Displacement Back Analysis Based on GA-BP and PSO-BP Neural Network

DOI

10.2991/coal-16.2016.34How to use a DOI?

Keywords

back analysis of displacements, BP neural network, genetic algorithm, particle swarm optimization

Abstract

In order to study the back analysis accuracy of different algorithms of neural network for back analysis of tunnels surrounding rock. Firstly, the main parameters influencing the deformation of tunnels surrounding rock are analyzed by orthogonal test. Secondly, a FLAC3D numerical model of the tunnels was established based on the field conditions for getting learning samples of the neural work. Thirdly, the BP neural network is optimized by genetic algorithm, particle swarm optimization and normalization method, respectively. At last, the back analysis of the tunnels displacement is carried out, and the mechanical parameters of the surrounding rock are forecasted for comparing accuracy of the three optimized methods. The results show that it is faster and precise to use artificial neural network to inverse mechanical parameters of the tunnels surrounding rock.

Copyright

© 2016, 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  - Gu Dongdong
AU  - Tan Yunliang
PY  - 2016/10
DA  - 2016/10
TI  - Displacement Back Analysis Based on GA-BP and PSO-BP Neural Network
BT  - Proceedings of the 8th Russian-Chinese Symposium “Coal in the 21st Century: Mining, Processing, Safety"
PB  - Atlantis Press
SP  - 168
EP  - 173
SN  - 2352-5401
UR  - https://doi.org/10.2991/coal-16.2016.34
DO  - 10.2991/coal-16.2016.34
ID  - Dongdong2016/10
ER  -