【作者】 张伟；

【导师】 林承焰；

【作者基本信息】 中国石油大学 ， 地质资源与地质工程， 2010， 博士

【摘要】 当前油藏地质模型已经受到国内外各大石油公司、科研院校的广泛重视,紧紧围绕降低地质模型的不确定性、提高地质模型的精度开展了大量研究工作,主要有相控储层建模、等时建模、四维储层建模、确定性和随机建模相结合、多点地质统计学等方法,本文在前人的基础上,提出使用地质模型跟踪方法来提高地质模型的预测精度。目前,为了进一步提高储层地质模型的精度,地质约束条件被广泛使用。然而,人们往往是孤立地建立各个时期的油藏地质模型,没有对油藏地质模型进行系统的跟踪研究,这使得大量新的反馈信息没有用于提高油藏地质模型的精度。油田处于不同的开发阶段所反馈回来的信息是不同的,其地质模型动态跟踪的特点和方法也不完全相同。本论文分油田的不同开发阶段提出了具有针对性的地质模型跟踪方法,包括基于随钻资料的地质模型动态实时跟踪方法、油田开发前期油藏地质模型跟踪方法、油田开发中后期地质模型综合一体化方法等。从狭义上讲,基于随钻资料的地质模型动态实时跟踪方法才是真正意义上的实时跟踪,其他地质模型跟踪方法则不能算是实时跟踪。为了实现地质模型动态实时跟踪,本论文使用Matlab软件(平台)自主设计并实现了GMUS(Geological Model Updating System)软件,基本实现了井资料加载、随钻数据采集、随钻曲线显示、随机模拟、地质模型自动更新等功能。将基于随钻资料的地质模型动态实时跟踪方法应用于埕海油田一区,依据Zh8Es-h3、Zh8Es-h4、Zh8Es-h5等井的随钻资料进行地质模型动态实时跟踪,钻井实践表明由于实时提供了可靠的地质模型,随钻井钻遇油层能力明显提高。将油田开发前期油藏地质模型跟踪方法应用于埕海油田二区,对埕海油田二区进行了四年的地质模型跟踪,查明了影响地质模型精度的因素主要集中在构造、变差函数、网格大小等方面,模型重建后砂体厚度预测相对误差减小了12.3%。将油田开发中后期地质模型动态跟踪方法应用于葡北四断块79-53井组,使用EnKF方法建立了动态地质模型,在此基础上建立了79-53井组一体化地质模型,一体化地质模型更加合理、可靠。动态地质模型和一体化地质模型基本再现了地下储层渗透率的变化,这对于深入分析和认识葡I组油层内部存在的窜流通道具有重要的意义。目前,基于随钻资料的地质模型动态实时跟踪方法已经在油田现场中得到应用,并取得了良好的应用效果。油藏地质模型动态跟踪方法可以应用于油田的不同开发阶段,可以提高油藏地质模型的精度和可靠性,使所建油藏地质模型不断逼近于地下真实情况,这对大港、大庆油田以外的其它类似油田也有一定的理论意义和现实价值。更多还原

【Abstract】 Nowadays, the reservoir geological model has earned widespread respect from the oil companies and research institutions both at home and abroad, which have carried out lots of studies for the purpose of reducing the uncertainty and improving the accuracy of the geological model. The method mainly includes facies-controlled reservoir modeling, isochronous modeling, four-dimensional reservoir modeling, the combination of deterministic and stochastic modeling, multi-point geostatistics and so on. Based on the previous study, tracking methods is presented in this article to improve the accurate predictions of geological model.The geological constraints are widely used now in order to further improve the accuracy of the reservoir geological model. However, periods of reservoir geological models are always set up seperately without systemic tracking research, causing numerous new feedback information not being used to improve the model’s accuracy. The feedback information is different in different development stages, which makes the model tracking methods and characteristics vary at the same time. The paper proposed targeted geological modeling tracking methods for each field stage, including the geological model dynamically real-time tracking method on the basis of logging while drilling, the geological model tracking method on the early development stage, and geological model integrated method on the middle and late development stage. In a narrow sense, the method based on logging while drilling is real-time tracking, while the others are not. For implementing dynamically real-time tracking, a new software--GMUS (Geological Model Updating System) is designed by using the Matlab software, which can realize the well data loading, data acquisition while drilling, drilling curve view, stochastic simulation, automatic update, etc.The dynamically real-time method based on logging while drilling is applied on the block one of Chenghai field and carried on the Zh8Es-h3, Zh8Es-h4, and Zh8Es-h5 well. Drilling practice showed that the drilling catching rate was obviously improved due to the accurate model. It took four years to track the Chenghai oilfield block two after applying the tracking methods on the early development stage and got the conclusion that the main factors on influencing the model accuracy were structure, variation function, grid size, etc. The sand thickness prediction in the restructured model had a decreased relative error 12.3 %. A more reliable integrated model was set up with the tracking methods on the middle and late development stage used in the 79-53 well group in PubeiⅣblock and the EnKF method in the dynamic modeling. Dynamic geologic model and integrated geologic model recreated the permeability change of the subsurface reservoir, which had important significance on in-depth analysis and understanding on the channeling path existed in the PuⅠoil group.The geological model dynamically real-time track has been applied in the oilfield and made the best results now. Reservoir geological model dynamic tracking method can be applied to various stages of field development to increase the precision and reliability, and get the most represent of the real situation underground. This research is of great significance not only in the peripheral oilfields in Dagang and Daqing, but also in the similar oilfields in the other areas.更多还原