DBN在测井解释中的研究与应用

doi:10.3969/j.issn.0255-8297.2018.04.012

应用科学学报 ›› 2018, Vol. 36 ›› Issue (4): 689-697.doi: 10.3969/j.issn.0255-8297.2018.04.012

DBN在测井解释中的研究与应用

段友祥, 徐冬胜, 孙歧峰, 李钰

中国石油大学(华东)计算机与通信工程学院, 山东青岛 266580

收稿日期:2017-08-04 修回日期:2017-10-05 出版日期:2018-07-31 发布日期:2018-07-31
通信作者: 段友祥,教授,研究方向:网络与服务计算、计算机技术在油气领域的应用等,E-mail:yxduan@upc.edu.cn E-mail:yxduan@upc.edu.cn
基金资助:
十三·五“重大专项”基金（No.2017ZX05009-001）资助

Research and Application on DBN for Well Log Interpretation

DUAN You-xiang, XU Dong-sheng, SUN Qi-feng, LI Yu

College of Computer and Communication Engineering, China University of Petroleum(East China), Qingdao 266580, China

Received:2017-08-04 Revised:2017-10-05 Online:2018-07-31 Published:2018-07-31

摘要/Abstract

摘要： 测井解释就是将测井信息加工解释成地质信息，以往大多通过数学方法建立解释模型或者使用最基本的反向传播（back propagation，BP）网络来完成这项工作.针对BP网络存在学习训练效率不高的问题，提出了将深度置信网络应用于测井曲线解释.针对测井解释的特点，选择4条测井曲线数据作为输入进行泥砂分层以及孔隙度的预测实验，并与BP网络的预测结果进行对比分析.实验表明，深度置信网络可用于测井曲线解释，其分类精度较一般BP算法有所提高并且训练时间有所降低.

关键词: 人工神经网络, 深度置信网络, 测井解释, 泥砂分层, 储层参数预测

Abstract: Well log interpretation refers to interpreting logging information into geological information, which was generally accomplished by establishing mathematical models or using the fundamental BP networks in the past. This study proposes to apply the deep belief network (DBN) to the interpretation of logging curve. We used four well log curves as input parameters, conducted the mudstone, and conducted the sandstone layering experiment and reservoir parameter prediction experiment with the DBN method. The results of experiment show that the DBN performs well in the interpretation of logging curve, with higher classification accuracy and shorter training time than that of BP algorithm.

Key words: artificial neural network, well log interpretation, prediction of reservoir parameters, deep belief network (DBN), mudstone and sandstone layering

中图分类号:

TP391

段友祥, 徐冬胜, 孙歧峰, 李钰. DBN在测井解释中的研究与应用[J]. 应用科学学报, 2018, 36(4): 689-697.

DUAN You-xiang, XU Dong-sheng, SUN Qi-feng, LI Yu. Research and Application on DBN for Well Log Interpretation[J]. Journal of Applied Sciences, 2018, 36(4): 689-697.

