近年来在石油工业领域大力推广数据挖掘技术。一方面，随着数据挖掘理论的研究与发展，数据挖掘技术从实验室走向了现实应用，在商业、工程技术等方面获得了广泛的应用，积累了大量的应用经验。另一方面，随着智能信息化的推进，石油工业中积累了大量的生产数据，尤其是实时监控系统和实时数据库的广泛应用，生产数据有了完备的记录，这使得数据挖掘技术的进入成为可能。数据挖掘技术作为一种全新的油井井况研究手段具有很大的优势。 本文以优化石油开采制度，提高生产效率为目标，借助MATLAB这个具有强大数值计算能力和可视化图形设计功能的平台，设计开发了油井数据挖掘系统，为油井井况研究提供技术支持。将MATLAB、数据挖掘技术、油井井况研究三者结合，提出了油井数据挖掘系统设计和开发的三层体系结构，实现了油井数据挖掘系统平台的搭建。在对某采油区53口油井的数据进行分析的基础上，设计了针对油井井况研究的油井数据主成分分析、聚类分析和决策树算法，实现了对油井井况规则、知识的提取。 本文在MATLAB平台下，借助于数据挖掘技术设计了针对油井井况研究的数据挖掘系统，该系统包括数据库信息模块、曲线拟合模块、主成分分析模块、聚类分析模块及分类模块，实现了从生产数据中提取反映油井井况的规则和知识的功能，为决策者优化生产制度、提高生产效率提供了可靠依据。

Data Mining is one of rhe promising information techniques in petroleum industry. On one hand, with the development and research of data mining theory, data mining has found its applications in various areas, from business to entertainment, from science to technology. On the other hand, due to the development of information systems in petroleum industry, there is a huge amount of data accumulated in the petroleum production, especially for the application of real time database, operation data of petroleum industry can be recorded completely in database. And this accumulation of process data makes it possible to employ data mining to improve productivity. Meanwhile, data mining, as a method of petroleum industry analyzing, has a great advantage. This thesis tried to apply data mining in petroleum industry. And this study on petroleum industry data mining system discussed technical approach form a new aspect, that is, it based on MATLAB which has a strong numerical computation and visual graphic design ability, so as to provide technical support for data mining from petroleum industry data. This study combined MATLAB, data mining, and oil condition, put forward the three-tier architecture, realized the construction of system framework, designed Principal Component Analysis, clustering and C4.5 algorithm at length, applied it to the 53 well of Long-er-zeng oil. To analyse the oil conditions, the thesis applies the data mining technique in petroleum industry data, and uses the Curve fitting, Principal Component Analysis, clustering, and classifying to analyse the petroleum industry data, the results well illustrates the conditions of petroleum industry, which provides a basis for decision-makers.