摘要
塔河油田奥陶系油气藏是大型碳酸盐岩溶洞型油气藏,其储渗空间主要为大小不同的溶洞、裂缝带、溶蚀孔隙等组成,该油气藏具有极强的非均质性,单纯用静态资料来认识这类油气藏是非常困难。文章提出利用生产动态资料和信息进行该类型油气藏研究的新思路,利用人工神经网络技术在处理非线性相关参数预测方面的优势,并以渗流理论为基础,结合试井成果,选用已知油井的产量、油嘴、油压、含水率、气油比、原油密度等6个开发动态参数作为样品输入数据,推导出影响油气藏开发的重要参数(地层系数)与生产信息的关系,建立了人工神经网络预测储层参数的结构模型。通过塔里木盆地塔河油气田实例研究,说明了利用动态信息评价油气藏技术在碳酸盐岩缝洞型油气藏储层预测和非均质性分析等方面具有较高的实用价值。
Ordovician oil and gas pools in Tahe oilfield belong to large karstic carbonate reservoirs, with the reservoir spaces mainly composed of dissolved caverns, fractures and dissolved pores. They have very strong heterogeneity, thus it is very difficult to understand them by simply using static data. A new idea is presented in this paper to study the type of reservoirs by using dynamic production data. Through fully making using of the advantages of artificial neural network technology in prediction of nonlinear relevant parameters and on the basis of filtration theory, the relationships between the major parameters (formation capacity) that can influence reservoir development and production data are derived and a structure model for predicting reservoir parameters with artificial neural network is built by using 6 dynamic production parameters, including the output, choke, tubing pressure, water cut, gas/oil ratio and crude density, as sample input data and in combination with well test results. Case study in Tahe oilfield in the Tarim Basin shows that the dynamic production data are valuable in prediction and heterogeneity analysis of fractured vuggy carbonate reservoirs.
出处
《天然气工业》
EI
CAS
CSCD
北大核心
2006年第8期53-55,共3页
Natural Gas Industry
关键词
油气藏动态
信息
碳酸盐岩油气藏
储集层研究
神经网络
塔河
油气田
reservoir behavior, information, carbonate reservoir, reservoir study, neural network, Tahe oil & gasfield
