基于Kriging与Closest Point融合算法的边坡岩土层界面拟合

Rock and soil layer interface fitting method based on a Kriging and Closest Point coupled algorithm

以先后发生过两次土质滑坡的平朔公司东露天矿边坡为研究背景,针对目前高精度岩土层界面拟合时有效数据不足和人工评价误差较大的现状,充分利用岩土层界面的地质调查点数据,有效扩充钻孔数据范围,为高精度的岩土层界面拟合提供有效数据,并利用基于Closest Point算法的结构分析模型进行Kriging插值计算,实现Kriging算法与Closest Point算法的融合,完成对露天矿边坡各岩土层的三维拟合.现场选取测量点的比对结果表明,基于Kriging与Closest Point融合算法的界面拟合,与传统的人工经验评价方法相比,能大幅提高拟合精度,且简单易行.

Taking the slope at the East Open Pit Mine of Pingshuo Coal Industry Co. where two soil landslides occurred succes- sively as the research background, in consideration of the current situation that valid data lacks and relatively large error exists in expe- rience evaluation, data of geological survey points in interfaces between rock and soil layers were fully used and the drilling data range was effectively expanded to provide valid data for high-precision interface fitting. A Closest Point algorithm-based structural analysis model was applied to conduct Kriging interpolation calculations, which made the two algorithms coupled. Thus a three-dimensional fit- ting model of all rock and soil layers can be completed. By comparing the data of measuring points selected from the field, the accuracy of the interface fitting method is much higher than the traditional method of human experience.