摘要
在分析和解剖已有方法的基础上,提出一种利用多元统计学与信息处理技术预测油气空间分布的方法。该方法用马氏距离判别法对信息进行集成,用贝叶斯公式计算已知样本的含油气概率,并由此建立不同马氏距离值下的含油气概率模板,然后采用该模版预测油气资源在空间分布的概率。南堡凹陷应用结果表明,在凹陷西北部的北堡和老爷庙油田(陆地区),预测结果与目前的含油气井分布吻合;对勘探程度较低的凹陷东南部滩海区进行预测,指出了老堡南、南堡南、蛤坨等有利含油气区块。2005年度已钻17口探井结果与预测符合率达81%,证明该预测模型对降低风险、提高勘探成功率有显著效果。
A multivariate statistical method is proposed to predict the geographic locations of undiscovered petroleum accumulations on the basis of improvements of the existing methods. The method involves a data integration using Mahalanobis distance, a probabilistic classification employing Bayesian statistics, and the estimation of probability of hydrocarbon occurrence at untested locations utilizing the established classification model. The application of this method is demonstrated through a case study in the Nanpu Sag. The method was validated by the successful predictions of the oil and gas bearing areas of Beipu and Laoyemiao blocks in the onshore part of the sag in the north. The predicted high potential areas in the Liaopunan, Nanpunan and Hatuo blocks of the offshore part of the Nanpu Sag in the south were confirmed by 17 new exploratory wells in 2005 with an 81% success rate, suggesting that this method is useful for risk reduction in exploration.
出处
《石油勘探与开发》
SCIE
EI
CAS
CSCD
北大核心
2007年第1期113-117,共5页
Petroleum Exploration and Development
基金
全国新一轮油气资源评价攻关项目"常规油气资源评价"
关键词
油气资源
空间分布
风险预测
南堡凹陷
油气勘探
hydrocarbon resources
spatial distribution
risk prediction
Nanpu Sag
petroleum exploration
