<<
[an error occurred while processing this directive] | [an error occurred while processing this directive] >>
Application of Chaotic Time Series Analysis to the Prediction of Tunnel Surrounding Rock Displacement
(1 School of Civil Engineering, Hexi University, Zhangye 734000; 2 Western Research Center of Civil Engineering Disaster Prevention and Mitigation, Ministry of Education, Lanzhou University of Technology, Lanzhou 730050)
Abstract During tunnel excavation, the surrounding rock displacement inevitably will be affected by many factors. Being a macro representation of the internal mechanical phenomenon of surrounding rock, the displacement possesses strong properties of chaotic dynamics. In this paper, the embedding dimension and optimum delay time are reasonably confirmed based on the one-dimensional data of in-situ measured displacement, and the chaotic time series is reconstructed to fully reflect the phase-space of the original dynamic system. Using the chaotic time series as a training sample, a chaotic time series fuzzy neural network model is set up for predicting the surrounding rock displacement. Surrounding rock displacement of the Zaoshugou No.2 tunnel on the new Lanzhou-Chongqing railway is predicted and analyzed, with the research results showing that the new model has a high precision of prediction, fast calculation rates for convergence and advantages regarding real time and stabilization.
Abstract：
During tunnel excavation, the surrounding rock displacement inevitably will be affected by many factors. Being a macro representation of the internal mechanical phenomenon of surrounding rock, the displacement possesses strong properties of chaotic dynamics. In this paper, the embedding dimension and optimum delay time are reasonably confirmed based on the one-dimensional data of in-situ measured displacement, and the chaotic time series is reconstructed to fully reflect the phase-space of the original dynamic system. Using the chaotic time series as a training sample, a chaotic time series fuzzy neural network model is set up for predicting the surrounding rock displacement. Surrounding rock displacement of the Zaoshugou No.2 tunnel on the new Lanzhou-Chongqing railway is predicted and analyzed, with the research results showing that the new model has a high precision of prediction, fast calculation rates for convergence and advantages regarding real time and stabilization.
CHEN Si-Yang-1,
ZHU Yan-Peng-2,
Huang-Li-Hua-1
.Application of Chaotic Time Series Analysis to the Prediction of Tunnel Surrounding Rock Displacement[J] MODERN TUNNELLING TECHNOLOGY, 2015,V52(3): 75-81
2015, Vol. 52 