STUDY OF PLS-LCF MODEL FOR ROCK BURST PREDICTION BASED ON PARTICLE SWARM OPTIMIZATION
YAN Zhixin1,2,HE Xiang2,GONG Bin2
(1. Key Laboratory of Mechanics on Disaster and Environment in Western China,Ministry of Education,Lanzhou University, Lanzhou,Gansu 730000,China;2. School of Civil Engineering and Mechanics,Lanzhou University,
Lanzhou,Gansu 730000,China)
Abstract:The prediction of rock burst is a complex systemic problem. It is necessary to consider the various factors affecting the rock burst comprehensively. But when variables to be considered are very large,the multicollinearity among variables will affect the objectivity of the analysis. In order to eliminate the adverse effect and effectively predict rock burst,the affecting factors for rock burst are analyzed by means of partial least-squares regression(PLSR). A new synthesis variable with better interpretation to the dependent variable is extracted and it can preferably overcome the multicollinearity among variables. The nonlinear relationship between composition values and rock burst grades is established according to the logistic curve function(LCF). The particle swarm optimization(PSO) with the global optimization is used to optimize the parameters of LCF. So far,a PLS-LCF prediction model of rock burst based on PSO is built. The test results of the model show a very good precision. The evaluation results obtained by applying the developed model to practical engineering are well consistent with the practical situation,which indicates that the model is feasible and effective for rock burst prediction.
言志信1,2,贺 香2,龚 斌2. 基于粒子群优化的PLS-LCF岩爆灾害预测模型研究[J]. 岩石力学与工程学报, 2013, 32(s2): 3180-3186.
YAN Zhixin1,2,HE Xiang2,GONG Bin2. STUDY OF PLS-LCF MODEL FOR ROCK BURST PREDICTION BASED ON PARTICLE SWARM OPTIMIZATION. , 2013, 32(s2): 3180-3186.
郭 雷,李夕兵,岩小明,等. 基于BP网络理论的岩爆预测方法[J]. 工业安全与环保,2005,31(10):32-34.(GUO Lei,LI Xibing,YAN Xiaoming,et al. Rock burst prediction methods based on BP network theory[J]. Industrial Safety and Environmental Protection,2005,31(10):32-34.(in Chinese))
Your browse does not support frame!