Deformation Prediction Model of Small Reservoirs Dams and Bank Slopes Based on PS-InSAR Technique and M-LSTM
Accurate prediction of dam and bank slope deformation of small reservoirs is an important link of reservoir modernization management.This paper proposes a method for predicting the deformation of small reservoir dams and slopes based on PS-InSAR technology and M-LSTM neural network(Multivariate Long Short-Term Memory).Firstly,the PS-InSAR technology was used to obtain the deformation characteristics of four typical small reservoir dams and slopes in Liuyang City.Then,three deformation influencing factors were optimized to establish a deformation prediction model of small reservoir dams and slopes based on M-LSTM.The accuracy of the model was verified.The results show that the PS-InSAR technology has good operability in monitoring the deformation of small reservoir dams and slopes.The M-LSTM model has better prediction performance compared to the LSTM model,with an average coefficient of determi-nation reaching 0.91.The average absolute error and root mean square error are only 0.012 and 0.010,respectively,in-dicating the good applicability of the M-LSTM model in predicting the deformation of small reservoir dams and slopes.
small reservoirPS-InSARM-LSTMdeformation prediction
万方数据
维普
