Time Forecast And Prediction Of Landslide Based On Wavelet And Grey Model

China is one of the countries which have the most serious landslide disaster in the world。Landslide disaster can always cause serious economic losses, which threatens the safety ofpeople’s life.According to the statistics, about70cities and460counties in our country havebeen threatened and damaged by the landslide hazard. As a result, the economic usuallysuffers from a loss between at least1.5billion and2.3billion every year. As the naturaldisasters and human engineering activities increase recently, the landslides also increaseaccordingly.If we know that the landslide occurred at certain time and take actions, then thelandslide hazard can be reduced at most, besides we can reduce unnecessary casualties andeconomic losses.Consequently, research of forecasting Landslide-time is significant.Based onthe purpose above, in this paper, for the problems in the prediction of landslide, we will makea detailed research. According to the introduction and study of wavelet theory and Greymodel, wavelet noise reduction and Grey model can be applied to landslide prediction.Themain research contents are as follows:（1）A succinct introduction of the landslide-time prediction and development tendency,including the landslide monitoring technology and landslide prediction model.（2）Before the Time prediction, In order to improve prediction accuracy, it is necessaryto process observed displacement time data and remove noise. In this paper, the technique ofwavelet is applied to remove the noise in the observed landslide displacement data.Wavelettheory, wavelet function and the selection rules of wavelet threshold are proposed before,Then conduct noise reduction for landslide data by different threshold values, besides,through the signal to noise ratio to compare the result of noise reduction.（3）Proposing Grey model succinctly, programming by MATLAB and using Greymodel to estimate the time of the noise reductive data, then analyzing the results.（4）Apply the noise reduction method for wavelet threshold and Grey model to the newbeach landslide, conduct noise reduction for the observed data and estimate the time, thencompare the results to the true value.