,2017:1-12.
[23] JIA G M, CHENG F Y, YANG J F, et al. Intelligent checking model of Chinese radiotelephony read-backs in civil aviation air traffic control[J]. Chinese Journal of Aeronautics, 2018, 31(12):2280-2289.
[24] SPIRIDONAKOS M D, FASSOIS S D. Non-stationary random vibration modelling and analysis via functional series time-dependent ARMA (FS-TARMA) models-A critical survey[J]. Mechanical Systems and Signal Pro-cessing, 2014, 47(1-2):175-224.
[25] HEYLEN W, LAMMENS S, SAS P. Modal analysis theory and testing[M]. Leuven:Katholieke Universiteit Leuven, 2007.
[26] 杨武. 时变结构模态参数的时域辨识方法及在线辨识技术研究[D]. 北京:北京理工大学, 2015:25-45. YANG W. Study on modal parameter estimation and online identification technology for linear time-varying structures in time-domain[D]. Beijing:Beijing Institute of Technology, 2015:25-45(in Chinese).
[27] GOODFELLOW I, BENGIO Y, COURVILLE A. Deep learning[M]. Cambridge:MIT Press, 2016:374-419.
[28] POULIMENOS A G, FASSOIS S D. Output-only sto-chastic identification of a time-varying structure via func-tional series TARMA models[J]. Mechanical Systems and Signal Processing, 2009, 23(4):1180-1204.
[29] GLOROT X, BENGIO Y. Understanding the difficulty of training deep feedforward neural networks[C]//Proceedings of 13th International Conference on Artificial Intelligence and Statistics, 2010:1-8.
[30] 陆宇平, 杨朝星,刘洋洋. 空中加油系统的建模与控制技术综述[J]. 航空学报, 2014, 35(9):2375-2389. LU Y P, YANG C X, LIU Y Y. A survey of modeling and control technologies for aerial refueling system[J]. Acta Aeronautica et Astronautica Sinica, 2014, 35(9):2375-2389(in Chinese).
[31] 陆正亮, 张翔,于永军,等. 纳卫星变轨段质量矩姿态控制系统设计[J]. 航空学报, 2016, 38(6):320778. LU Z L, ZHANG X, YU Y J, et al. Design of moving-mass attitude control system for nanosatellites in orbital transfer stage[J]. Acta Aeronautica et Astronautica Sinica, 2016, 38(6):320778(in Chinese).
[32] POULIMENOS A G, FASSOIS S D. Parametric time-domain methods for non-stationary random vibration modelling and analysis-A critical survey and comparison[J]. Mechanical Systems and Signal Processing, 2006, 20(4):763-816.
[33] REYNDERS E, HOUBRECHTS J, DE ROECK G. Fully automated (operational) modal analysis[J]. Mechanical Systems and Signal Processing, 2012, 29:228-250.
[34] SHAO M, NIKIAS C L. Signal-processing with fractional lower order moments-Stable processes and their applications[J]. Proceedings of the IEEE, 1993, 81(7):986-1010.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献