A Robust Deep Learning-Based Damage Identification Approach for SHM Considering Missing Data

被引:6
|
作者
Deng, Fan [1 ,2 ,3 ]
Tao, Xiaoming [1 ,2 ,3 ]
Wei, Pengxiang [1 ,2 ,3 ]
Wei, Shiyin [1 ,2 ,3 ]
机构
[1] Harbin Inst Technol, Key Lab Smart Prevent & Mitigat Civil Engn Disaste, Minist Ind & Informat Technol, Harbin 150090, Peoples R China
[2] Harbin Inst Technol, Minist Educ, Key Lab Struct Dynam Behav & Control, Harbin 150090, Peoples R China
[3] Harbin Inst Technol, Sch Civil Engn, Harbin 150090, Peoples R China
来源
APPLIED SCIENCES-BASEL | 2023年 / 13卷 / 09期
基金
中国国家自然科学基金;
关键词
structural health monitoring; missing data; damage identification; deep learning;
D O I
10.3390/app13095421
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
Data-driven methods have shown promising results in structural health monitoring (SHM) applications. However, most of these approaches rely on the ideal dataset assumption and do not account for missing data, which can significantly impact their real-world performance. Missing data is a frequently encountered issue in time series data, which hinders standardized data mining and downstream tasks such as damage identification and condition assessment. While imputation approaches based on spatiotemporal relations among monitoring data have been proposed to handle this issue, they do not provide additional helpful information for downstream tasks. This paper proposes a robust deep learning-based method that unifies missing data imputation and damage identification tasks into a single framework. The proposed approach is based on a long short-term memory (LSTM) structured autoencoder (AE) framework, and missing data is simulated using the dropout mechanism by randomly dropping the input channels. Reconstruction errors serve as the loss function and damage indicator. The proposed method is validated using the quasi-static response (cable tension) of a cable-stayed bridge released in the 1st IPC-SHM, and results show that missing data imputation and damage identification can be effectively integrated into the proposed unified framework.
引用
收藏
页数:18
相关论文
共 50 条
  • [31] Deep learning-based algorithm for automatic identification and classification of surface damage of agricultural products
    Liu, Weili
    Journal of Biotech Research, 2024, 17 : 138 - 145
  • [32] Deep Learning-Based Building Footprint Extraction With Missing Annotations
    Kang, Jian
    Fernandez-Beltran, Ruben
    Sun, Xian
    Ni, Jingen
    Plaza, Antonio
    IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2022, 19
  • [33] Deep Learning-Based Specific Emitter Identification
    Srinivasulu, N.B.
    Chalamalasetti, Yaswanth
    Ramkumar, Barathram
    Lecture Notes in Networks and Systems, 2023, 554 : 283 - 290
  • [34] Deep learning-based bacterial genus identification
    Khan, Shafiur Rahman
    Khan, Ishrat
    Bag, Md. Abdus Sattar
    Uddin, Machbah
    Hassan, Md. Rakib
    Hassan, Jayedul
    JOURNAL OF ADVANCED VETERINARY AND ANIMAL RESEARCH, 2022, 9 (04) : 573 - 582
  • [35] Development and application of a deep learning-based sparse autoencoder framework for structural damage identification
    Pathirage, Chathurdara Sri Nadith
    Li, Jun
    Li, Ling
    Hao, Hong
    Liu, Wanquan
    Wang, Ruhua
    STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL, 2019, 18 (01): : 103 - 122
  • [36] Deep learning-based segmentation for disease identification
    Mzoughi, Olfa
    Yahiaoui, Itheri
    ECOLOGICAL INFORMATICS, 2023, 75
  • [37] Deep learning-based vehicle event identification
    Yen-Yu Chen
    Jui-Chi Chen
    Zhen-You Lian
    Hsin-You Chiang
    Chung-Lin Huang
    Cheng-Hung Chuang
    Multimedia Tools and Applications, 2024, 83 (41) : 89439 - 89457
  • [38] Deep learning-based recovery method for missing structural temperature data using LSTM network
    Liu, Hao
    Ding, You-Liang
    Zhao, Han-Wei
    Wang, Man-Ya
    Geng, Fang-Fang
    STRUCTURAL MONITORING AND MAINTENANCE, AN INTERNATIONAL JOURNAL, 2020, 7 (02): : 109 - 124
  • [39] DNASimCLR: a contrastive learning-based deep learning approach for gene sequence data classification
    Yang, Minghao
    Wang, Zehua
    Yan, Zizhuo
    Wang, Wenxiang
    Zhu, Qian
    Jin, Changlong
    BMC BIOINFORMATICS, 2024, 25 (01):
  • [40] A Deep Learning-Based Approach for the Identification of a Multi-Parameter BWBN Model
    Li, Zele
    Noori, Mohammad
    Wan, Chunfeng
    Yu, Bo
    Wang, Bochen
    Altabey, Wael A.
    APPLIED SCIENCES-BASEL, 2022, 12 (19):