西南石油大学学报(自然科学版) ›› 2007, Vol. 29 ›› Issue (6): 27-30.DOI: 10.3863/j.issn.1000-2634.2007.06.006

• 地质勘探 • 上一篇    下一篇

测试井数据资料的数据清洗技术研究

张允 姚军 王子胜   

  1. 中国石油大学石油工程学院,山东 东营 257061
  • 收稿日期:2007-06-12 修回日期:1900-01-01 出版日期:2007-12-20 发布日期:2007-12-20
  • 通讯作者: 张允

THE TECHNOLOGY OF WELL LOGGING AND WELL TESTING DATA CLEANING

ZHANG Yun YAO Jun WANG Zi-sheng   

  1. School of Petrolum Engineering,China University of Petroleum,Dongying Shandong 257061,China
  • Received:2007-06-12 Revised:1900-01-01 Online:2007-12-20 Published:2007-12-20
  • Contact: ZHANG Yun

摘要:

针对测井和试井资料中存在数据质量的问题,提出了一种基于聚类分析和神经网络预测技术的数据清洗新方法。该方法首先检测测试井数据中存在空缺项的记录数据,对无空缺数据项的记录数据采用模糊聚类分析技术进行数据分类,再对各类数据分别进行蚁群聚类分析和神经网络学习并矫正噪声数据。将该数据清洗方法运用到试井分析中进行检验,取得了良好的效果。为提高测试井数据质量进行正确的解释评价提供了保证。

关键词: 测井数据, 试井数据, 数据清洗, 模糊聚类, 蚁群算法, 神经网络

Abstract:

In terms of the quality problem of well logging and well testing data, the authors of this paper put forward a new data cleaning method based on clustering analysis and nerve network prediction technology with the method, the missing data in well logging data and well testing data are firstly checked up, clustering analysis adopted to classify the data that don't exist missing data, and then ant colony clustering analysis and neutral network study are carried on to correct the wrong data and fill in the missing data. The data cleaning method is proved to have good effects in the application of well testing analysis.

Key words: well logging data, well testing data, data cleaning, fuzzy clustering, ant colony algorithm, neutral network

中图分类号: