|本期目录/Table of Contents|

[1]李书进,张杰玲,赵源.基于响应数据图像化和深度残差收缩网络的结构损伤诊断[J].建筑科学与工程学报,2026,(01):28-40.[doi:10.19815/j.jace.2025.01031]
　LI Shujin,ZHANG Jieling,ZHAO Yuan.Structural damage diagnosis based on image processing of response data and deep residual shrinkage network[J].Journal of Architecture and Civil Engineering,2026,(01):28-40.[doi:10.19815/j.jace.2025.01031]
点击复制

基于响应数据图像化和深度残差收缩网络的结构损伤诊断(PDF)
分享到：

《建筑科学与工程学报》[ISSN:1673-2049/CN:61-1442/TU]

卷:
期数:
2026年01期
页码:
28-40
栏目:
智能检测与建造技术专栏
出版日期:
2026-01-20

文章信息/Info

Title:
Structural damage diagnosis based on image processing of response data and deep residual shrinkage network
文章编号:
1673-2049(2026)01-0028-13
作者:
李书进1，张杰玲1，赵源2
1. 武汉理工大学 土木工程与建筑学院，湖北 武汉 430070; 2. 武汉理工大学三亚科教创新园，海南 三亚 572025
Author(s):
LI Shujin1, ZHANG Jieling1, ZHAO Yuan2
1. School of Civil Engineering and Architecture, Wuhan University of Technology, Wuhan 430070, Hubei, China; 2. Sanya Science and Education Innovation Park of Wuhan University of Technology, Sanya 572025, Hainan, China
关键词:
损伤诊断; 深度残差收缩网络; 卷积神经网络; 格拉姆角场; 多标签分类; 数据增强
Keywords:
damage diagnosis;  deep residual shrinkage network;  convolutional neural network;  Gramian angular field;  multilabel classification;  data enhancement
分类号:
TU312.3
DOI:
10.19815/j.jace.2025.01031
文献标志码:
A
摘要:
利用卷积神经网络(CNN)处理二维图像的优势以及深度残差收缩网络(DRSN)在抗噪性、稳定性和鲁棒性上的良好表现，提出一种将动力响应信号图像化处理后利用DRSN对结构损伤进行诊断的方法。以复杂损伤工况下的平面和空间框架结构的节点损伤诊断问题为研究对象，从模型的样本输入和特征学习两方面出发，通过格拉姆角场(GAF)变换和数据增强处理将各监测点的一维结构动力响应信号构造为图像增强样本集，同时构建了适用于框架结构节点损伤位置和损伤程度诊断的DRSN多标签分类模型，并从训练收敛速度、诊断准确率、训练样本类别及网络结构几方面对其诊断性能进行了研究。通过对所提方法在强噪声干扰下的抗噪性能及处理新样本时的泛化性能进行研究，验证其有效性和实用性。结果表明：采用图像增强样本集训练的DRSN模型在诊断准确率、迭代收敛速度和稳定性方面的表现优于普通卷积神经网络模型，且在不同的诊断对象上表现出了良好的鲁棒性和适应性；DRSN的自适应调整阈值降噪机制具有出色的抗噪性能和泛化性能，使其在强噪声、小样本情况下的表现更具优势。
Abstract:
Utilizing the advantages of convolutional neural network(CNN) in processing two-dimensional images and the good performance of deep residual shrinkage network(DRSN) in noise resistance, stability and robustness, a new method of structural damage diagnosis using DRSN after two-dimensional processing on dynamic response signal was proposed. Taking the node damage diagnosis problem of planar and spatial frame structures under complex damage conditions as the research object, starting from two aspects of model sample input and feature learning, the one-dimensional structural dynamic response signals were constructed into image enhancement samples by means of Gramian angular field(GAF) transformation and data enhancement processing. At the same time, a DRSN multi-label classification model suitable for the diagnosis of damage location and damage degree of frame structure nodes was constructed, and the damage diagnosis performance was studied from the aspects of training convergence speed, diagnosis accuracy, training sample category and network structure. In addition, in order to further verify the effectiveness and practicability of the proposed method, the anti-noise performance under strong noise interference and the generalization performance when processing new samples were studied. The results show that the DRSN model trained with image enhanced samples performs better than the common CNN model in terms of diagnostic accuracy, iterative convergence speed and stability, and shows good robustness and adaptability on different diagnostic objects. The adaptive threshold reduction mechanism of DRSN has excellent anti-noise performance and generalization performance, which makes it more advantageous in the case of strong noise and small sample.

