数据挖掘技术在油田生产决策支持中的研究及应用

A Study and Application of Data Mining Technology to the Decision Support of Oilfield Production

油田生产过程积累了大量的历史数据,从大量沉淀的数据环境中挖掘有用的决策信息,探索出油田生产中的规律性,能够更有效地指导生产,为石油企业提供生产决策支持。文中提出了基于油田生产数据采用数据抽取、转换和加载技术的数据仓库的构建策略以及基于多维数据集的数据挖掘的实施方案,并对包括数据仓库体系结构的设计、数据仓库的构建、多维数据集数据存储模式的优选策略、在线分析处理以及基于分层聚类分析的方法实现数据挖掘等在内的各主要环节进行了系统详细的阐述。最后结合油田生产数据,综合运用数据仓库、联机分析处理和数据挖掘技术构建了一套油田企业生产决策支持系统,并提取和挖掘出了对于油田生产决策支持有用的信息。

Large amount of past data has been accumulated in oilfield production. Mining useful decision information from large amount of past data and exploring the roles of oil production can guide production more effectively and provide decision support for oil corporation. This paper proposes a configuration strategy of the data warehouse of oil production and a multidimensional datasets based data mining implementation scheme, and expounds the design of the architecture of a data warehouse, establishment of a data warehouse, optimum selection strategy of storage schema of multidimensional datasets, OLAP and realization of data mining, etc. in detail. We finally put oilfield production data into practice, and get some useful information for the decision support of oilfield production.