A feature learning-based method for impact load reconstruction and localization of the plate-rib assembled structure

Abstract

Impact load is the load that machines frequently experienced in engineering applications. Its time-history reconstruction and localization are crucial for structural health monitoring and reliability analysis. However, when identifying random impact loads, conventional inversion methods usually do not perform well because of complex formula derivation, infeasibility of nonlinear structure, and ill-posed problem. Deep learning methods have great ability of feature learning and nonlinear representation as well as comprehensive regularization mechanism. Therefore, a new feature learning-based method is proposed to conduct impact load reconstruction and localization. The proposed method mainly includes two parts. The first part is designed to reconstruct impact load, named convolutional-recurrent encoder–decoder neural network (ED-CRNN). The other part is constructed to localize impact load, called deep convolutional-recurrent neural network (DCRNN). The ED-CRNN utilizes the one-dimensional (1-D) convolutional encoder–decoder to obtain low-dimension feature representations of input signals. Two long short-term memory (LSTM) layers and a bidirectional LSTM (BiLSTM) layer are uniformly distributed in this network to learn the relationship between input features and the output load in time steps. The DCRNN is constructed mainly by two 1-D convolutional neural network (CNN) layers and two BiLSTM layers to learn high-hidden-level spatial as well as temporal features. The fully connected layers are placed at the end to localize an impact load. The effectiveness of the proposed method was demonstrated by two numerical studies and two experiments. The results reveal that the proposed method has the ability to accurately and quickly reconstruct and localize the impact load of complex assembled structure. Furthermore, the performance of the DCRNN is related to the number of sensors and the architecture of the network. Meanwhile, the strategy of alternating layout is proposed to reduce the number of training locations.

Keywords

Identification of impact load feature learning convolutional-recurrent encoder–decoder convolutional-recurrent neural network assembled structure

Get full access to this article

View all access options for this article.

References

Yanyutin

Yanchevsky

. Identification of an impulse load acting on an axisymmetrical hemispherical shell. Int J Sol Struct 2004; 41: 3643–3652.

Uhl

. The inverse identification problem and its technical application. Arch Appl Mech 2007; 77: 325–337.

Park

Kim

Jun

S-M

, et al. Localizations and force reconstruction of low-velocity impact in a composite panel using optical fiber sensors. Adv Compos Mater 2012; 21: 357–369.

Huang

Zeng

Lin

, et al. Baseline-free damage detection in composite plates using edge-reflected Lamb waves. Compos Struct 2020; 247: 112423.

Wang

, et al. A novel deep learning-based method for damage identification of smart building structures. Structural Health Monitoring 2019; 18: 143–163. DOI: 10.1177/1475921718804132.

Qiu

Zhang

Xie

, et al. Localisation of unknown impact loads on a steel plate using a pattern recognition method combined with the similarity metric via structural stress responses in the time domain. Mech Syst Signal Process 2019; 128: 429–445.

Mao

Xie

Zhang

. Identification of impact load using transient statistical energy analysis method. Int J Appl Electromagn Mech 2014; 45: 353–361.

Jiang

. Dynamic load identification for interval structures under a presupposition of ‘being included prior to being measured’. Appl Math Model 2020; 85: 107–123.

Liu

Sun

Han

, et al. Dynamic load identification for stochastic structures based on Gegenbauer polynomial approximation and regularization method. Mech Syst Signal Process 2015; 56–57: 35–54.

10.

Qiu

Zhang

, et al. Unknown impact force localisation and reconstruction in experimental plate structure using time-series analysis and pattern recognition. Int J Mech Sci 2020; 166: 105231. DOI: 10.1016/j.ijmecsci.2019.105231.

11.

Hollandsworth

Busby

. Impact force identification using the general inverse technique. Int J Impact Eng 1989; 8: 315–322.

12.

Movahedian

Boroomand

. Inverse identification of time-harmonic loads acting on thin plates using approximated Green’s functions. Inverse Probl Sci Eng 2016; 24: 1475–1493.

13.

Zhao

Chen

, et al. Identification of distributed dynamic excitation based on Taylor polynomial iteration and cubic Catmull–Rom spline interpolation. Inverse Probl Sci Eng 2020; 28: 220–237.

14.

Zhang

X-N

Xie

S-L

. A new method of load identification in time domain. Int J Appl Electromagn Mech 2010; 33: 1025–1032.

15.

Inoue

Harrigan

Reid

. Review of inverse analysis for indirect measurement of impact force. Appl Mech Rev 2001; 54: 503–524.

16.

Sun

Liu

Han

, et al. A new improved regularization method for dynamic load identification. Inverse Probl Sci Eng 2014; 22: 1062–1076. DOI: 10.1080/17415977.2013.854353.

17.

Mao

Guo

Zhao

. A state space force identification method based on Markov parameters precise computation and regularization technique. J Sound Vib 2010; 329: 3008–3019.

