储层横向预测的多元地质统计方法

dono

储层横向预测的多元地质统计方法

刘国明  董敏煜（石油大学 北京）

苑书金  孔炜（吉林油田研究院）

Multivariate geologic geologicstatistical method for horizontal reservoir prediction

Liu Guoming, Dong Minyu, Yuan Shujin,Kong Wei

摘要 本文应用多元统计方法，利用地震属性并辅以钻井、测井资料对某地区某层段砂岩的厚度和孔隙率进行了横向预测。

ABSTRACR:The horizontal prediction of thickness and porosity of a sand member in aregion was made by applying multivariate statistical method to seismicattribute, drilling and logging data.

在多元统计方法中，主要应用了有效数计算、逐步回归分析和Krig插值等方法。

Themultivariate statistical method mainly involves significant figure calculation,stepped regressive analysis and Krig interpolation.

有效数可以定量判定哪些地震属性能进行储层预测，从而筛选出地震属性；

Significantfigure determines which seismic attribute may be applied to reservoirprediction, and useful seismic attribute can be thus chosen   properly.

逐步回归分析是利用已筛选后的地震属性计算储层参数，用它可以提高预测效率和精度，

Thestepped regressive analysis uses the chosen seismic attribute to estimatereservoir parameter, by which efficiency and accuracy of prediction can be improved obviously.

但计算结果仍与建模景点处的真值存在误差；

However,the estimation result slightly deviates from real result at the borehole pointwhere model is constructed.

再利用剩余误差的结构性进行Krig插值，可使误差进一步减小。

Thedeviation will be reduced further by doing Krig interpolation according to the structuralproperty of residual deviation.

用本方法对陆相薄互层的计算结果表明，其精度可以达到勘探甚至开发阶段储层描述的要求。

Its computationresult of a land-facies thin interbed shows that the accuracy meets therequirements of exploration or development phase.

主题词 有效数 回归分析  地震属性 储层 横向预测

Subjectheading: significant figure, regressive analysis, seismic attribute, reservoir,horizontal prediction