基于自编码器无监督学习结构损伤量化检测研究
作者:
作者单位:

1.天津城建大学 计算机与信息工程学院 天津 300384;2.天津城建大学计算机与信息工程学院 天津 300384

中图分类号:

TP212;TN911.72

基金项目:

天津市科技计划项目 23YDTPJC00350

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [18]
  • | |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    结构健康检测指通过实时或周期性监测评估工程结构的健康状态,深度学习方法因能从原始数据中提取高层特征而备受关注。针对实际应用中损伤类别的多样性,缺乏对损伤状态进行定量分析,本文提出了部分跳跃卷积自编码器损伤判断量化方法。本方法使用卷积自编码器处理结构响应,将高维数据降维至低维特征空间,通过重构误差设定损伤指标,以判断健康状态;基于低维特征构建损伤系数,实现结构损伤量化。利用IASC-ASC I和IASC-ASCE II数据集验证了算法在损伤判断和量化方面的有效性。实验结果表明,损伤指标对大部分损伤状态的判定准确率达到100%,个别损伤状态下的准确率为96%,对不同损伤状态的量化均符合预期。

    Abstract:

    Structural health monitoring refers to the evaluation of the health condition of engineering structures through real-time or periodic monitoring. Deep learning methods have gained attention due to their ability to extract high-level features from raw data. However, the diversity of damage types in practical applications and the lack of quantitative analysis for damage states remain challenging. In this paper, a partial skip-connected convolutional autoencoder-based approach for damage assessment and quantification is proposed. This method utilizes a convolutional autoencoder to process structural responses, reducing high-dimensional data to a low-dimensional feature space. A damage index is defined based on reconstruction error to assess health status, while a damage coefficient constructed from the low-dimensional features enables quantitative damage assessment. The effectiveness of the algorithm in damage detection and quantification is validated using the IASC-ASCE Benchmark Structures I and II datasets. Experimental results demonstrate that the damage index achieves 100% accuracy in identifying most damage states, with 96% accuracy in certain specific cases, and that the quantification aligns well with expected values across different damage states.

    参考文献
    [1] BROWNJOHN J M. Structural health monitoring of civil infrastructure [J]. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2007, 365(1851): 589-622.
    [2] TALAEI KHOEI T, OULD SLIMANE H, KAABOUCH N. Deep learning: Systematic review, models, challenges, and research directions [J]. Neural Computing and Applications, 2023, 35(31): 23103-24.
    [3] JIA J, LI Y. Deep learning for structural health monitoring: Data, algorithms, applications, challenges, and trends [J]. Sensors, 2023, 23(21): 8824.
    [4] ZHAO X, WANG L, ZHANG Y, et al. A review of convolutional neural networks in computer vision [J]. Artificial Intelligence Review, 2024, 57(4): 99.
    [5] 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(154-70.
    [6] 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(01): 159-67.Li Xuesong, Ma Hongwei, Lin Yizhou. Structural Damage Identification Based on Convolutional Neural Networks [J]. Journal of Vibration and Shock, 2019, 38(01): 159-167.
    [7] 骆勇鹏, 王林堃, 廖飞宇, et al. 基于一维卷积神经网络的结构损伤识别 [J]. 地震工程与工程振动, 2021, 4): 145-56.Luo Yongpeng, Wang Linkun, Liao Feiyu, et al. Structural Damage Identification Based on One-Dimensional Convolutional Neural Networks [J]. Earthquake Engineering and Engineering Vibration, 2021, 4): 145-156.
    [8] 杨渊, 练继建, 周观根, et al. 基于一维卷积神经网络的钢桁架结构损伤识别 [J]. 天津大学, 天津市钢结构学会 第二十届全国现代结构工程学术研讨会论文集, 2020,Yang Yuan, Lian Jijian, Zhou Guangen, et al. Structural Damage Identification of Steel Truss Structures Based on One-Dimensional Convolutional Neural Networks [J]. Proceedings of the 20th National Symposium on Modern Structural Engineering, Tianjin University, Tianjin Steel Structure Society, 2020.
    [9] POLLASTRO A, TESTA G, BILOTTA A, et al. Semi-supervised detection of structural damage using variational autoencoder and a one-class support vector machine [J]. IEEE Access, 2023,
    [10] WANG Z, CHA Y-J. Unsupervised deep learning approach using a deep auto-encoder with a one-class support vector machine to detect damage [J]. Structural Health Monitoring, 2021, 20(1): 406-25.
    [11] 刘玉驰, 蒋玉峰, 王树青, et al. 基于数据融合及残差卷积自编码器的结构损伤识别方法 [J]. 振动与冲击, 2023, 42(04): 194-203.Liu Yuchi, Jiang Yufeng, Wang Shuqing, et al. Structural Damage Identification Method Based on Data Fusion and Residual Convolutional Autoencoder [J]. Journal of Vibration and Shock, 2023, 42(04): 194-203.
    [12] LI Z, LIN W, ZHANG Y. Real-time drive-by bridge damage detection using deep auto-encoder. Structures 2023; 47: 1167–81 [M].
    [13] MA X, LIN Y, NIE Z, et al. Structural damage identification based on unsupervised feature-extraction via variational auto-encoder [J]. Measurement, 2020, 160(107811.
    [14] WU J, NIE Z. Damage Detection of Beam Bridge Under a Moving Load Using Auto-encoder [J]. Journal of Building Technology, 2021, 3(1): 13-25.
    [15] DANG H V, TRAN-NGOC H, NGUYEN T V, et al. Data-driven structural health monitoring using feature fusion and hybrid deep learning [J]. IEEE Transactions on Automation Science and Engineering, 2020, 18(4): 2087-103.
    [16] FERNANDEZ-NAVAMUEL A, PARDO D, MAGALH?ES F, et al. Bridge damage identification under varying environmental and operational conditions combining Deep Learning and numerical simulations [J]. Mechanical Systems and Signal Processing, 2023, 200(110471.
    [17] JOHNSON E A, LAM H-F, KATAFYGIOTIS L S, et al. Phase I IASC-ASCE structural health monitoring benchmark problem using simulated data [J]. Journal of engineering mechanics, 2004, 130(1): 3-15.
    [18] DYKE S J, BERNAL D, BECK J, et al. Experimental phase II of the structural health monitoring benchmark problem; proceedings of the Proceedings of the 16th ASCE engineering mechanics conference, F, 2003 [C].
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文
分享
文章指标
  • 点击次数:3
  • 下载次数: 0
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2024-09-18
  • 最后修改日期:2024-11-20
  • 录用日期:2024-11-21
文章二维码
×
《国外电子测量技术》
2025年投稿方式有变