Application of nonlinear time series analysis in slope deformation analysis and forecast

The slope was a nonlinear dissipative dynamic system, which was controlled by the condition of rock mass and was influenced by the terrain, groundwater, earthquake and human projects. The slope deformation took on nonlinear evolution features. In this paper the method of nonlinear time series analysis was discussed and the real slope displacements were used to forecast the coming deformation. By reconstructing the phase space the attractors can be renew for researching original dynamic system. After analyzing the radial basis function, its network model was build for forecasting the deformation and compared with the BP neural network. The results showed that the radial basis function model had well generalization ability. It was much better than BP network in the convergence speed and the local maxima and had the advantages in accuracy and speed training.