Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Research on Landslide Displacement Prediction Based on Transformer
AbstractLandslides are a typical geological disaster, often causing significant damage to agriculture, industrial production, human life, and property, sometimes leading to catastrophic outcomes. This study uses the data from the Bazimen landslide in the Three Gorges area, along with hydrometeorological data. Five machine learning models—Transformer, Random Forest (RF), Backpropagation Neural Network (BP), Decision Tree (DT), and Support Vector Regression (SVR)—are employed for landslide displacement prediction. The models use preprocessed data, including rainfall, reservoir levels, and cumulative displacement sequences, as inputs. The output displacement predictions for the next 22 months in a time series autoregressive manner. Comparative analysis of the five models' results reveals that the Transformer model achieves MAE, RMSE, and MAPE values of 0.07, 0.08, and 6.23%, respectively.
Research on Landslide Displacement Prediction Based on Transformer
AbstractLandslides are a typical geological disaster, often causing significant damage to agriculture, industrial production, human life, and property, sometimes leading to catastrophic outcomes. This study uses the data from the Bazimen landslide in the Three Gorges area, along with hydrometeorological data. Five machine learning models—Transformer, Random Forest (RF), Backpropagation Neural Network (BP), Decision Tree (DT), and Support Vector Regression (SVR)—are employed for landslide displacement prediction. The models use preprocessed data, including rainfall, reservoir levels, and cumulative displacement sequences, as inputs. The output displacement predictions for the next 22 months in a time series autoregressive manner. Comparative analysis of the five models' results reveals that the Transformer model achieves MAE, RMSE, and MAPE values of 0.07, 0.08, and 6.23%, respectively.
Research on Landslide Displacement Prediction Based on Transformer
ce papers
Yao, Mingxia (Autor:in) / Tian, Dongdong (Autor:in) / Hu, Guohui (Autor:in) / Kong, Qiang (Autor:in) / Li, Yingxue (Autor:in) / Shu, Jiao (Autor:in)
ce/papers ; 8 ; 613-619
01.03.2025
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Landslide Displacement Prediction of Shuping Landslide Combining PSO and LSSVM Model
| DOAJ | 2023
Multiple neural networks switched prediction for landslide displacement
| Elsevier | 2014