基于大数据挖掘技术的页岩气井压裂液产出规律分析  

Analysis of Output Law of Fracturing Fluid in Shale Gas Well Based on Big Data Mining Technology

作　　者：李思辰 张公社[1] 纪国法[1] LI Si-chen;ZHANG Gong-she;JI Guo-fa(Collaborative Innovation Center of Unconventional Oil and Gas Hubei Province,Hubei Key Laboratory of Oil and Gas Drilling and Production Engineering, Yangtze University, Wuhan 430100, China)

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

出　　处：《科学技术与工程》2019年第25期130-134,共5页Science Technology and Engineering

基　　金：国家自然科学基金(51804042);油气资源与勘探技术教育部重点实验室项目(K2018-09)资助

摘　　要：为了准确分析页岩气井压裂液产出规律,通过大数据的筛选和分析对影响页岩气井生产的26个因素进行得分排序,得出影响页岩气气井压裂液产出率的主控因素为地质储量、平均单段砂量、孔隙度、A靶点和B靶点深度。通过主成分分析和多元线性回归建立气井压裂液产出规律预测数学模型,计算表明,压裂液产出率的预测值与实际生产数据相比,预测精确度在90%。In order to accurately analyze the law of fracturing fluid production in shale gas wells,this paper sorts and ranks 26 factors affecting shale gas well production through big data screening and analysis,and obtains the main factors affecting the fracturing fluid yield of shale gas wells. The controlling factors are geological reserves,average single-stage sand volume,porosity,A target point and B target point depth. Through the principal component analysis and multiple linear regression,the mathematical model for predicting the production law of gas well fracturing fluid is established. The calculation shows that the predicted value of fracturing fluid yield is 90% compared with the actual production data.

关 键 词：页岩气气井 压裂液产出率 大数据挖掘 主成分分析 多元线性回归 

分 类 号：TE332[石油与天然气工程—油气田开发工程]