Displacement prediction model of landslide based on time series and visual simulation of the landslide evolution

The time series of landslide displacement reflects the evolution of landslide, and is monotonously increasing and nonlinear. With ensemble empirical mode decomposition (EEMD) methods the monitored displacement series is decomposed into relatively simple components, then BP neural network was applied to forecasting the evolution of each component, and the individual predictions were summed to get the final result of prediction. 3-dimensional modeling software is adopted to plot the 3-dimensional terrain model of the monitored areas, based on the data collected from these areas. Finally, with the combination of mathematical models for landslide prediction and 3-d terrain model, the visual simulation of landslide evolution is realized in visualizing software.