基于信息量与Logistic模型的矿区滑坡灾害易发性评价

Susceptibility Evaluation of Landslide Hazard in Mining Area Based on Information Content and Logistic Model

针对矿区长期煤矿开采引起的滑坡灾害频发问题,快速高效地模拟和评价矿致滑坡灾害易发性是实现采矿地区科学防灾减灾的关键。基于此,本文应用信息量与Logistic回归模型结合多源高分辨率光学遥感数据等,选取相对高差、坡度、坡向、距断层距离、NDVI、距采空区距离6个滑坡影响因子来评价采煤矿区滑坡灾害易发性。结果表明:(1)信息量与Logistic回归模型耦合的综合预测准确率为96%,信息量模型滑坡预测准确率为95%,实验结果表明耦合模型的预测精度优于单一信息量评价模型,评价模型的合理性和预测精度皆符合检验要求;(2)研究结果也表明了采用信息量+Logistic回归模型耦合能较为客观准确、快速高效地评价地下采矿引起的滑坡灾害易发范围,评价结果可为类似地区高效快速划定滑坡灾害易发区间提供技术支撑。

According to the frequent occurrence of landslides caused by long-term coal mining in mining areas,the key to scientific disaster prevention and mitigation in mining areas is to quickly and efficiently simulate and evaluate the susceptibility of landslides caused by mining.Based on this,this paper applies information content and Logistic regression model combined with multi-source high-resolution optical remote sensing data,and selects six landslide influencing factors,including relative height difference,slope,slope aspect,distance to fault,NDVI and distance to goaf,to evaluate the susceptibility of landslide disaster in coal mining area.The results show that:(1)The comprehensive prediction accuracy of the coupling of information content and Logistic regression model is 96%,and the landslide prediction accuracy of the information content model is 95%.The experimental results show that the prediction accuracy of the coupling model is better than that of the single information content evaluation model,and the rationality and prediction accuracy of the evaluation model meet the test requirements;(2)The research results also show that the coupling of information content and Logistic regression model can objectively,accurately,quickly and efficiently evaluate the landslide disaster prone area caused by underground mining,and the evaluation results can provide technical support for the efficient and rapid delineation of landslide disaster prone area in similar areas.