Time: 23rd, November, 2017

Author: School of Mechanics and Construction Engineering

Publisher: JNU News Center

A JNU research team led by Prof. Ma Hongwei has made a breakthrough in big-data methodology for bridge safety inspection, and the relevant scientific paper was published in Computer-Aided Civil and Infrastructure Engineering, an A-class journal ranked first in its field, with an impact factor of 5.79. This paper put forward a new method for using big-data methodology for large bridge inspection for the first time. Lin Yizhou is the first author, Nie Zhenhua the second author and Prof. Ma the project leader.

Pic 1. Convolutional Neural Network

Pic 2. Accuracy of Single Damage Location: Column 1 is calculating method of Convolutional Neural Network; Columns 2, 3 and 4 are other methods

Pic 3. Identify Result of Multiple Damage

Pic 4. Concept of Simple Beam (Third Order Modal) derived from Network Intermediate Layer