Structural damage identification based on the wavelet scattering convolution neural network

PDF(3396 KB)

Journal of Vibration and Shock ›› 2023, Vol. 42 ›› Issue (14) : 138-146.

MA Yafei1，LI Cheng1，HE Yu1，WANG Lei1，TU Ronghui1,2

Author information +

History +

Abstract

Damage identification is one of the key issues in the field of structural condition assessment, which is of great importance to ensure structural safety. Deep learning algorithm has led to many breakthroughs in vibration-based structural damage identification, but it is still an urgent technical challenge to obtain the key information on structural damage from massive amounts of data. This paper proposes a multi-type structural damage identification model based on one-dimensional convolutional network (1DCNN) deep learning. The wavelet scattering transform is used to replace the convolutional filter in the first layer of the 1DCNN architecture. The scattering coefficients are used to achieve dimensionality reduction and feature extraction of the original data in the input layer, and the CNN convolutional layer, activation layer and pooling layer are combined to achieve feature enhancement processing of monitoring data. The 1DCNN fully-connected layer and Softmax function are combined to classify the feature data, thus realizing the location and quantitative identification of multi type structural damage. The above frame is verified by two numerical models of steel truss structure and cable-stayed bridge. The results show that compared with the normal convolutional neural network model, the accuracy of structural damage identification based on the wavelet scattering based convolutional neural network is significantly improved, and the accuracy of damage classification is more than 95%. In addition, with the increase of the proportion of environmental noise of sensor data, the accuracy of the wavelet scattering convolutional neural network damage classification slightly decreases but still has high accuracy, indicating that this method has strong robustness and anti noise ability.

Key words

structural condition assessment / deep learning / wavelet scattering transform / convolutional neural network / damage identification

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

MA Yafei1，LI Cheng1，HE Yu1，WANG Lei1，TU Ronghui1,2. Structural damage identification based on the wavelet scattering convolution neural network[J]. Journal of Vibration and Shock, 2023, 42(14): 138-146

