The Application of Wavelet Analysis and Support Vector Machine Coupling Model in Displacement Prediction of Landslide

Considering features, such as obvious nonlinear and chaotic characteristics, of the time series of landslide displacement, this paper proposed a wavelet analysis and support vector machine coupling model (WA-SVM) to predict the displacement of a landslide. The monitored time series of displacement firstly is decomposed into several components, including high frequency components and low frequency components, by using the wallet analysis (WA). Subsequently, these components, with different frequencies, are predicted separately by the support vector machine (SVM). The final predicted values can be obtained by superimposing all predicted components. In order to evaluate the predicting performance of this model, this paper applies the WA-BP model to BaiShuihe landslide, located in the Three Gorges Reservoir in China. The result shows that the prediction performance of both models is acceptable and the WA-SVM model seems to be superior the WA-BP model in terms of the main accuracy evaluation indexes.