LAUSR

Abstract To evaluate the deformation of surrounding rocks of the underground caverns in the Xiangjiaba hydropower station during excavation, a least squares support vector machine (LSSVM) method based on particle swarm optimization (PSO) algorithm is proposed in this study. The PSO algorithm was employed in optimizing the regularization and kernel parameters of the LSSVM. To develop the proposed PSO-LSSVM model, several important factors, such as the geological conditions, location of monitoring instruments, space and time condition before and after measuring, were used as the input parameters, while the displacement of surrounding rocks was the output parameter. Further, the numerical results of the deformations of surrounding rocks were compared with the measured data. The results obtained demonstrate that the proposed PSO-LSSVM model has potential in accurately forecasting the deformation of surrounding rocks of underground caverns subjected to excavation.