Abstract
The volatility of the cumulative displacement of landslides is related to the influence of external factors. To improve the prediction of nonlinear changes in landslide displacement caused by external influences, a new combined forecasting model of landslide displacement has been proposed. Variational modal decomposition (VMD) was used to obtain the trend and fluctuation sequences of the original sequence of landslide displacement. First, we established a stacked long short time memory (LSTM) network model and introduced rainfall and reservoir water levels as influencing factors to predict the fluctuation sequence; next, we used a threshold autoregressive (TAR) model to predict the trend sequence, following which the trend and fluctuation prediction sequence were superimposed to obtain the cumulative predicted displacement of the landslide. Finally, the VMD-stacked LSTM-TAR combination model based on the variational modal decomposition, stacked long short time memory network, and a threshold autoregressive model was built. Taking the landslide of Baishuihe in the Three Gorges Reservoir area as an example, through comparison with the prediction results of the VMD-recurrent neural network-TAR, VMD-back propagation neural network-TAR, and VMD-LSTM-TAR, the proposed combined prediction model was noted to have high accuracy, and it provided a novel approach for the prediction of volatile landslide displacement.
Highlights
There are many landslide disasters, wide coverage areas, and frequent activities globally which cause serious economic losses and casualties every year
By focusing on landslides affected by rainfall and reservoir water levels in this study, a combined prediction method of cumulative landslide displacement based on Variational modal decomposition (VMD), stacked long short time memory (LSTM) network models, and threshold autoregressive (TAR) models was proposed, taking the Baishuihe landslide as an example
The combined VMD-stacked LSTM-TAR prediction model proposed in this study processes and predicts data based on the VMD method, the stacked LSTM network model, and the TAR model
Summary
There are many landslide disasters, wide coverage areas, and frequent activities globally which cause serious economic losses and casualties every year. In addition to a linear natural change trend, the cumulative displacement of landslides has a sharp increase This seasonal change, which may be caused by changes in rainfall and reservoir water levels, is difficult to predict. By focusing on landslides affected by rainfall and reservoir water levels in this study, a combined prediction method of cumulative landslide displacement based on VMD, stacked LSTM network models, and TAR models was proposed, taking the Baishuihe landslide as an example. In Zigui County in the Three Gorges Reservoir Area, using the time series decomposition method based on VMD to decompose the cumulative landslide displacement, we built a stacked LSTM network and introduced rainfall and reservoir water level influencing factors to establish a prediction model for the cumulative displacement fluctuation term of the landslide. The TAR model was used to predict the trend item, and the validity of the combined VMD-LSTM-TAR landslide displacement prediction model was demonstrated through the model prediction accuracy verification and by comparison with other models
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