References

[1] 朱宏平, 余璟, 张俊兵. 结构损伤动力检测与健康监测研究现状与展望 [J]. 工程力学, 2011, 28(2): 1-11.
ZHU Hongping, YU Jing, ZHANG Junbing. A summary review and advantages of vibration-based damage identification methods in structural health monitoring [J]. Engineering Mechanics, 2011, 28(2): 1-11.
[2] BAO Y, CHEN Z, WEI S, et al. The state of the art of data science and engineering in structural health monitoring [J]. Engineering, 2019, 5(2): 34-38.
[3] LIU S, LIU T, ZHOU J, et al. Relationship between shear-stress distribution and resulting acoustic-emission variation along concrete joints in prefabricated girder structures [J]. Engineering Structures, 2019, 196(1): 109319.
[4] PANDEY A, BISWAS M, SAMMAN M. Damage detection from changes in curvature mode shapes [J]. Journal of Sound and Vibration, 1991, 145(2): 321-332.
[5] SALAWU O. Detection of structural damage through changes in frequency: a review [J]. Engineering Structures, 1997, 19(9): 718-723.
[6] DöHLER M, HILLE F, MEVEL L, et al. Structural health monitoring with statistical methods during progressive damage test of S101 Bridge [J]. Engineering Structures, 2014, 69(15): 183-193.
[7] 赵俊, 程良彦, 马宏伟. 基于曲率模态的拱板结构损伤识别 [J]. 暨南大学学报(自然科学版), 2008, 29(5): 470-477.
ZHAO Jun, CHEN Liangyan, MA Hongwei. The damage detection in the arch basing on the changes in curvature mode shape [J]. Journal of Jinan University (Natural Science), 2008, 29(5): 470-477.
[8] 聂振华, 程良彦, 马宏伟. 基于结构动力特性的损伤检测可视化方法 [J]. 振动与冲击, 2011, 30(12): 7-13.
NIE Zhenhua, CHENG Liangyan, MA Hongwei. Visualization method for structural damage detection based on its dynamic characteristics [J]. Journal of Vibration and Shock, 2011, 30(12): 7-13
[9] 周计祥, 吴邵庆, 董萼良, 等. 基于应变模态的模态应变能损伤识别方法 [J]. 振动､测试与诊断, 2019, 39(1): 25-31+218-219.
ZHOU Jixiang, WU Shaoqing, DONG Eliang, et al. A strain mode-based approach to modal strain energy damage identification [J]. Jouranl of Vibration, Measurement & Diagnosis, 2019, 39(1): 28-31+218-219.
[10] HOU Z, NOORI M, AMAND R S. Wavelet-based approach for structural damage detection [J]. Journal of Engineering Mechanics, 2000, 126(7): 677-683.
[11] LI Z, PARK H S, ADELI H. New method for modal identification of super high‐rise building structures using discretized synchrosqueezed wavelet and Hilbert transforms [J]. The Structural Design of Tall and Special Buildings, 2017, 26(3): e1312.
[12] SONY S, GAMAGE S, SADHU A, et al. Vibration-based multiclass damage detection and localization using long short-term memory networks [C] // Structures. Elsevier, 2022, 35: 436-451.
[13] TIBADUIZA D A, TORRES-ARREDONDO M A, MUJICA L, et al. A study of two unsupervised data driven statistical methodologies for detecting and classifying damages in structural health monitoring [J]. Mechanical Systems and Signal Processing, 2013, 41(1-2): 467-484.
[14] WHEELER B J, KARIMI H A. Deep learning-enabled semantic inference of individual building damage magnitude from satellite images [J]. Algorithms, 2020, 13(8): 195.
[15] XU J, WANG H, CUI C, et al. Oil spill monitoring of shipborne radar image features using SVM and local adaptive threshold [J]. Algorithms, 2020, 13(3): 69.
[16] SATPAL S B, GUHA A, BANERJEE S. Damage identification in aluminum beams using support vector machine: Numerical and experimental studies [J]. Structural Control and Health Monitoring, 2016, 23(3): 446-457.
[17] ARANGIO S, BONTEMPI F. Structural health monitoring of a cable-stayed bridge with Bayesian neural networks [J]. Structure and Infrastructure Engineering, 2015, 11(4): 575-587.
[18] 孙志军, 薛磊, 许阳明, 等. 深度学习研究综述 [J]. 计算机应用研究, 2012, 29(8): 2806-2810.
SUN Zhijun, XUE Lei, XU Yangming, et al. Overview of deep learning [J]. Application Research of computers, 2012, 29(8): 2806-2810.
[19] ZHANG Y, MIYAMORI Y, MIKAMI S, et al. Vibration‐based structural state identification by a 1‐dimensional convolutional neural network [J]. Computer-Aided Civil and Infrastructure Engineering, 2019, 34(9): 822-839.
[20] LIN Y Z, NIE Z H, MA H W. Structural damage detection with automatic feature‐extraction through deep learning [J]. Computer-Aided Civil and Infrastructure Engineering, 2017, 32(12): 1025-1046.
[21] ABDELJABER O, AVCI O, KIRANYAZ S, et al. Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks [J]. Journal of Sound and Vibration, 2017, 388:154-170.
[22] LI F, LIU M, ZHAO Y, et al. Feature extraction and classification of heart sound using 1D convolutional neural networks [J]. EURASIP Journal on Advances in Signal Processing, 2019, 2019(1): 1-11.
[23] KRIZHEVSKY A, SUTSKEVER I, HINTON G. ImageNet Classification with Deep Convolutional Neural Networks [J]. Advances in Neural Information Processing Systems, 2012, 25(2).
[24] BRUNA J, MALLAT S. Invariant scattering convolution networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8): 1872-86.
[25] OYALLON E, MALLAT S. Deep roto-translation scattering for object classification [C] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015: 2865-2873.
[26] MALLAT S. Group invariant scattering [J]. Communications on Pure and Applied Mathematics, 2012, 65(10): 1331-1398.
[27] WIATOWSKI T, BÖLCSKEI H. A mathematical theory of deep convolutional neural networks for feature extraction [J]. IEEE Transactions on Information Theory, 2017, 64(3): 1845-1866.
[28] WIATOWSKI T, TSCHANNEN M, STANIC A, et al. Discrete deep feature extraction: A theory and new architectures [C] // International Conference on Machine Learning. PMLR, 2016: 2149-2158.
[29] IOFFE S, SZEGEDY C. Batch Normalization: accelerating deep network training by reducing internal covariate shift [J/OL]. Proceedings of the 32nd International Conference on International Conference on Machine Learning. 2015. 03167.
[30] GIMPEL K, SMITH N A. Softmax-Margin CRFs: Training Log-Linear Models with Cost Functions [C] // Human Language Technologies: Conference of the North American Chapter of the Association of Computational Linguistics. Los Angeles, DBLP, 2010.