Using Multi-dimensional Features of Structure to Construct Graph Convolutional Neural Network for Structural Damage Identification
Jianhui YANG, Qingxuan ZHAO, Pulin PU
Author information+
Show less
文章历史+
Received
Published
2024-11-13
2025-08-25
Issue Date
2026-02-12
PDF (7886K)
摘要
以数据为驱动的深度学习结构损伤识别(structural damage identification,SDI)效果受结构复杂程度、模型构建方法及数据规模等因素影响较大.本文引入图卷积神经网络(graph convolutional neural network,GCN)以整合结构节点间的属性特征,从图的视角挖掘节点间的复杂属性关系,为SDI提供多维度学习信息.为此,设计了一种融合结构多维特征的图卷积神经网络模型(graph convolutional neural network integrating multi-dimensional features of structure,S-GCN),基于结构振动数据构造损伤特征矩阵,并通过衍生图网络,以图的节点和边表征结构节点的连接关系,构建边索引矩阵,将结构损伤状态、振动数据及节点属性等多维特征信息输入GCN进行结构损伤特征提取及预测识别,探索结构多维特征信息驱动下的GCN在损伤预测中的应用效果.通过两个钢结构验证方法的可行性及有效性,结果表明,S-GCN能够整合结构多维特征信息,对两个结构对象均实现了较高的损伤预测准确性,并展现出良好的噪声鲁棒性.进一步的对比分析显示,相较于三种非GCN模型,S-GCN能够高效地依托节点间关系快速更新节点特征并预测节点损伤状态,其损伤识别准确率、计算效率及网络各层演进过程均优于对比模型,验证了在结构损伤识别中融合结构空间特征的有效性.
Abstract
The effect of data-driven deep learning structural damage identification (SDI) is greatly affected by factors such as structural complexity, model construction method and data size, etc. We introduce graph convolutional neural network (GCN) to integrate the attribute features between structural nodes, explore the complex attribute relationships between nodes from a graph perspective, and provide multi-dimensional learning information for SDI. To this end, a graph convolutional neural network integrating multi-dimensional features of structure (S-GCN) model was designed, which integrates multidimensional structural features. Based on structural vibration data, a damage feature matrix was constructed, and through a derived graph network, the connection relationship between nodes and edges in the graph was represented. An edge index matrix was constructed, and multidimensional feature information such as structural damage status, vibration data, and node attributes was input into GCN for structural damage feature extraction and identification. The application effect of GCN in damage identification driven by multidimensional structural feature information was explored. The feasibility and effectiveness of two steel structure verification methods were verified, and the results showed that S-GCN can integrate multi-dimensional structural feature information, achieve high damage identification accuracy for both structural objects, and demonstrate good noise robustness. Further comparative analysis shows that S-GCN can efficiently update node features and predict node damage status based on inter-node relationships compared with the three non-GCN models. Its damage recognition accuracy, computational efficiency, and network layer evolution process are all better than the comparative models, verifying the effectiveness of integrating structural spatial features in structural damage recognition.
特征提取层是图卷积神经网络的核心组成部分,它通过接收输入层的一维振动信号和邻接矩阵,并对输入的特征矩阵和边索引矩阵进行卷积操作来提取结构的损伤特征,进一步通过ReLU(rectified linear unit)激活函数对卷积层的输出进行处理.本文所设计的S-GCN模型特征提取层主要由两层图卷积、ReLU和全连接层组成.
LIH M, DONGM M, XIONGX, et al .State-of-the-art review of vibration-based damage identification framework for structures[J].Building Structure, 2021, 51(4): 45-50.(in Chinese)
[3]
LIJ, HAOH .A review of recent research advances on structural health monitoring in Western Australia[J].Structural Monitoring and Maintenance,2016,3(1):33-49.
[4]
AGDASD, RICEJ A, MARTINEZJ R,et al .Comparison of visual inspection and structural-health monitoring As bridge condition assessment methods[J]. Journal of Performance of Constructed Facilities, 2016, 30(3): 04015049.
ZHAOJ, CHENGL Y, MAH W .The damage detection in the arch basing on the changes in curvature mode shape[J].Journal of Jinan University (Natural Science Edition), 2008, 29(5):470-477.(in Chinese)
NIEZ H, CHENGL Y, MAH W. Visualization method for structural damage detection based on its dynamic characteristics[J]. Journal of Vibration and Shock, 2011, 30(12): 7-13.(in Chinese)
[9]
ZHANGJ Q, JINZ H, TENGS,et al. Structural damage detection based on decision-level fusion with multi-vibration signals[J]. Measurement Science and Technology,2022,33(10):105112.
