Thetime series analysis and prediction of landslide GNSS monitoring displacement is of great significance to the early warning research of landslide disasters. To improve the accuracy of landslide cumulative displacement prediction, this study proposes a combination of variational mode decomposition (VMD) algorithm, long short term memory (LSTM) network model and grey model (GM) to integrated landslide cumulative displacement prediction model. Based on the variational modal decomposition algorithm, the cumulative displacement of the GNSS landslide is decomposed to obtain the trend displacement and the fluctuation displacement. The LSTM model considering the rainfall influence factor is constructed to predict the fluctuation displacement, and the dynamic GM(1,1) prediction model is established to predict the trend displacement. Decomposition predictions are then superimposed to obtain predicted values. Taking the Bazimen landslide as an example, compared with VMD-BPNN-GM and LSTM models, the experimental results show that the established combined model has the highest prediction accuracy, and the prediction results conform to the change of landslide displacement, which has certain engineering application value in the monitoring of landslide disasters.
Landslide Displacement Prediction Based on VMD-LSTM-GM Model Considering Rainfall
Lect. Notes Electrical Eng.
China Satellite Navigation Conference ; 2022 ; Beijing, China May 22, 2022 - May 25, 2022
2022-05-07
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
GNSS landslide cumulative displacement prediction , Variational mode decomposition , Long short term memory network , Dynamic GM(1,1) Engineering , Communications Engineering, Networks , Signal, Image and Speech Processing , Aerospace Technology and Astronautics , Control, Robotics, Mechatronics , Computer Applications , Geography, general
Statistically-based rainfall thresholds for landslide triggering
British Library Conference Proceedings | 2004
|British Library Online Contents | 2010
|LSTM LSTM-based future threat prediction method and apparatus
Europäisches Patentamt | 2021
|Traffic situation prediction method based on LSTM model
Europäisches Patentamt | 2024
|