This dataset was developed under the National Key R&D Program project "Stability Analysis and Early Warning Technology for Loess Slopes Under Extreme Rainfall." It employs a self-developed Visco-Elastic-Visco-Plastic (VE-VP) constitutive model to conduct systematic numerical simulations via the finite element software CODE_BRIGHT, generating standardized data for machine learning-based prediction of loess landslide slip lines. By varying parameter combinations including slope height, slope angle, soil strength, matric suction, rainfall intensity, and duration, the dataset simulates the mechanical-hydraulic coupled responses of slopes under various conditions. The primary output consists of displacement contour maps that visually depict slip line distribution. Data is named using a "parameter-coding" system that clearly identifies each simulation scenario. Characterized by its advanced coupled mechanism, comprehensive parameter coverage, and high visual clarity of results, this dataset is suitable for research on loess landslide mechanisms, the development of early warning models for landslides under extreme rainfall, and intelligent prediction of slip lines based on machine learning.
| # | number | name | type |
| 1 | 41971293 | Snow accumulation processes in the permafrost region of the Tibetan Plateau and snow parameters inversion using multi-source remote sensing. | National Natural Science Foundation of China |
| 2 | 2022YFC3003404 |
This work is licensed under
CC BY 4.0 (Creative Commons Attribution 4.0 International License).
CSTR:11738.11.ncdc.loess.db7396.2026
10.12072/ncdc.loess.db7396.2026
