摘要
针对油藏中油气规律分布复杂的地区,在以往油层含油判别分析的基础上,提出了一种基于支持向量机的油藏建模方法。应用已知油层的岩层厚度、泥质含量等6个参数作为训练样本的输入,油气特征作为训练样本的输出,对支持向量机进行训练,从而建立基于支持向量分类机的识别模型,并在此基础上对某油区油气水层分布规律进行了预测。结果显示,采用支持向量机判断的油气分布规律与实际试油结果完全一致,即将支持向量机用于油层油气识别是有效的。
A method of modeling in the oil field system based on the support vector machine （SVM） is proposed, based on knowledge of oilbearing characteristics. The method for recognizing oil/gas/water zone＇is applied to interpret the law of oil and gas distribution. Simulation shows that the law of oil and gas distribution coincides with oil testing results, and SVM is effective for recognizing oil and gas distribution in oil field.
出处
《控制工程》
CSCD
2006年第4期355-357,380,共4页
Control Engineering of China
关键词
支持向量机
油气分布
模式识别
support vector machine （SVM）
oil and gas distribution
pattern recognition
