Experimental study on damage identification for grid structure based on BP neural network

DOI

10.2991/ifeesm-17.2018.329How to use a DOI?

Keywords

damage identification; modal parameter; BP neural network

Abstract

Aiming at the difficulties of modal concentration and high degree of freedom in the damage identification of the truss structure and the good fault tolerance and robustness of the BP network, based on the theory of the change of the modal parameters of the truss structure before and after the damage, the modal parameters and BP neural Network structure damage identification method. Taking a 6m × 7.5m square pyramidal grid structure as the research object, the importance coefficient of each bar was calculated according to the theory of continuous collapse, and the position of the damaged bar was simulated. Then, the square of the normalized frequency of the structure before and after damage And the combination of normalized vibration mode parameters as damage indicators to train, test and test BP neural network. The results show that this method can well identify the location and extent of damage to the grid structure.Damage identification, including damage judgment, location and degree, is one of the core of SHM [1]. The change of modal parameters before and after the damage can be regarded as the sign of structural damage to diagnose the position and degree of structural damage.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Zhe Xing
AU  - Chun-He Yang
AU  - Bin Yang
PY  - 2018/02
DA  - 2018/02
TI  - Experimental study on damage identification for grid structure based on BP neural network
BT  - Proceedings of the 2017 3rd International Forum on Energy, Environment Science and Materials (IFEESM 2017)
PB  - Atlantis Press
SP  - 1823
EP  - 1826
SN  - 2352-5401
UR  - https://doi.org/10.2991/ifeesm-17.2018.329
DO  - 10.2991/ifeesm-17.2018.329
ID  - Xing2018/02
ER  -