参考文献/References:

［ 1 ］ CASAS J R, APARICIO A C. Structural damage identification from dynamic - test data ［ J ］ . Journal of Structural Engineering, 1994, 120(8): 2437 - 2450.

［ 2 ］冯紫科 , 杨璐 , 柳美玉 , 等 . 基于 BP 神经网络的桁架结构损伤识别研究［ J ］ . 建筑科学与工程学报 ,2024,41(6):41 - 48.

FENG Zike, YANG Lu, LIU Meiyu, et al. Research on damage identification of truss structures based on BP neural network ［ J ］ . Journal of Architecture and Civil Engineering, 2024, 41(6): 41 - 48.

［ 3 ］ HOU R R, XIA Y. Review on the new development of vibration - based damage identification for civil engineering structures: 2010 — 2019 ［ J ］ . Journal of Sound and Vibration, 2021, 491: 115741.

［ 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):159 - 167.

LI Xuesong, MA Hongwei, LIN Yizhou. Structural damage identification based on convolution neural network ［ J ］ . Journal of Vibration and Shock, 2019, 38(1) : 159 - 167.

［ 5 ］ ZHANG Y Q, 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.

［ 6 ］秦世强 , 苏晟 , 杨睿 . 基于多标签卷积神经网络的结构损伤识别［ J ］ . 建筑科学与工程学报 ,2024, 41(3) :108 - 119.

QIN Shiqiang, SU Sheng, YANG Rui. Structural damage identification based on multi - label convolution neural network ［ J ］ . Journal of Architecture and Civil Engineering, 2024, 41(3): 108 - 119.

［ 7 ］骆俊锦 , 王万良 , 王铮 , 等 . 基于时序二维化和卷积特征融合的表面肌电信号分类方法［ J ］ . 模式识别与人工智能 ,2020,33(7):588 - 599.

LUO Junjin, WANG Wanliang, WANG Zheng, et al . Surface electromyography classification method based on temporal two - dimensionalization and convolution feature fusion ［ J ］ . Pattern Recognition and Artificial Intelligence, 2020, 33(7): 588 - 599.

［ 8 ］ DEBAYLE J, HATAMI N, GAVET Y. Classification of time - series images using deep convolutional neural networks ［ C ］ //SPIE. Tenth International Conference on Machine Vision (ICMV 2017). Vienna: SPIE, 2017: 10696 - 23.

［ 9 ］ HUANG J S, CHEN B Q, YAO B, et al. ECG arrhythmia classification using STFT - based spectrogram and convolutional neural network ［ J ］ . IEEE Access, 2019, 7: 92871 - 92880.

［ 10 ］ WANG Z, OATES T. Encoding time series as images for visual inspection and classification using tiled convolutional neural networks ［ C ］ //AAAI. Workshops at the Twenty - ninth conference on artificial intelligence. Austin: AAAI, 2015: 1 - 7.

［ 11 ］ HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition ［ C ］ //IEEE. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016: 770 - 778.

［ 12 ］李书进 , 熊书琪 , 范沛然 , 等 . 考虑残差学习的深层卷积神经网络在结构损伤识别中的应用研究［ J ］ . 工业建筑 ,2022,52(7):192 - 198.

LI Shujin, XIONG Shuqi, FAN Peiran, et al. Application research on deep convolutional neural network considering residual learning in structural damage identification ［ J ］ . Industrial Construction, 2022, 52(7) : 192 - 198.

［ 13 ］ ZHAO M H, ZHONG S S, FU X Y, et al. Deep residual shrinkage networks for fault diagnosis ［ J ］ . IEEE Transactions on Industrial Informatics, 2020, 16(7): 4681 - 4690.

