Landslide Displacement Prediction of Shuping Landslide Combining PSO and LSSVM Model

Predicting the deformation of landslides is significant for landslide early warning. Taking Shuping in Three Gorges Reservoir area (TGRA) as a case, displacement decomposed into two components by time series model (TSM). The least squares support vector machine (LSSVM) optimized particle swarm optimization (PSO) selected to predict prediction based on rainfall and reservoir water level (RWL). Five parameters, including over previous month, months, RWL, change RWL month period half year, are input variables. relationships between five parameters revealed grey correlation analysis. PSO-LSSVM used periodic term (PTD), method applied trend (TTD). With same variables, back propagation (BP) PSO-SVM also developed comparative In model, R2 three monitoring stations larger than 0.98, MAE values RMSE smallest among models. outcomes demonstrate that has high accuracy predicting displacement.