Deformation Characteristics And Analysis Of Landslide Displacement Prediction Based On VMD-PNN Model

Landslide was one of the most common geological hazards and it often cause catastrophic damage to people.Therefore,the prediction of landslide displacement is an important approach for risk control and prevention of property loss.In this paper,Baishuihe slope,SanXia was taken as a research case.Firstly,the geological data and the deformation were obtained through on-site investigation and ccontinuous monitoring.Based on that,we used Geo Studio to build the physical model of the slope,and simulated the landslide deformation trend under different triggering factor.Secondly,this paper proposes a combined forecasting model based on variational mode decomposition and PNN(Periodic neural network)to predict the displacement of the slope.In addition,we also compared the precision between the prediction results and the actual landslide displacement.The joint model is verified with the deformation trend of the previous physical model,which proves the effectiveness of the joint model.The main work and results of this paper are as follows:(1)By consulting a large number of literatures,the research status of landslide monitoring and displacement prediction is analyzed,and it is pointed out that the existing prediction models have defects such as complicated calculation and poor generalization ability,which leads to the background of the model proposed in this paper.(2)The basic principle and prediction process of the joint prediction model based on variational modal decomposition and periodic neural network are expounded.Theoretically,it is introduced that the model has the characteristics of high prediction accuracy and efficient calculation,and its application in the field of landslide displacement prediction.Advantage.(3)Based on the geological data of the study area,a geological model was established using GeoStudio.Through the numerical simulation of different working conditions and the changes of the seepage field and stability coefficient of the landslide,it is pointed out that the change of reservoir water level and rainfall are the key factors affecting the deformation of the landslide,and the triggering factor of the joint prediction model in the displacement prediction of the Baishuihe landslide is determined.enter.(4)The joint prediction model proposed in this paper is applied to the prediction of Baishuihe landslide.The comparison of the error between the predicted displacement value of the model and the actual displacement value of the Baishuihe landslide shows that the model has high prediction accuracy;The three aspects of accuracy,computational efficiency,and ability to resist missing values are compared with common slope displacement prediction models,and the results show that.The joint prediction model proposed in this paper outperforms other models in these aspects,and can provide a more efficient and robust prediction method for landslide displacement prediction.