基于特征参数的聚合物驱开发指标组合预测方法  

Combination prediction method of polymer flooding development index based on characteristics parameters

作　　者：赵云飞[1] 孙洪国[1] 张雪玲[1] 周丛丛[1] 冯程程[1] 王志新[1] 周子健[1] 李一宇轩 ZHAO Yunfei;SUN Hongguo;ZHANG Xueling;ZHOU Congcong;FENG Chengcheng;WANG Zhixin;ZHOU Zijian;LI Yiyuxuan(Exploration and Development Research Institute of PetroChina Daqing Oilfield Co Ltd,Daqing 163712,China)

机构地区：[1]中国石油大庆油田有限责任公司勘探开发研究院,黑龙江大庆163712

出　　处：《大庆石油地质与开发》2023年第1期57-63,共7页Petroleum Geology & Oilfield Development in Daqing

基　　金：中国石油天然气集团有限公司“十三五”科技开发基金项目“高-特高含水油田改善水驱效果关键技术”(2019B-1209)。

摘　　要：大庆油田聚合物驱开发对象已转向储层物性更差的二、三类油层,受砂体规模、剩余油分布、开发方式等影响,开发效果差异大,油田开发指标预测不确定性增强。从聚合物驱相渗曲线与区块采出程度变化趋势对应关系出发,通过大量矿场数据统计,明确了聚合物驱阶段采出程度变化规律。同时,为了减少开发方式的影响,以聚合物驱注入站为研究单元,按照不同油层类型,通过机器学习、数据拟合等手段,建立了采油量生长曲线特征参数与聚合物驱阶段采出程度的关联关系,并给出了特征参数的物理意义。针对大庆油田二、三类油层区块,应用大数据分析手段,分别给出了以注入速度为核心的采油量和阶段采出程度预测模型,从而实现驱替全过程开发指标精准预测。应用结果表明,开发指标组合预测方法对4大开发区(萨中、萨北、喇嘛甸、萨南)的预测符合率均在90%以上,能够满足年度及长远开发规划编制需求,可以为二、三类油层聚合物驱开发方案调整提供技术支持。研究成果为油田开发规划编制与开发调整提供了理论依据。The polymer flooding development targets in Daqing Oilfield have shifted to Class 2 and Class 3 reser-voirs with poorer property.The influence of sand body size,remaining oil distribution and development mode causes much difference of development effect and increasing prediction uncertainty of oilfield development index.Based on corresponding relationship between relative permeability curve and block recovery percent(OOIP)trend of polymer flooding,the change law of recovery percent(OOIP)in polymer flooding stage is determined by large amount of pilot data statistics.Meanwhile,in order to reduce the influence of development mode,and with polymer flooding injection station as research unit,the relationship between characteristics parameters of oil production growth curve and recovery percent(OOIP)for different reservoir types in polymer flooding stage is established using machine learning and data fitting,and the physical significance of characteristics parameters is given.In view of Class 2 and Class 3 reservoir blocks of Daqing Oilfield,prediction models of oil production and stage recovery per-cent(OOIP)with injection rate as the core are given by using big data analysis method,so as to achieve accurate prediction of development indexes in the whole process of displacement.The results show that prediction coinci-dence rate of development index combination prediction method for 4 development zones is>90%,which meets the needs of annual and long-term development planning,and provides technical support for adjustment of polymer flooding development plan for Class 2 and Class 3 reservoirs.The research provides theoretical basis for oilfield de-velopment planning and adjustment.

关 键 词：聚合物驱 大数据 特征参数 预测模型 开发指标 

分 类 号：TE357.46[石油与天然气工程—油气田开发工程]