基于张量积样条混合算法的计算机建模可视化-《计算机技术与发展》

文章信息/Info

Title:: Modeling and Visualization of Computer Based on Mixed Algorithm of Tensor Product Spline

Author(s):: ZHANG Qing-hong; CHENG Guo-jian; [1]Department of Computer Project,Xi＇an Areotechnology University[2]School of Computer Science,Xi＇an Shiyou University

Keywords:: tensor product spline; Kriging estimates; computer model; reservoir characterization; smoothness transact

摘要:: 为获取精确光滑的地层面图形,首先利用测井数据对其进行克里金插值建模,然后根据数据的疏密性,获取两种网格数据,细化克里金估计网格;稀疏克里金估计网格。为了使其光滑可视化,在克里金估计基础上采用张量积样条算法对细化网格进行光滑处理;对稀疏网格进行y与x方向插值处理。该方法不仅利用克里金估计建模的精确性与Matlab样条函数的封装性及其光滑处理效果,而且采用混编技术;极大地提高了地层面构造的光滑度且不失估计精度,因此该方法在油藏表征与计算机建模中具有一定的实用意义

Abstract:: Firstly used Kriging interpolation to model with log data to obtain smooth and precise ground level graphics,then according to the number of log data,got two kinds of grid datum： dense grids estimated by Kriging,sparse grids estimated by Kriging.In order to smooth visualization,based on the Kriging estimates,adopted the tensor product spline algorithm to smooth the dense grids surface,and interpolate from the direction of x-axis and y-axis for the sparse grid surface to construct the smooth surface.It not only used the encapsulation of Matlab spline functions and accuracy of Kriging compute,but also made use of mixed programming technology;Enhanced the smoothness of stratum surface largely and maintained the estimated accuracy.So the method has practical significance at reservoir characterization and modeling of computer