Deep learning based nondestructive evaluation of reinforcement bars using ground‐penetrating radar and electromagnetic induction data .Computer-Aided Civil and Infrastructure Engineering (2021) Li Xiaofeng, Zhou Feng et al

Deep learning based nondestructive evaluation of reinforcement bars using ground‐penetrating radar and electromagnetic induction data

Computer-Aided Civil and Infrastructure Engineering (2021)

Li Xiaofeng, Zhou Feng et al, School of Mechanical Engineering and Electronic Information

Recently, Associate Professor Zhou Feng’s team from our School published a paper entitled “Deep learning-based nondestructive evaluation of reinforcement bars using ground-penetrating radar and electromagnetic induction data” online on “Computer-Aided Civil and Infrastructure Engineering”, a top journal in the field of civil engineering. Li Xiaofeng, a graduate student, is the first author, associate professor Zhou Feng is the corresponding author, and other co-authors are from School of Civil Engineering, Guangzhou University, School of Earth Sciences, University of Aberdeen, UK, and School of Civil Engineering and Earth Sciences, Delft University of Technology, The Netherlands. According to the Journal Index Report of Clarivate Analytics in 2020, the ranking of “Computer Aided Civil and Infrastructure Engineering” in various fields is as follows: 2/112 (Computer Science, Interdisciplinary Applications), 1/67 (Construction & Building Technology), 1/137 (Engineering Civil), 1/38 (Transportation Science & Technology), with the average annual number of articles published of about 100, and the latest impact factor of 11.775.

Aiming at the low accuracy and large evaluation error in nondestructive testing of reinforcement bars in the acceptance stage of the project, this paper proposes the use of ground-penetrating radar and electromagnetic induction for simultaneous data acquisition and information fusion, and also adopts deep learning-based algorithms for automatic information processing and interpretation of multi-source electromagnetic data. The experimental results show that the method can simultaneously achieve nondestructive evaluation of the protective layer thickness and diameter of reinforcement with 1 mm accuracy without a priori information, which is higher than the existing nondestructive testing methods in terms of both detection accuracy and detection efficiency.

The research results are funded by the National Natural Science Foundation of China.

Paper information:

Title: Deep learning-based nondestructive evaluation of reinforcement bars using ground-penetrating radar and electromagnetic induction data

Authors: Xiaofeng Li, Hai Liu, Feng Zhou, Zhongchang Chen, Iraklis Giannakis, Evert Slob

Source: Computer-Aided Civil and Infrastructure Engineering

DOI: 10.1111/mice.12798

Published online: 27 November 2021