松辽盆地深层火山岩含气储层产能预测  

Gas productivity prediction for deep-seated volcanic reservoir in Songliao Basin

作　　者：李河[1] 王祝文[1] 刘菁华[1] 许延清[1] 

机构地区：[1]吉林大学地球探测科学与技术学院,长春130026

出　　处：《地球物理学进展》2003年第4期729-736,共8页Progress in Geophysics

基　　金：国家自然科学基金"九五"重大项目;大庆石油管理局项目(49894190 42)联合资助.

摘　　要：　松辽盆地深层火山岩是当前大庆地球物理、地质、地球化学研究的主要领域之一,已取得丰硕的成果.火成岩含气储层产能作为一个表示动态特征的参数,是储层评价的重要指标之一.本文讨论了火成岩含气储层的产能与测井响应之间的关系,探讨了根据测井资料应用人工神经网络技术预测火成岩含气储层产能的方法.利用已知气井测试结果和测井资料作为网络的训练样本,根据网络学习训练结果,输入测井资料等静态参数,可预测储集层的产能.根据这种关系采用神经网络技术实现了测井对产能的预测评价,从而为大庆深部火成岩含气储层的开发提供了一定的依据.The deep-seated volcanic reservoir of Songliao Basin is one of the main research areas of the geophysics, geology and geochemistry in Daqing at present. And now it had obtained plentiful and substantial achievements. As a parameter representing the dynamic characteristic, gas reservoir performance is one of main indices for reservoir evaluation. This paper discussed the potential relationship between the productive capacity and the logging response of the deep-seated volcanic gas-bearing formation in Daqing oilfield,and probed the method to predict the productivity of the gas-bearing formation making use of the neural network technique based on the logging data. The gas-bearing formation test results and the logging data were taken as the neural network training samples. Based on the training result the static parameters logging data such as neutron porosity, density etc. was taken as the input parameters of the network, the productive capacity of the gas-bearing formation could be predicted. Hence, it provided a new parameter or tool for the development design of the deep-seated volcanic gas-bearing formation in Daqing Oilfield,and had widen the application range of well-logging in oil/gas exploration and exploitation.

关 键 词：火山岩储层 神经网络 产能预测 

分 类 号：TE1[石油与天然气工程—油气勘探] P631[天文地球—地质矿产勘探]