基于多尺度地理加权回归与空间异质性分区的滑坡易发性预测方法
By employing multi-scale geographically weighted regression and spatial heterogeneity zoning methods, the problems of low local accuracy and insufficient stability in landslide susceptibility prediction were solved, achieving high-precision and robust landslide susceptibility prediction and enhancing the engineering application value of the prediction results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA HYDROELECTRIC ENGINEERING CONSULTING GROUP CHENGDU RESEARCH HYDROELECTRIC INVESTIGATION DESIGN AND INSTITUTE
- Filing Date
- 2026-05-09
- Publication Date
- 2026-07-17
AI Technical Summary
Existing landslide susceptibility prediction models suffer from low local accuracy, poor area efficiency, and insufficient prediction stability due to their single spatial scale and homogeneous modeling, and thus cannot effectively characterize the spatial heterogeneity of the landslide formation process.
A multi-scale geographically weighted regression model is used to adaptively allocate independent bandwidth to each disaster-prone factor. A spatially continuous regression coefficient surface is generated by Kriging interpolation, and spatial heterogeneity is partitioned based on the regression coefficient surface. Finally, landslide susceptibility is extrapolated by partition, and quality is evaluated by multi-dimensional indicators.
It improves the accuracy and stability of landslide prediction, enhances the identification of local high-risk areas and the engineering application value of prediction results, and ensures the robustness and efficiency of prediction results.
Smart Images
Figure CN122155444B_ABST