JIAR, LIUY, WANGL Y,et al .Research on real-time evaluation methods for main shaft running load of special vehicle transmission[J]. Journal of Hunan University (Natural Sciences),2024, 51(10): 58-66.(in Chinese)
MAY F, LIC, HEY, et al. Structural damage identification based on the wavelet scattering convolution neural network[J].Journal of Vibration and Shock,2023, 42(14): 138-146.(in Chinese)
LIUY C, JIANGY F, WANGS Q,et al .Data fusion and residual convolutional auto-encoder based structural damage identification[J].Journal of Vibration and Shock,2023,42(4):194-203.(in Chinese)
LUOY P, WANGL K, LIAOF Y,et al .Vibration-based structural damage identification by 1-dimensional convolutional neural network[J].Earthquake Engineering and Engineering Dynamics,2021,41(4):145-156.(in Chinese)
[20]
YANGJ H, PENGZ R .A new convolutional neural network-based framework and data construction method for structural damage identification considering sensor placement[J].Measurement Science and Technology,2023,34(7): 075008.
[21]
TRUONGT T, LEEJ, NGUYEN-THOIT. Joint damage detection of structures with noisy data by an effective deep learning framework using autoencoder-convolutional gated recurrent unit[J].Ocean Engineering,2022,243:110142.
[22]
LIT F, ZHOUZ, LIS N,et al .The emerging graph neural networks for intelligent fault diagnostics and prognostics:a guideline and a benchmark study[J]. Mechanical Systems and Signal Processing,2022,168:108653.
[23]
WUZ H, PANS R, CHENF W,et al .A comprehensive survey on graph neural networks[J]. IEEE Transactions on Neural Networks and Learning Systems,2021,32(1):4-24.
[24]
DANGV H, VUT C, NGUYENB D,et al .Structural damage detection framework based on graph convolutional network directly using vibration data[J]. Structures,2022,38: 40-51.
[25]
AGGELISD G, PATSONAKISC, STIROSS C. Vibration‐based structural health monitoring of an old masonry arch bridge From automated OMA to damage detection[J]. Structural Control and Health Monitoring, 2014,21(7): 918-940.
[26]
WANGL, WANGW. Damage identification method based on frequency response function and support vector machine[J]. Mechanical Systems and Signal Processing, 2018, 99: 760-775.
[27]
NGUYENK H, KIMS Y, KANGS H. Enhanced damage detection of a structure using statistical damage index and artificial neural network[J]. Structural Control and Health Monitoring, 2019, 26(4):23-37.
YANGJ X, ZHANGL K, LIR,et al .Research on bridge structural damage detection based on convolutional and long short-term memory neural networks[J].Journal of Railway Science and Engineering,2020,17(8):1893-1902.(in Chinese)
[30]
BAOQ, XUY. Structural damage identification via modal parameters estimation based on cross-entropy optimization algorithm[J]. Structural Health Monitoring, 2021, 20(5):12.
TANGQ Z, XINJ Z, ZHOUJ T,et al .Structural damage identification method based on AR-GP model[J]. Journal of Vibration and Shock,2021,40(9):102-109.(in Chinese)
[33]
AVCIO, ABDELJABERO, KIRANYAZS,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.
[34]
ABDELJABERO, AVCIO, KIRANYAZM S,et al .1-D CNNs for structural damage detection:Verification on a structural health monitoring benchmark data[J]. Neurocomputing,2018,275:1308-1317.
[35]
ABDELJABERO, AVCIO, KIRANYAZS,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.
[36]
AVCIO, ABDELJABERO, KIRANYAZS,et al .Structural damage detection in real time: implementation of 1D convolutional neural networks for SHM applications[M]//Structural Health Monitoring & Damage Detection,Volume 7.Cham:Springer International Publishing,2017:49-54.
[37]
KIRANYAZS, INCET, ABDELJABERO,et al. 1-D convolutional neural networks for signal processing applications[C]//ICASSP 2019-2019 IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP). Brighton,UK. IEEE,2019:8360-8364.
[38]
AVCIO, ABDELJABERO, KIRANYAZS,et al .Convolutional neural networks for real-time and wireless damage detection[M]//Dynamics of Civil Structures,Volume 2. Cham:Springer International Publishing,2019:129-136.
[39]
LAURENSV D M, HINTONG.Visualizing data using t-SNE[J].Journal of Machine Learning Research, 2008, 9(2605):2579-2605.