［ 14 ］ LI Y, CHEN H G. Image recognition based on deep residual shrinkage Network ［ C ］ //IEEE. 2021 International Conference on Artificial Intelligence and Electromechanical Automation (AIEA). Guangzhou: IEEE, 2021: 334 - 337.

［ 15 ］ LI J Y, TAN L, ZHOU Y C, et al. Voice - face cross - modal association learning based on deep residual shrinkage network ［ C ］ //IEEE. 2023 IEEE International Conference on Image Processing and Computer Applications (ICIPCA). Changchun: IEEE, 2023: 140 - 145.

［ 16 ］ TONG J Y, TANG S Y, WU Y, et al. A fault diagnosis method of rolling bearing based on improved deep residual shrinkage networks ［ J ］ . Measurement, 2023, 206: 112282.

［ 17 ］周赣 , 华济民 , 李铭钧 , 等 . 基于图转换和混合卷积神经网络的窃电检测方法［ J ］ .2022(19):78 - 86.

ZHOU Gan, HUA Jimin, LI Mingjun, et al. Detection method of stealing electricity based on graph transformation and mixed convolutional neural network ［ J ］ . Automation of Electric Power Systems, 2022(19): 78 - 86.

［ 18 ］ ZHANG C Y, BENGIO S, HARDT M, et al. Understanding deep learning (still) requires rethinking generalization ［ J ］ . Communications of the ACM, 2021, 64(3): 107 - 115.

［ 19 ］ SHORTEN C, KHOSHGOFTAAR T M. A survey on image data augmentation for deep learning ［ J ］ . Journal of Big Data, 2019, 6: 60.

［ 20 ］ ZHAO X Q, ZHANG Y Z. An intelligent diagnosis method of rolling bearing based on multi - scale residual shrinkage convolutional neural network ［ J ］ . Measurement Science and Technology, 2022, 33(8): 085103.

［ 21 ］ HU J, SHEN L, SUN G. Squeeze - and - excitation networks ［ C ］ //IEEE. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 7132 - 7141.

［ 22 ］卞文彬 , 邓艾东 , 刘东川 , 等 . 基于改进深度残差收缩网络的风电机组滚动轴承故障诊断方法［ J ］ . 机械工程学报 ,2023,59(12):202 - 214.

BIAN Wenbin, DENG Aidong, LIU Dongchuan, et al . Fault diagnosis method of wind turbine rolling bearing based on improved deep residual shrinkage network ［ J ］ . Journal of Mechanical Engineering, 2023, 59(12): 202 - 214.

［ 23 ］ DEVKAR R, SHIRAVALE S. A survey on multi - label classification for images ［ J ］ .International Journal of Computer Applications, 2017, 162(8): 39 - 42.

［ 24 ］ RASTOGI R, MORTAZA S. Imbalance multi - label data learning with label specific features ［ J ］ . Neurocomputing, 2022, 513: 395 - 408.

［ 25 ］欧进萍 , 何政 , 吴斌 , 等 . 钢筋混凝土结构基于地震损伤性能的设计［ J ］ . 地震工程与工程振动 ,1999, 19(1) :21 - 30.

OU Jinping, HE Zheng, WU Bin, et al. Seismic damage performance - based design of reinforced concrete structures ［ J ］ . Earthquake Engineering and Engineering Vibration, 1999, 19(1): 21 - 30.

相似文献/References:

[1]李书进,杨繁繁,张远进.基于卷积神经网络和多标签分类的复杂结构损伤诊断[J].建筑科学与工程学报,2025,42(01):101.[doi:10.19815/j.jace.2023.02027]
　LI Shujin,YANG Fanfan,ZHANG Yuanjin.Damage diagnosis of complex structure based on convolution neural network and multi-label classification[J].Journal of Architecture and Civil Engineering,2025,42(01):101.[doi:10.19815/j.jace.2023.02027]

备注/Memo

备注/Memo:
-
更新日期/Last Update: 2026-01-20