Cross-domain structural damage identification using transfer learning strategy

doi:10.1016/j.engstruct.2024.118171

2024-07-01

PRE

刘泳 (Yang Liu)

Sheng-En Fang *

DOI：10.1016/j.engstruct.2024.118171

原文链接

原文求助

摘要

En 中文

In damage identification of civil structures, training deep neural networks (DNNs) often requires a large volume of annotated training data. However, artificial damage to real-world structures is always forbidden, resulting in insufficient data for training. Transfer learning provides a solution that allows structural damage information in a data-rich domain to be transferred and shared as the prior knowledge to a data-scarce domain, thereby indirectly augmenting available training data in the latter domain. To this end, a multi-task transfer learning strategy has been adopted for the purpose of achieving cross-domain damage identification between a plane frame in the source domain and a three-dimensional frame in the target domain. The strategy can share the learning knowledge from the domain with sufficient training data to the target domain with insufficient data. Thereby, the DNN (named DNN#2) in the target domain can be trained on a small amount of training data. Owing to their ability in data feature extraction, stacked auto-encoders (SAEs) are used to construct the desirable DNN#1 and DNN#2 corresponding to different damage identification tasks, named Task#1 and Task#2, in the two domains. The two DNNs share the hidden layers in order to share damage feature information between the source and target domains. The training datasets of the two domains are first used to jointly pre-train the auto-encoders' parameters in an unsupervised learning way. Afterwards, a supervised fine-tuning step is carried out to retraining the entire SAEs for better performance. By these means, Task#2 receives some prior knowledge from Task#1, and thus, it is accomplished on limited training data when Task#1 is synchronously achieved. The analysis results demonstrate that the proposed multi-task learning strategy requires only a single training session to simultaneously realize the cross-domain damage identification of two different steel frames.

Keyword:

Cross-domain damage identification

Transfer learning

Multi-task learning

Fine-tuning

Frame structure

期刊

Engineering Structures

IF:

6.4

论文数:

2.1W

被引数:

8.7W

机构

fuzhou university

学者数:

3.3W

论文数: 2.1W

被引数: 31

引用论文

Damage identification by response surface based model updating using D-optimal design

MECHANICAL SYSTEMS AND SIGNAL PROCESSING

IF8.9

2011-02-01

125

PREAI

Fang, Sheng-En; Perera, Ricardo

A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications 关于处理数据稀缺的深度学习工具的调查: 定义，挑战，解决方案，技巧和应用

JOURNAL OF BIG DATA

IF6.4

2023-04-14

196

OAAI

Alzubaidi, Laith; Bai, Jinshuai; Al-Sabaawi, Aiman; Santamaria, Jose; Albahri, A. S.; Al-dabbagh, Bashar Sami Nayyef; Fadhel, Mohammed. A. A.; Manoufali, Mohamed; Zhang, Jinglan; Al-Timemy, Ali. H. H.; Duan, Ye; Abdullah, Amjed; Farhan, Laith; Lu, Yi; Gupta, Ashish; Albu, Felix; Abbosh, Amin; Gu, Yuantong

Autoencoder-based representation learning and its application in intelligent fault diagnosis: A review 基于自动编码器的表示学习及其在智能故障诊断中的应用: 综述

MEASUREMENT

IF5.6

2022-02-01

130

PREAI

Yang, Zheng; Xu, Binbin; Luo, Wei; Chen, Fei

Analysis and Prediction of Drug Transfer into Human Milk Taking into Consideration Secretion and Reuptake Clearances across the Mammary Epithelia

Drug Metabolism and Disposition

IF0

2011-12-01

PREAI