%0 Journal Article
%A WU Kai
%A YANG Xue-lian
%A LI Jia
%T Intelligent Optimization of Supporting Parameters for Large Underground Caverns Based on DE-LSSVM
%D 2019
%R 10.11988/ckyyb.20180147
%J Journal of Yangtze River Scientific Research Institute
%P 115-120
%V 36
%N 9
%X To tackle the time-consuming heavy workload in  optimizing the supporting parameters for large caverns, an intelligent optimization method is proposed by combining differential evolution algorithm (DE) and the least squares support vector machine (LSSVM). The learning samples are produced by orthogonal design and FLAC<sup>3D</sup> numerical simulation, and the optimal parameters of LSSVM are determined in global ranges by DE algorithm. Thus, the LSSVM with optimal parameters are used to describe the nonlinear relationship between supporting parameters and evaluation index. The DE algorithm is used again to search for the optimal supporting parameters in global ranges. The present method is applied to optimize the supporting parameters of underground caverns, and results demonstrate that the optimization method is of good application value in optimizing the supporting parameters of large underground caverns.
%U http://ckyyb.crsri.cn/EN/10.11988/ckyyb.20180147