摘要
建立准确可靠的三维孔隙度分布模型,对火山岩气藏有效储层的预测和建立合理的开发方案具有重要意义。针对长岭气田火山岩储层特点,利用神经网络技术建立单井孔隙度解释模型,并将目的层火山岩划分为三个火山体,提出用火山岩"体控+相控"的建模策略,采用基于象元的序贯高斯随机模拟方法和协同克立金法,建立三维孔隙度参数模型,其结果符合地质认识和生产实际,并为生产单位所采用。
To build the accurate and reliable three dimensional po-rosity distribution model accurately is very important to predict volcanic rock effective reservoir and built up reasonable development plan.According to the characteristic of the vol-canic reservoir of Changling gas field,the neural network method has been used to build the individual well porosity in-terpretation model,the volcanics were divided into three vol-canical bodies,and further more,the volcanics and lithology controlled modeling strategy has been put forward to build the porosity distribution model,by using sequential gauss simulation method and Co-kriging method.The three di-mensional porosity parameter modeling has been built up.In result,the porosity distribution model has proved to fit the geological precognition and production reality,which has been applied for production.
出处
《石油地质与工程》
CAS
2010年第6期38-41,140,共4页
Petroleum Geology and Engineering
关键词
长岭气田
火山岩
孔隙度建模
神经网络
单井解释模型
Changling gas field
volcanic rock
porosity modeling
neural network
individual well model
