摘要
将支持向量机和变异函数结合,提出一种新的空间插值方法,改进支持向量机法。该方法将变异函数作为支持向量机目标函数的约束条件,利用了变异函数的空间相关结构重建能力,保留了支持向量机较强的非线性回归能力。为了评价该方法的插值精度,将它和4种已存在的加权距离反比法、径向基函数法、克里金法和支持向量机法等空间插值方法应用于测井的地温场数据、煤层厚度数据以及测井声波时差曲线的插值重构,结果表明,改进支持向量机算法相对其他插值重构算法具有较高的插值精度和较好的相似度,能够很好地实现对空间区域变量的插值重构。
Based on the combination of support vector machine and the semivariogram, a spatial interpolation method, named improved support vector machine, is proposed in the paper. Considering the semivariogram as the constraint of the objective function of support vector machine can not only use the spatial correlation structure reconstruction ability of variation, but also retain the strong nonlinear regression ability of support vector machine. To evaluate the interpolation accuracy of the proposed method, this method as well as the four existing methods, such as inverse distance weight, radius function, Kriging and support vector machine, are then used in the interpolation reconstruction of log geothermal field data, coal seam thickness data and log acoustic time curve. The results indicate that the improved support vector machine algorithm has higher accuracy and better similarity, which can achieve a good reconstruction of spatial variables, compared to other interpolation algorithms.
出处
《测井技术》
CAS
CSCD
北大核心
2012年第6期575-580,共6页
Well Logging Technology
基金
国家高技术发展计划863项目(2006AA06A109-1)
国家高技术发展计划863项目(2006AA060105)
中国石化重大科技攻关项目(JP04014)
山东省自然科学基金(ZR2009FL029)资助
关键词
测井曲线
支持向量机
变异函数
空间变量
重构
插值
log curve, support vector machine, semivariogram, spatial variable, reconstruction, interpolation
