摘要
鄂尔多斯盆地W油田21井区低渗透储集层非均质性较强,注水开发效果不理想。为了精确描述其储集层物性及展布特征,在地质、测井和开发动态等资料的基础上,综合应用相控建模技术,针对鄂尔多斯盆地低渗透油藏采用多种建模方法相结合的手段进行了建模研究,依次建立了研究区的构造、岩相、沉积微相以及微相控制下的储层属性等相关模型。同时对油气储集层相控与非相控建模结果进行了对比,对比结果显示相控属性模型对油藏特征的描述和表征更加合理、准确,应用Petrel地质建模与Eclipse数值模拟的一体化无缝连接,使后续的CO2驱油数值模拟研究也取得了良好的应用效果。这一研究结果直接体现出相控建模技术在鄂尔多斯盆地低渗透油藏的成功应用,也预示着相控建模技术是解决多种沉积微相和非均质性较强的低渗透储集层物性参数相关模拟的有效途径,能清晰地认识和识别砂体连通性及其展布规律,可为低渗透油藏后期开发提供可靠的地质基础。
The W low permeability oilfield with well 21 block in the Ordos Basin is characterized by ultra-low permeability oil reservoir with strong heterogeneity, and the effect of water flooding is not promising. In order to characterize the reservoir heterogeneity and its through-going characteristics, this paper establishes the structural model, lithofacies model, the sedimentary microfacies model, and the microfacies-controlled property model step by step using facies-controlled modeling technologies with multiple stochastic modeling methods based on the geology investigations, petrographic observations, log analysis, and production performances. The comparison between facies-controlled modeling and non-facies-controlled modeling reveals that facies-controlled property modeling resulted in a more reliable and realistic descriptions of the targeted reservoir heterogeneities. An seamless integration of 3D model established by Petrel and Eclipse simulations gets good application in later research on CO2-EOR numerical simulation. The successful application of the facies-controlled property modeling technologies in the Ordos Basin shows that such approach can characterize the properties of the low permeability reservoir with multi sedimentary mirco-facies and strong heterogeneity, and can describe connectivity and distribution feature of sand, providing a reliable petrophysical property model for optimizing the development strategy of ultra-low permeability reservoirs.
出处
《现代地质》
CAS
CSCD
北大核心
2016年第3期655-662,共8页
Geoscience
基金
国家科技重大专项(2011ZX05004-006-004)
关键词
鄂尔多斯盆地
低渗透油藏
相控建模
变差函数
Ordos Basin
low permeability oilfield
facies-controlled modeling
variogram function
