A Sampling-Based Probabilistic Prediction Method of Rockbursts in Deep-buried Tunnels
(1 Zhejiang Provincial Institute of Communications Planning, Design and Research Co., Ltd., Hangzhou 310006; 2 State Key Laboratory of Geomechanics and Geotechnical Engineering, Institute of Rock and Soil Mechanics, Chinese Academy of Sciences, Wuhan 430071)
Abstract：
With the increase of depth of underground works, lots of rock engineering are threatened by rockburst di? sasters. The study of rockburst prediction in deep tunnels is of big significance to controlling disasters during construction and determining construction parameters. However, rockburst events occur randomly and the randomness of rock mechanical parameters caused by the inhomogeneity of actual rock mass is one of the main factors leading to the randomness of rockbursts. A sampling-based probabilistic prediction method of rock bursts in deep tunnels is proposed based on the method combining reliability theory with numerical simulation. Mechanical parameters of rock mass are considered as stochastic variables, and the distribution types and numerical characteristics are obtained based on indoor tests and Bayes method; a group of numerical calculation sample are constructed by using Monte Carlo method of reliability theory; a numerical analysis of sample and a statistical analysis of numerical calculation results are carried out, the probability of rockbursts is obtained. An analysis and prediction of rockbursts occurred in branch tunnel F in the No.3 test tunnel of Jinping II hydropower station is conducted by using the sampling-based probabilistic prediction method. The results show that simulation results are consistent with actual rock? bursts, indicating that this prediction method is convenient and practical and can be taken as a useful references for prediction of rock bursts in the deep tunnels.
2018, Vol. 55 