【作者】 高倩；

【导师】 刘建朝；

【作者基本信息】 长安大学 ， 地学信息工程， 2015， 博士

【摘要】 随着数字油田数据建设的深入发展,油田数据对于油田企业的价值越来越大。自1999年大庆油田提出数字油田这15年来,人们都在致力于油田历史数据数字化与油田数字化管理建设,在取得了巨大的成就的同时,由于缺乏顶层设计,应用中又出现了很多的问题。本文以研究数字油田与数字油田多源数据为开端,在数字油田先进理论指导下,以数字油田数据多源性与融合为研究目标,着重研究了数字油田多源数据存在的方式、产生的原因及其特征,建立了数字油田数据融合技术方法体系,同时以陕北薛岔油田为例,对该油田地质研究中的多源数据进行关联、整理与融合,建立了构造模型、沉积相模型和储层模型等,实现了多源油田数据融合方法在油气地质研究与决策中的应用。本次论文研究的主要内容与成果如下:1、充分研究了数字油田数据,将数字油田数据进行了分类,提出油田源数据与计算机管理数据的概念。认为油田源数据是油田企业在勘探、开发、生产与集输过程从源头上采集的数据,定义为“会说话石头的语言”。具有可复用,能够用于对地下油气藏的识别与表征,对油气生产过程智能动态管理等特征。计算机管理数据是指按照计算机原理所管理的数据,其特征是较庞杂,大部分是结构化的数据,基本上是解释、推断和地质研究成果报告。定义为知识体数据。2、在充分研究数字油田数据的基础上,研究了数据的多源性。数据多源性分为油田数据的多源性与计算机管理数据的多源性。油田数据多源性产生的原因是由于油田数据多元采集方式与多学科、多领域、多技术构成的各种专业数据,如地震、测井、岩心测试、产量、温度、压力等。计算机管理数据的多源性产生的原因是由于在数字油田建设过程中多期次建设、多部门主导、多开发商水平不一和多种格式与多种数据库及多种管理信息系统造成的数据不兼容的多源性。3、在研究数字油田数据多源性的基础上,研究了多源数据融合问题。从油田数据应用的本质中得出油田数据融合是数据组织、合并、计算后产生新的数据体,并应用在地质研究上,发现地下地质体、构造与油气藏,建立各种地质模型,是数据转化为信息过程的基本结论。4、油田多源数据融合研究是一个复杂而重要的过程,因此,需要正确的理论与方法论作指导,在此提出了油田数据学的概念,以及建立了数字油田数据融合技术方法体系,以此作为油田数据研究的基本理论和方法。同时针对油田多源数据提取效率低的问题,研发了Datist(数据专家)技术及软件,能比较好地解决了这一难题,构成了油田多源数据融合研究的方法论。5、应用本课题提出的油田多源数据融合的理论、模型和方法,以陕北薛岔油田为例,对油田数据进行了关联整理和融合,并建立了该研究区的构造模型、沉积相模型与油藏储层模型,并对此做了地质研究与分析,从真正意义上研究了油田数据的多源性及实现了数据融合在地质研究中的应用,为大大提高地质研究效率,寻找更多的油气资源,发现剩余油,提高采收率做出贡献。更多还原

【Abstract】 Today, with Digital O ilfield Data Construction developing depth, field data plays more and more important role. Since 1999 Daqing O ilfield proposed Digital O ilfeild, all of oil enterprises strive to make historical data digitalised and digital management constructed, however, go with a lot of achievements there are also a lot of problems because of lacking top-level design.This paper took digital oilfield and its multi- source data research for beginning, under the guidance of advanced digital oilfield theories, highlighted the way of multi-source data existing, the reason and feature of generating. Meanwhile, this paper took a field in shaanxi for example, established relevant tectonic model, sedimentary facies model, reservoir model based on its geological multi-source data neaten and fusion, to complete its applied research in reservoir decision-making. The main contents and results of this thesis are as following:1.After a detailed research in digital oilfield data and had them classified, then put forward “Digital O ilfield Source Data(DIS Data) ” and “Computer Management Data(CM Data)” concepts. This paper argues that DIS Data is ” the language of talkable stone ”, the data which collected from source during exploration, exploitation, production and transportation process, has several features as reusable, Identification and characterization of underground reservoirs, Intelligent dynamic management of oil and gas production process, ect. CM Data referes to the data according to the principle of computer management, Its characteristic is more complex; most are structured data and to be used as explain, inference and geological research reports.2. Adequately studied the digital oilfield data and its polyphyly, then proposed that data polyphyly has different meaning to oilfield source data and computer manageme nt data.O ilfield source data is caused by oil field multivariate data acquisition methods, and all kinds of professional data,which is from multidisciplinary and multi- field, such as Seismic, logging, core test, output, temperature, pressure, etc. However, the cause of computer management of multi-source data is due to many-time, many-sector and many-developer digital oilfield construction, which have different levels and a variety of formats and database and management information systems caused data management gap.3. After a detailed research in digital oilfield data and its polyphyly, we studied multi-source data fusion problems. O ilfield data fusion is a process of a new data volume which is generated after the merger, calculated and organized, applied in geological research, and then, used to find the underground geological body, structure and reservoir, and establish all kinds of geological models.4. Oil field research of multi- source data fusion is a complex and important process which needs correct theories and medthod as guidance,in hence,we put forward O ilfield Data Science concept and knowledge hierarchy as the basic theory in oilfield data research. Meanwhile, we developed Datist Experts technology and software in accordance with the need of oil field multi-source data fusion in the trouble of multi-source data extracting, which obtain good results and constitute the methodology of multi-source data fusion research.5. Applied oilfield multi-source data theory and method, the paper took X O ilfield in Shaanxi for example, completes data mining and fusion, and thus established the structure modle, sedimentary facies model and reservoir model of the study area, and completes geological research and analysis. All of these work embodies real significance of the oilfield data polyphyly and realizes the application of data fusion in geological research, making a contribution to finding the remaining oil, improving recovery and looking for more oil and gas resources.更多还原