[1]郭 晨,陈 龙.基于机器学习方法的数字岩芯电导率预测[J].计算机技术与发展,2020,30(07):100-103.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 022]
　GUO Chen,CHEN Long.Prediction of Digital Core Electrical Conductivity Using Machine Learning Method[J].,2020,30(07):100-103.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 022]
点击复制

基于机器学习方法的数字岩芯电导率预测()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
30
期数:
2020年07期
页码:
100-103
栏目:
应用开发研究
出版日期:
2020-07-10

文章信息/Info

Title:
Prediction of Digital Core Electrical Conductivity Using Machine Learning Method
文章编号:
1673-629X(2020)07-0100-04
作者:
郭 晨; 陈 龙
长安大学 信息工程学院,陕西 西安 710064
Author(s):
GUO Chen; CHEN Long
School of Information Engineering,Chang’an University,Xi’an 710064,China
关键词:
机器学习; 岩石物理; 神经网络; 集成学习; 电导率
Keywords:
machine learning; digital rock physics; neural network; ensemble learning; electrical conductivity
分类号:
TP39
DOI:
10. 3969 / j. issn. 1673-629X. 2020. 07. 022
摘要:
岩芯电导率是地下油气判断及岩石物理分析的重要参数。 传统的岩石物理特性分析一般通过对岩芯样本的实验测量分析或对数字岩芯图像数据的有限元数值计算等方法,来获得岩石样本的电导率等宏观物理特性。 但这两类分析方法均需耗费较高的人工、时间成本或大量计算资源。 随着人工智能等一代数字化分析方法的发展, 文中拟开发一种采用机器学习对岩石电特性进行预测的新方法。 在研究中,将针对岩芯的三维图像数据使用集成学习(ensemble learning) 和人工神经网络(ANN) 来预测电导率。 其中传统机器学习与多层神经网络的输入特征是几何参数,而三维卷积 神经网络输入特征是三维二元分割图像。 在研究中比较各种机器学习方法的优劣。 实验表明,以三维二元分割图像作输入特征的三维卷积神经网络(3DCNN)比采用 Minkowski 泛函作输入特征的学习模型在预测岩芯电导率上性能更优。
Abstract:
The electrical conductivity of digital core is an important parameter for underground oil and gas evaluation and rock physics analysis. Traditional analysis of rock physical properties usually obtains macroscopic physical properties such as conductivity of rock samples by means  of experimental measurement and analysis of core samples or finite element numerical calculation of digital core image data. However,both methods require high labor,time cost or a large amount of computing resources. With the development of artificial intelligence and other digital analysis methods,we intend to develop a new method for predicting rock electrical properties by machine learning. In this study, the integrated learning (ensemble learning ) and artificial neural network (ANN ) are used to predict the conductivity of three-dimensional image data of core. The input features of traditional machine learning and multi-layer neural network are geometric parameters, while the input features of three - dimensional convolution neural network are three-dimensional binary segmentation images. In the research,the advantages and disadvantages of various machine learning methods are compared. Experiment shows that the three-dimensional convolution neural network (3DCNN) with three-dimensional binary segmentation image as input feature has better performance than the learning model with Minkowski functional as input feature in predicting core conductivity.

相似文献/References:

[1]陈全 赵文辉 李洁 江雨燕.选择性集成学习算法的研究[J].计算机技术与发展,2010,(02):87.
　CHEN Quan,ZHAO Wen-hui,LI Jie,et al.Research of Selective Ensemble Learning Algorithm[J].,2010,(07):87.
[2]黄秀丽 王蔚.SVM在非平衡数据集中的应用[J].计算机技术与发展,2009,(06):190.
　HUANG Xiu-li,WANG Wei.Application of SVM in Imbalances Dataset[J].,2009,(07):190.
[3]鲁晓南 接标.一种基于个性化邮件特征的反垃圾邮件系统[J].计算机技术与发展,2009,(08):155.
　LU Xiao-nan,JIE Biao.An Individual Anti- Spam Technology[J].,2009,(07):155.
[4]张苗 张德贤.多类支持向量机文本分类方法[J].计算机技术与发展,2008,(03):139.
　ZHANG Miao,ZHANG De-xian.Research on Text Categorization Based on. M- SVMs[J].,2008,(07):139.
[5]汤萍萍 王红兵.基于强化学习的Web服务组合[J].计算机技术与发展,2008,(03):142.
　TANG Ping-ping,WANG Hong-bing.Web Service Composition Based on Reinforcement -Learning[J].,2008,(07):142.
[6]杨雪洁 赵姝 张燕平.基于商空间理论的冬小麦产量预测和分析[J].计算机技术与发展,2008,(03):249.
　YANG Xue-jie,ZHAO Shu,ZHANG Yan-ping.Analysis on Winter Wheat Yield Based on Quotient Space Theory[J].,2008,(07):249.
[7]汤伟 程家兴 纪霞.一种基于概率推理的邮件过滤系统的研究与设计[J].计算机技术与发展,2008,(08):76.
　TANG Wei,CHENG Jia-xing,JI Xia.Research and Design of a Spam Filtering System Based on Probability Inference[J].,2008,(07):76.
[8]孙海虹 丁华福.基于模糊粗糙集的Web文本分类[J].计算机技术与发展,2010,(07):21.
　SUN Hai-hong,DING Hua-fu.Web Document Classification Based on Fuzzy-Rough Set[J].,2010,(07):21.
[9]汤伟 程家兴 纪霞.统计学理论在邮件分类中的应用研究[J].计算机技术与发展,2008,(12):231.
　TANG Wei,CHENG Jia-xing,JI Xia.Research and Design of a Spam Filtering System Based on Statistical Learning Theory[J].,2008,(07):231.
[10]张高胤 谭成翔 汪海航.基于K-近邻算法的网页自动分类系统的研究及实现[J].计算机技术与发展,2007,(01):21.
　ZHANG Gao-yin,TAN Cheng-xiang,WANG Hai-hang.Design and Implementation of Web Page Automation Classification System Based on K- Nearest Neighbor Algorithm[J].,2007,(07):21.

更新日期/Last Update: 2020-07-10