18.

Chen

Qin

Zhao

, et al. Toward efficacy of piecewise polynomial truncated singular value decomposition algorithm in moving force identification. Adv Struct Eng 2019; 22: 2687–2698.

19.

Chen

Chan

THT

. A truncated generalized singular value decomposition algorithm for moving force identification with ill-posed problems. J Sound Vib 2017; 401: 297–310.

20.

Liu

Shepard

. Dynamic force identification based on enhanced least squares and total least-squares schemes in the frequency domain. J Sound Vib 2005; 282: 37–60.

21.

Jang

Baek

Choi

, et al. A new method for measuring nonharmonic periodic excitation forces in nonlinear damped systems. Mech Syst Signal Process 2011; 25: 2219–2228.

22.

Hochstenbach

Noschese

Reichel

. Fractional regularization matrices for linear discrete ill-posed problems. J Eng Math 2015; 93: 113–129.

23.

Hochstenbach

Reichel

. Combining approximate solutions for linear discrete ill-posed problems. J Comput Appl Math 2012; 236: 2179–2185.

24.

Loubes

J-M

Pelletier

. Maximum entropy solution to ill-posed inverse problems with approximately known operator. J Math Anal Appl 2008; 344: 260–273.

25.

Chen

Chan

THT

Nguyen

. Moving force identification based on modified preconditioned conjugate gradient method. J Sound Vib 2018; 423: 100–117.

26.

Qiao

Chen

Luo

, et al. A novel method for force identification based on the discrete cosine transform. J Vib Acoust Trans ASME 2015; 137. DOI: 10.1115/1.4030616.

27.

Alqam

HaM

Dhingra

. Frequency response-based indirect load identification using optimum placement of Strain Gages and accelerometers. J Vib Acoust Trans ASME 2019; 141. DOI: 10.1115/1.4042709.

28.

Zhang

Qiu

Zhu

, et al. Novel computation method of reducing ill-posedness for structural static distributed load identification by optimising strain gauge locations. Mech Syst Signal Process 2019; 124: 83–110.

29.

Tran

Inoue

. Further development and experimental verification of wavelet deconvolution technique for impact force reconstruction. Mech Syst Signal Process 2021; 148: 107165.

30.

Mao

Yan

. A new SVM regression approach for mechanical load identification. Int J Appl Electromagn Mech 2010; 33: 1001–1008.

31.

Yan

Sun

Büyüköztürk

. Impact load identification for composite structures using Bayesian regularization and unscented Kalman filter. Struct Control Heal Monit 2017; 24: 1–18.

32.

Sarego

Zaccariotto

Galvanetto

. Artificial neural networks for impact force reconstruction on composite plates and relevant uncertainty propagation. IEEE Aerosp Electron Syst Mag 2018; 33: 38–47.

33.

Yang

Yan

. Parameters identification of moving load using ANN and dynamic strain. Shock Vib 2016; 2016: 1–13. DOI: 10.1155/2016/8249851.

34.

Ren

Chen

, et al. A deep learning-based computational algorithm for identifying damage load condition: an artificial intelligence inverse problem solution for failure analysis. C–Comput Model Eng Sci 2018; 117: 287–307.

35.

Cooper

DiMaio

. Static load estimation using artificial neural network: application on a wing rib. Adv Eng Softw 2018; 125: 113–125.

36.

Zhou

Dong

Guan

, et al. Impact load identification of nonlinear structures using deep recurrent neural network. Mech Syst Signal Process 2019; 133: 106292.

37.

Wang

Yang

, et al. A new fault diagnosis method based on adaptive spectrum mode extraction. Struct Heal Monit 2021; 107: 147592172098694. DOI: 10.1177/1475921720986945.

38.

Badrinarayanan

Kendall

Cipolla

. SegNet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans Pattern Anal Mach Intell 2017; 39: 2481–2495. DOI: 10.1109/TPAMI.2016.2644615.

39.

Cho

Van Merriënboer

Gulcehre

, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation. In: EMNLP 2014–2014 conference on empirical methods in natural language processing, proceedings of the conference, Doha, Qatar, 2014. DOI: 10.3115/v1/d14-1179.

40.

Guo

Lei

Xing

, et al. Deep convolutional transfer learning network: a new method for intelligent fault diagnosis of machines with unlabeled data. IEEE Trans Ind Electron 2019; 66: 7316–7325.

41.

Chen

Xie

Yuan

, et al. Research on a real-time monitoring method for the wear state of a tool based on a convolutional bidirectional LSTM model. Symmetry (Basel) 2019; 11: 1233. DOI: 10.3390/sym11101233.

42.

Zeng

Huang

Luo

, et al. Damage imaging that exploits multipath scattered lamb waves. Struct Health Monitoring; 19: 1629–1644, 2020. DOI: 10.1177/1475921719892828.