[1]李 阳,胡晓东,张庭山,等.数据挖掘技术在录井解释评价中的应用[J].复杂油气藏,2020,13(01):5-9.[doi:10.16181/j.cnki.fzyqc.2020.01.002]
 LI Yang,HU Xiaodong,ZHANG Tingshan,et al.Application of data mining technology in interpretation and evaluation of well-logging[J].Complex Hydrocarbon Reservoirs,2020,13(01):5-9.[doi:10.16181/j.cnki.fzyqc.2020.01.002]
点击复制

数据挖掘技术在录井解释评价中的应用()
分享到:

《复杂油气藏》[ISSN:1674-4667/CN:31-2019/TQ]

卷:
13卷
期数:
2020年01期
页码:
5-9
栏目:
油气勘探
出版日期:
2020-03-25

文章信息/Info

Title:
Application of data mining technology in interpretation and evaluation of well-logging
作者:
李 阳1胡晓东2张庭山1刘 宁1
1.西南石油大学地球科学与技术学院,四川 成都 610500;2.中国石油西南油气田分公司川东北气矿,四川 达州635000
Author(s):
LI Yang1HU Xiaodong2ZHANG Tingshan1LIU Ning1
1. School of Geosciences and Technology, Southwest Petroleum University, Chengdu 610500, China;2. Northeast Sichuan Gasfield, Southwest Oil & Gas Field Company, PetroChina, Dazhou 635000, China
关键词:
数据挖掘分类归纳录井解释数据降维渤海油田
Keywords:
data miningclassificationlogging interpretationdata dimensionality reductionBohai Oilfield
分类号:
TE122
DOI:
10.16181/j.cnki.fzyqc.2020.01.002
文献标志码:
A
摘要:
为了更好地研究渤海油田各个二级构造带的储层流体性质并做出合理准确的解释评价结论,通过整理不同地化录井技术的特征参数及其试油结论,利用数据挖掘中的分类归纳方法建立分类解释模型,包括BP神经网络和支持向量机,此种分类方法快捷准确;同时应用数据降维方法对高维度的录井特征参数进行降维处理,使分类结果最终显示在二维可视化图版上;利用支持向量机绘制解释图版的不同流体性质边界,解决人为绘制的随意性,提高解释图版的实用性。通过使用数据挖掘技术与地化录井有机结合,对提高渤海地区储层油气水层解释精度具有深远的意义。
Abstract:
To better study the reservoir fluid properties of each second-order structural zone in Bohai Oilfield and make the reasonable and accurate conclusions of interpretation and evaluation,it was sorted out the characteristics parameters of different geochemical logging techniques and their testing-conclusions.Using the sorted generalization method in data mining(DM),it was established the classification and interpretation model including BP neural network and support vector machine(SVM),which is fast and accurate.Meanwhile,the dimensionality of high dimensional logging characteristic parameter was reduced by using the data dimension reduction method. Finally, the classification results were displayed on a two-dimensional visualization chart. Using SVM, the boundaries of different fluid properties for interpreting charts were drawn,which can solve the arbitrariness of artificial drawing and improve the practicability of the interpretation chart.Combined with the DM technology and the geochemical logging,it has profound significance for improving the interpretation accuracy of oil, gas and water layers in the Bohai area.

参考文献/References:

[1]LiX,XuJ.TheImprovementofBPArtificialNeuralNetworkAlgorithmandItsApplication[C].2010InternationalConferenceonE-BusinessandE-Government,Guangzhou,2010:2568-2571.
[2]周黄斌,周永华,朱丽娟.基于MATLAB的改进BP神经网络的实现与比较[J].计算技术与自动化,2008(1):28-31.
[3]范磊,张运陶,程正军.基于Matlab的改进BP神经网络及其应用[J].西华师范大学学报(自然科学版),2005(1):70-73.
[4]顾亚祥,丁世飞.支持向量机研究进展[J].计算机科学,2011,38(2):14-17.
[5]祁亨年.支持向量机及其应用研究综述[J].计算机工程,2004,30(10):6-9.
[6]丁世飞,齐丙娟,谭红艳.支持向量机理论与算法研究综述[J].电子科技大学学报,2011,40(1):2-10.
[7]肖艳,姜琦刚,王斌,等.基于ReliefF和PSO混合特征选择的面向对象土地利用分类[J].农业工程学报,2016(4):211-216.
[8]赵宇,黄思明,陈锐.数据分类中的特征选择算法研究[J].中国管理科学,2013,21(6):38-46.
[9]董虎胜.主成分分析与线性判别分析两种数据降维算法的对比研究[J].现代计算机(专业版),2016(29):36-40.
[10]张文盛,刘忠宝.基于Matlab仿真的数据降维实验设计[J].实验技术与管理,2016(9):119-121.
[11]UTKINLV,ZHUKYA.Aone-classclassificationsupportvectormachinemodelbyinterval-valuedtrainingdata[J].Knowledge-BasedSystems,2017,120:43-56.

相似文献/References:

[1]李昌恒,陈德春,向刚,等.基于数据挖掘的蒸汽吞吐井注采参数优化设计[J].复杂油气藏,2011,(02):80.
[2]苗 涛,王凤琴,高利军,等.基于数据挖掘技术的碎屑岩岩性识别方法及应用[J].复杂油气藏,2021,14(01):39.[doi:10.16181/j.cnki.fzyqc.2021.01.008]
 MIAO Tao,WANG Fengqin,GAO Lijun,et al.Lithology identification method for clastic rock based on data mining technology and its application[J].Complex Hydrocarbon Reservoirs,2021,14(01):39.[doi:10.16181/j.cnki.fzyqc.2021.01.008]

备注/Memo

备注/Memo:
收稿日期:2019-04-22;改回日期:2019-11-11。
第一作者简介: 李阳(1988—),硕士学位,从事石油地质勘探及数据挖掘方面工作。E-mail:7891235@qq.com。
基金项目:中海石油(中国)有限公司综合科研项目“细分构造带的录井油气水解释模型及评价方法研究”(编号:ZZKJ-2016-TJ-01)。
更新日期/Last Update: 2020-03-15