基于SFLA-Kriging算法的三维地质建模  

Research on 3D geological modeling based on shuffled frog leaping algorithm and Kriging

作　　者：王英博[1,2] 王栋[3] 李仲学[2] 王琳[3] 

机构地区：[1]辽宁工程技术大学创新实践学院,阜新123000 [2]北京科技大学金属矿山高校开采与安全教育部重点实验室,北京100083 [3]辽宁工程技术大学软件学院,葫芦岛125105

出　　处：《系统工程理论与实践》2014年第11期2913-2920,共8页Systems Engineering-Theory & Practice

基　　金：国家科技支撑计划(2013BAH12F00)

摘　　要：Kriging方法可对离散数据点进行内插和外推,近年来在地质、测绘各方面得到了普遍应用,但Kriging方法中变异函数理论模型的选择和实验变异函数参数的设定大部分依赖于地质工作人员的专业技能,具有一定的客观性.引入混合蛙跳算法,对Kriging实验变异函数进行优化,提出SFLA-Kriging算法并结合某露天矿钻孔数据进行三维建模.实验结果标明,SFLA-Kriging算法与传统Kriging方法相比,克服了一般方法的局限性,误差降低27.89%、三维地质模型精度更高.Kriging algorithm can process discrete data with interpolation and extrapolation and it has been applied universally in the field of geology, surveying and mapping etc. But the selection of variogram theory model and the setting of experimental variogram parameters largely depend on the geological staff＇s professional skills which has certain objectivity. Hybrid leapfrog algorithm was introduced to optimize the Kriging experimental variogram and 3D model was built with proposed SFLA-Kriging algorithm and a certain open-pit mine drilling data. Experiment has shown that, SFLA-Kriging algorithm has overcome the limitation of general methods and has smaller error and higher precision of 3D geological model comparing with the traditional Kriging method.

关 键 词：露天矿 KRIGING 混合蛙跳 三维地质 优化算法 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]