大数据挖掘技术评价致密气藏水平井产能  

Horizontal-Well Productivity Estimation by Big Data Mining in Tight Gas Reservoir

作　　者：孙敬[1,2,3] 刘德华[1,2,3] 董春艳[1,3] 

机构地区：[1]长江大学,湖北武汉430100 [2]非常规油气湖北省协同创新中心,湖北武汉430100 [3]油气钻采工程湖北省重点实验室,湖北武汉430100

出　　处：《特种油气藏》2016年第5期74-77,共4页Special Oil & Gas Reservoirs

基　　金：国家自然科学基金"致密油水平井SRV与基质耦合变质量流动模型研究"(51504038);湖北省自然科学基金"页岩气水平分段压裂井流动机理和产能预测研究"(2015CFC857);中国石油天然气股份有限公司科技创新基金"致密油体积压裂缝网与基质耦合流动机制研究"(2015D-5006-0208)

摘　　要：为准确评价致密气藏压裂水平井的产能,通过分析致密气藏压裂水平井产能的影响因素,采用大数据技术进行数据过滤,给出了影响产能的2个大类9个子类40个因素,并优选出主要因素。运用主成分分析方法、线性回归方法进行影响因素的排序,并建立多因素影响下的产能方程。研究表明,采用大数据挖掘技术建立的产能评价方程预测的气井产量与实际产量相比,精度达到90%。研究内容为大数据技术在致密气藏储集层地质认识、大规模压裂技术、气井动态特征等领域的应用拓宽了思路。The fractured horizontal-well productivity sensitivity in tight gas reservoir is investigated to accurately estimate its productivity. Big data mining is used for data filtering,2 major classes of 9sub-classes containing 40 factors are provided and the key factors are optimized. The influencing factors are sorted by applying principal component analysis and linear regression methods to establish multi-factor productivity equation for fractured horizontal-well. Research shows that the gas production estimation accuracy of productivity equation derived from big data mining is up to 90% by taking actual field gas production as a reference. This research could provide new ideas for the application of big data mining in tight gas reservoir geology,large-scale fracturing technology,gas production performance,etc.

关 键 词：致密气藏 大数据挖掘技术 水平井 压裂气井产能评价 

分 类 号：TE349[石油与天然气工程—油气田开发工程]