A Deep Transfer Learning Network for Structural Condition Identification with Limited Real-World Training Data

被引:32
|
作者
Bao, Nengxin [1 ,2 ]
Zhang, Tong [3 ]
Huang, Ruizhi [1 ,2 ]
Biswal, Suryakanta [4 ]
Su, Jingyong [1 ]
Wang, Ying [1 ,2 ]
机构
[1] Harbin Inst Technol, Shenzhen, Guangdong, Peoples R China
[2] Guangdong Prov Key Lab Intelligent & Resilient Str, Shenzhen, Guangdong, Peoples R China
[3] Peng Cheng Lab, Shenzhen, Guangdong, Peoples R China
[4] Univ Surrey, Guildford, England
来源
STRUCTURAL CONTROL & HEALTH MONITORING | 2023年 / 2023卷
关键词
Condition identification - Identification accuracy - Learning architectures - Learning network - Performance - Pre-training - Real-world - Structural condition - Training data - Transfer learning;
D O I
10.1155/2023/8899806
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Structural condition identification based on monitoring data is important for automatic civil infrastructure asset management. Nevertheless, the monitoring data are almost always insufficient because the real-time monitoring data of a structure only reflect a limited number of structural conditions, while the number of possible structural conditions is infinite. With insufficient monitoring data, the identification performance may significantly degrade. This study aims to tackle this challenge by proposing a deep transfer learning (TL) approach for structural condition identification. It effectively integrates physics-based and data-driven methods by generating various training data based on the calibrated finite element (FE) model, pretraining a deep learning (DL) network, and transferring its embedded knowledge to the real monitoring/testing domain. Its performance is demonstrated in a challenging case, vibration-based condition identification of steel frame structures with bolted connection damage. First, disparate subsets of test data are used as training data, and the identification accuracy of the whole dataset is evaluated. The results demonstrate that the proposed approach can achieve high identification accuracy with limited types of training data, with the identification accuracy increasing up to 8.57%. Second, numerical simulation data are used as training data, and then different TL strategies and different DL architectures are compared on the performance of structural condition identification. The results show that even though the training data are from a different domain and with different types of labels, intrinsic physics can be learned through the pretraining process, and the TL results can be clearly improved, with the identification accuracy increasing from 81.8% to 89.1%. The comparative studies show that SHMnet with three convolutional layers stands out as the pretraining DL architecture, with 21.8% and 25.5% higher identification accuracy values over the other two networks, VGGNet-16 and ResNet-18. The findings of this study advance the potential application of the proposed approach towards expert-level condition identification based on limited real-world training data.
引用
收藏
页数:18
相关论文
共 50 条
  • [11] Configuring a federated network of real-world patient health data for multimodal deep learning prediction of health outcomes*
    Haudenschild, Christian
    Vaickus, Louis
    Levy, Joshua
    37TH ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING, 2022, : 627 - 635
  • [12] Deep learning from “passive feeding” to “selective eating” of real-world data
    Zhongwen Li
    Chong Guo
    Danyao Nie
    Duoru Lin
    Yi Zhu
    Chuan Chen
    Lanqin Zhao
    Xiaohang Wu
    Meimei Dongye
    Fabao Xu
    Chenjin Jin
    Ping Zhang
    Yu Han
    Pisong Yan
    Haotian Lin
    npj Digital Medicine, 3
  • [13] Deep learning from "passive feeding" to "selective eating" of real-world data
    Li, Zhongwen
    Guo, Chong
    Nie, Danyao
    Lin, Duoru
    Zhu, Yi
    Chen, Chuan
    Zhao, Lanqin
    Wu, Xiaohang
    Dongye, Meimei
    Xu, Fabao
    Jin, Chenjin
    Zhang, Ping
    Han, Yu
    Yan, Pisong
    Lin, Haotian
    NPJ DIGITAL MEDICINE, 2020, 3 (01)
  • [14] Battery Identification Based on Real-World Data
    Zhang, Miao
    Miao, Zhixin
    Fan, Lingling
    2017 NORTH AMERICAN POWER SYMPOSIUM (NAPS), 2017,
  • [15] Transfer Learning of Individualized Treatment Rules from Experimental to Real-World Data
    Wu, Lili
    Yang, Shu
    JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS, 2023, 32 (03) : 1036 - 1045
  • [16] The Network Service Platform for Real-World Data
    Kato, Yuka
    Narita, Masahiko
    Akiguchi, Chuzo
    2009 INTERNATIONAL CONFERENCE ON ADVANCED INFORMATION NETWORKING AND APPLICATIONS WORKSHOPS: WAINA, VOLS 1 AND 2, 2009, : 55 - 60
  • [17] Toward Real-World Implementation of Deep Learning for Smartphone-Crowdsourced Pavement Condition Assessment
    Jeong, Jong-Hyun
    Jo, Hongki
    IEEE INTERNET OF THINGS JOURNAL, 2024, 11 (04): : 6328 - 6337
  • [18] Structural damage identification by sparse deep belief network using uncertain and limited data
    Ding, Zhenghao
    Li, Jun
    Hao, Hong
    STRUCTURAL CONTROL & HEALTH MONITORING, 2020, 27 (05):
  • [19] Deep Real-world and Real-time Face Identification System
    Ahmadvand, Pouya
    Ebrahimpour, Reza
    Ahmadvand, Payam
    2019 27TH IRANIAN CONFERENCE ON ELECTRICAL ENGINEERING (ICEE 2019), 2019, : 1989 - 1993
  • [20] Real-World Anomaly Detection Using Deep Learning
    Koppikar, Unnati
    Sujatha, C.
    Patil, Prakashgoud
    Mudenagudi, Uma
    INTELLIGENT COMPUTING AND COMMUNICATION, ICICC 2019, 2020, 1034 : 333 - 342