基于一维空洞卷积神经网络的结构损伤检测方法研究

李睿; 张纯

doi:10.13204/j.gyjzg20103011

基于一维空洞卷积神经网络的结构损伤检测方法研究

doi: 10.13204/j.gyjzg20103011

李睿,
张纯^,

南昌大学建筑工程学院, 南昌 330031

基金项目:

国家自然科学基金项目（51469016）；江西省自然科学基金项目（20202BAB204029）；江西省研究生教改项目（JXYJG-2019-018）；江西省研究生创新专项资金项目（CX2018057）

详细信息

作者简介:
李睿,男,1996年出生,硕士

通讯作者:
张纯,男,1976年出生,教授,zhangchun@ncu.edu.cn。

计量
- 文章访问数: 140
- HTML全文浏览量: 14
- PDF下载量: 4
- 被引次数: 0
出版历程
- 收稿日期: 2020-10-30
- 网络出版日期: 2022-02-21

RESEARCH ON STRUCTURAL DAMAGE DETECTION METHOD BASED ON ONE-DIMENSIONAL DILATED CONVOLUTION NEURAL NETWORK

LI Rui,
ZHANG Chun^,

School of Civil Engineering and Architecture, Nanchang University, Nanchang 330031, China

摘要

摘要: 利用振动响应对结构损伤进行智能检测和诊断意义重大，但传统深度卷积神经网络模型在处理结构振动信号时存在模型参数多、信号细节信息损失、泛化性能不好等问题，因此提出一种基于一维空洞卷积神经网络的结构损伤检测方法。该方法所使用的神经网络模型通过空洞卷积来代替传统的卷积与池化的组合层，保持参数数量不变的情况下增大感受野；同时采用全局池化代替传统的全连接层来减少模型参数，以防止过拟合的出现，进而针对实际振动信号数据集类别不平衡的现象，通过对不同类别信号设置惩罚权重来训练代价敏感分类器，能有效提高样本不平衡情况下的结构损伤检测精度。卡塔尔大学看台缩尺模型损伤试验的验证与应用分析表明，利用该方法能够在不损失信号细节信息的情况下，从原始加速度信号中自动提取最优特征并分类，达到较高的识别准确率和分类成功率，以用于实时损伤检测。
- 损伤检测 /
- 空洞卷积 /
- 全局池化 /
- 样本不平衡 /
- 代价敏感分类器
Abstract: It is of great significance to use vibration response for intelligent detection and diagnosis of structural damage. However, the traditional deep convolution neural network model has some problems in processing structural vibration signal, such as large model parameters, loss of signal detail information and poor generalization performance. Therefore, a structural damage detection method based on one-dimensional dilated convolution neural network was proposed. The neural network model used in this method used dilated convolution to replace the traditional combination layer of convolution and pooling, and increased the receptive field while keeping the number of parameters unchanged; global pooling layer was used to replace the traditional full connection layer to reduce the model parameters and prevent the occurrence of over fitting, and then according to the phenomenon of class imbalance in the actual vibration signal data set, the cost sensitive classifier was trained by setting penalty weights for different types of signals, which could effectively improve the structural damage detection accuracy under unbalanced samples. The verification and application analysis of the damage experiment on the grandstand scale model of the University of Qatar showed that the method proposed in the paper could automatically extract the optimal features from the original acceleration signals and classify them without losing the detailed information of the signals, thus achieving high recognition accuracy and classification success ratio, which could be used for real-time damage detection.
- damage detection /
- dilated convolution /
- global pooling /
- sample imbalance /
- cost-sensitive classifier

HTML全文

参考文献(18)

[1]	BAO Y, TANG Z, LI H, et al.Computer Vision and Deep Learning-Based Data Anomaly Detection Method for Structural Health Monitoring[J].Structural Health Monitoring, 2019, 18(2):401-421.
[2]	卢宏涛, 张秦川.深度卷积神经网络在计算机视觉中的应用研究综述[J].数据采集与处理, 2016(1):1-17.
[3]	CHA Y J, CHOI W, BÜYÜKÖZTÜRK O.Deep Learning-Based Crack Damage Detection Using Convolutional Neural Networks[J].Computer-Aided Civil & Infrastructure Engineering, 2017, 32(5):361-378.
[4]	李雪松, 马宏伟, 林逸洲.基于卷积神经网络的结构损伤识别[J].振动与冲击, 2019, 38(1):167-175.
[5]	BAO Y, TANG Z, LI H, et al.Computer Vision and Deep Learning-Based Data Anomaly Detection Method for Structural Health Monitoring[J].Structural Health Monitoring an International Journal, 2018, 18(2):401-421.
[6]	KIRANYAZ S, AVCI O, ABDELJABER O, et al.1D Convolutional Neural Networks and Applications:A Survey[J].Mechanical Systems and Signal Processing, 2021, 151.DOI: 10.1016/j.ymssp.2020.107398.
[7]	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.
[8]	LECUN Y, BOTTOU L.Gradient-Based Learning Applied to Document Recognition[J].Proceedings of the IEEE, 1998, 86(11):2278-2324.
[9]	ABDELJABER O, AVCI O, KIRANYAZ S, et al.Real-Time Vibration-Basedstructural Damage Detection Using One-Dimensional Convolutional Neural Networks[J].Journal of Sound & Vibration, 2017, 388:154-170.
[10]	AVCI O, ABDELJABER O, KIRANYAZ S, et al.Structural Damage Detection in Real Time:Implementation of 1D Convolutional Neural Networks for SHM Applications[C]//Structural Health Monitoring & Damage Detection, Conference Proceedings of the Society for Experimental Mechanics Series.Garden Grove:2017:49-54.
[11]	AVCI O, ABDELJABER O, KIRANYAZ S, et al.Wireless and Real-Time Structural Damage Detection:A Novel Decentralized Method for Wireless Sensor Networks[J].Journal of Sound and Vibration, 2018, 424:158-172.
[12]	周飞燕, 金林鹏, 董军.卷积神经网络研究综述[J].计算机学报, 2017, 40(6):1230-1251.
[13]	YU F, KOLTUN V.Multi-Scale Context Aggregation by Dilated Convolutions[C]//International Conference on Learning Representation.San Juan:2016.
[14]	CHEN L C, PAPANDREOU G, KOKKINOS I, et al.Deeplab:Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected Crfs[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 40(4):834-848.
[15]	潘辉, 郑威, 张莹莹.基于改进残差网络对心电信号的识别[J].数据采集与处理, 2020, 35(4):682-692.
[16]	HINTON G E, SRIVASTAVA N, KRIZHEVSKY A, et al.Improving Neural Networks by Preventing Co-Adaptation of Feature Detectors[J].Computer Science, 2012, 3(4):212-223.
[17]	LIN M, CHEN Q, YAN S.Network in Network[J].Computer Science, 2013, 4(5):1-10.
[18]	HE H, GARCIA E A.Learning from Imbalanced Data[J].IEEE Transactions on Knowledge and Data Engineering, 2009, 21(9):1263-1284.