摘要
It is difficult to predict the time of instability and failure for the progressive landslide whose displacement-time curve is irregular.Therefore,it is difficult to accurately judge its located deformation stage.For this reason,this paper divides the landslide total displacement into the trend displacement and random displacement based on the R/S analysis method,and apply the neural network model and a polynomial fitting to forecast trend displacement and random displacement respectively.The total prediction displacement is obtained by adding the two parts prediction displacement.Taking the typical gradual landslide-Bazimen landslide as an example,this paper validates the above prediction model and compare the results with those of GM(1,1) model.The results show that the prediction error of the above prediction model is less than 2%,while that of the GM(1,1) model is close to 40%.The model is more suitable for predicting the displacement of a gradual landslide.