基于支持向量机的调剖效果预测方法研究  

Prediction of profile control result using support vector machines

作　　者：徐耀东[1] 任允鹏[1] 丁良成[1] 唐晓红[1] 

机构地区：[1]胜利油田分公司地质科学研究院,山东东营257015

出　　处：《断块油气田》2007年第2期50-52,共3页Fault-Block Oil & Gas Field

摘　　要：在统计分析大量已实施的调剖效果基础上,提出了基于支持向量机的调剖效果预测方法。通过分析影响调剖效果因素,运用支持向量机理论建立调剖效果预测模型。检验结果表明,基于支持向量机的调剖效果预测方法准确率较高,科学可行,应用前景广泛。In this paper,a new method of support vector machine(SVM) is put forward based on the statistical analysis of the result of massive implemented profile control and for predicting the result of profile control.It uses the support victor machine theory to represent nonlinear relationship between the result of profile control and its factors.This model studies the case histories.It can predict the fast result of profile control for similar reservoir conditions.Several support victor machine models were established for oil production increment,water production decrease,climb of water cut and so on.The result indicated that the models give the accurate predictions for the novel conditions and the method has the widespread application prospect.

关 键 词：支持向量机 调剖 预测 统计学理论 

分 类 号：O211.67[理学—概率论与数理统计]