Displacement prediction for mountain road slopes based on variational mode decomposition and LSTM

Mountain roads serve as vital links connecting remote areas with the outside world. The stability of road slopes in these regions is particularly significant due to complex terrain and variable geological conditions. Accurately predicting slope displacement is crucial for disaster prevention and ensuring traffic safety. This paper proposes a predictive model that combines variational mode decomposition (VMD) and long short-term memory neural network (LSTM) to improve the accuracy and reliability of slope displacement prediction for mountain roads. The original slope displacement data is preprocessed using VMD, decomposing it into trend, periodic, and random modal components. Then, the LSTM model is utilized to predict each modal component. Experimental results show that the proposed VMD-LSTM model outperforms single-use LSTM, XGBoost, and ELM models in terms of prediction accuracy, with RMSE, MAE, and MAPE values of 0.102, 0.098, and 2.854%, respectively, demonstrating the effectiveness and superiority of the model.