摘要
对控制油田产油潜力各种影响因素进行了系统的分析,选择地质构造、储层孔隙度、储层渗透率和原油性质4个因子作为控制油田产油潜力的主控因素。应用地理信息系统(GIS)的空间分析和处理操作功能,构建了各个主控因素的子专题图层,并建立了主控因素与每米采油指数间的非线性人工神经网络(ANN)分析模型,最终提出了评价产油潜力的GIS与ANN耦合模型。应用该评价模型对埕北30潜山油藏的产油潜力进行了评价,并应用灵敏度分析方法对该地区各个主控因素的灵敏度进行了系统分析,有效地解决了人工神经网络难以通过权重系数矩阵来判定各个影响因子影响程度的难题。
The factors for controlling oil productivity of oilfield were analyzed. The geological structure, permeability and porosity of reservoir bed and property of crude oil were taken as the dominant factors for oil productivity. The subject maps of every dominant factor were built using the geography information system (GIS). A nonlinear artificial neural network (ANN) model for analyses of dominant factors and oil-production index per meter was constructed. A coupled model of GIS and ANN for evaluating oil productivity was also established. The oil productivity in Chengbei-30 buried hill reservoir was assessed by the coupled model effectively. The sensitivity of every dominant factor in this area was analyzed with the sensitivity analysis method. In this way, the weight coefficient matrix of ANN can be used to determine the influence degree of all dominant factors.
出处
《石油大学学报（自然科学版）》
EI
CSCD
北大核心
2004年第5期18-22,共5页
Journal of the University of Petroleum,China(Edition of Natural Science)
基金
教育部跨世纪优秀人才基金资助(2000 3)
教育部青年骨干教师基金资助(2000 65)