参考文献

[1] 张平,潘保芝,张莹,王鹏,董瑞新. 自组织神经网络在火成岩岩性识别中的应用[J]. 石油物探,2009, 48(1):53-56. Ping Z, Pan B Z, Zhang Y, Wang P, Dong R X. Application of self-organization maps network in identifying the lithology of igneous rock[J]. Geophysical Prospecting for Petroleum, 2009, 48(1):53-56. (in Chinese)
[2] 陈蓉,王峰. 基于MATLAB的BP神经网络在储层物性预测中的应用[J]. 测井技术,2009, 33(1):75-78. Chen R, Wang F. Application of MATLAB-based of BP neural network in reservoir parameters prediction[J]. Well Logging Technology, 2009, 33(1):75-78. (in Chinese)
[3] Hamidi H, Rafati R. Prediction of oil reservoir porosity based on BP-ANN[C]//IEEE International Conference on Innovation Management and Technology Research, 2012:241-246.
[4] 肖波,韩学辉,周开金,支乐菲. 测井曲线自动分层方法回顾与展望[J]. 地球物理学进展,2010, 25(5):1802-1810. Xiao B, Han X H, Zhou K J, Zhi L F. A review and outlook of automatic zonation methods of well log[J]. Progress in Geophysics, 2010, 25(5):1802-1810. (in Chinese)
[5] 靳玉萍,李保霖. 基于遗传优化径向基概率神经网络的岩性识别应用[J]. 计算机应用,2013, 33(2):353-356. Jin Y P, Li B L. Lithology identification based on genetic optimized radial basis probabilistic neural network[J]. Journal of Computer Applications, 2013, 33(2):353-356. (in Chinese)
[6] 胡启华,范晶晶,张新. 应用BP神经网络预测油页岩含油率[J]. 计算机应用,2014(S2):186-189. Hu Q H, Fan J J, Zhang X. Application of BP neural network in oil content prediction[J]. Journal of Computer Applications, 2014(S2):186-189. (in Chinese)
[7] 潘少伟,梁鸿军,李良,王家华. 改进PSO-BP神经网络对储层参数的动态预测研究[J]. 计算机工程与应用,2014, 50(10):52-56. Pan S W, Liang H J, Liang L, Wang J H. Dynamic prediction on reservoir parameter by improved PSO-BP neural network[J]. Computer Engineering & Applications, 2014, 50(10):52-56. (in Chinese)
[8] 单敬福,陈欣欣,赵忠军,葛雪,张芸. 利用BP神经网络法对致密砂岩气藏储集层复杂岩性的识别[J]. 地球物理学进展,2015(3):1257-1263.) Shan J F, Chen X X, Zhao Z J, Ge X, Zhang Y. Identification of complex lithology for tight sandstone gas reservoirs SASE on BP neural net[J]. Progress in Geophysics, 30(3):1257-1263. doi:10.6038/pg20150335. (in Chinese)
[9] 瞿晓婷,张蕾,冯宏伟,王惠亚,张涛,冯筠. 面向复杂储层的非均衡测井数据的岩性识别[J]. 地球物理学进展,2016, 31(5):2128-2132. Qu X T, Zhang L, Feng H W, Wang H Y, Zhang T, Feng Y. Lithology identification for imbalanced logging data on complex reservoirs[J]. Progress in Geophysics, 2016, 31(5):2128-2132, doi:10.6038/pg20160533. (in Chinese)
[10] Hinton G E, Osindero S, Teh Y W. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18(7):1527-1554.
[11] 吕启,窦勇,牛新,徐佳庆,夏飞. 基于DBN模型的遥感图像分类[J]. 计算机研究与发展,2014, 51(9):1911-1918. Lü Q, Dou Y, Niu X, Xu J Q, Xia F. Remote sensing image classification based on DBN model[J]. Journal of Computer Research & Development, 2014, 51(9):1911-1918. (in Chinese)
[12] 孙劲光,蒋金叶,孟祥福,李秀娟. 一种数值属性的深度置信网络分类方法[J]. 计算机工程与应用,2014, 50(2):112-115. Sun J G, Jiang J Y, Meng X F, Li X J. DBN classification algorithm for numerical attribute[J]. Computer Engineering and Applications, 2014, 50(2):112-115. (in Chinese)
[13] 孙劲光,蒋金叶,孟祥福,李秀娟. 深度置信网络在垃圾邮件过滤中的应用[J]. 计算机应用,2014, 34(4):1122-1125. Xun J G, Jiang J X, Meng X F, Li X J. Application of deep belief nets in spam filtering[J]. Journal of Computer Applications, 2014, 34(4):1122-1125. (in Chinese)
[14] 樊雅琴,王炳皓,王伟,唐烨伟. 深度学习国内研究综述[J]. 中国远程教育,2015(6):27-33. Fan Y Q, Wang B H, Wang W, Tang Y W. A review of Chinese literature on deep learning[J]. Distance Education in China, 2015(6):27-33. (in Chinese)
[15] 焦李成,杨淑媛,刘芳,王士刚,冯志玺. 神经网络七十年:回顾与展望[J]. 计算机学报,2016, 39(8):1697-1716. Jiao L C, Yang S Y, Liu F, Wang S G, Feng Z X. Seventy years beyond neural networks:retrospect and prospect[J]. Chinese Journal of Computers, 2016, 39(8):1697-1716. (in Chinese)

DBN在测井解释中的研究与应用

Research and Application on DBN for Well Log Interpretation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 10

编辑推荐

Metrics

本文评价

[1]	蔡安辉, 潘冶, 刘永刚, 孙国雄. 基于人工神经网络的L⁹(3⁴)正交实验预测特点研究[J]. 应用科学学报, 2004, 22(2): 228-232.
[2]	夏善木, 李大勇. 用于铸铁力学性能快速预测的人工神经网络[J]. 应用科学学报, 2002, 20(3): 309-312.
[3]	訾炳涛, 姚可夫, 崔建忠, 巴启先. 铝合金凝固晶粒尺寸的人工神经网络研究[J]. 应用科学学报, 2001, 19(4): 353-356.
[4]	周川, 胡寿松, 马勤弟, 胡维礼. 基于神经网络辨识的表面温度传感器动态标定及瞬态温度测量[J]. 应用科学学报, 1999, 17(1): 45-50.
[5]	杨宏, 唐晓军, 刘勇, 顾凡, 徐益谦. 基于火焰图像特性的神经网络诊断系统研究[J]. 应用科学学报, 1998, 16(1): 81-87.
[6]	任雪斌, 杨建良, 黄德修. 不对称刻纹光纤传感阵列及其神经网络处理[J]. 应用科学学报, 1997, 15(4): 408-412.
[7]	钱诗智, 陆继东, 李广. 煤焦非均相着火温度与煤种关系的神经网络模型[J]. 应用科学学报, 1997, 15(2): 211-216.
[8]	陆文聪, 方建慧, 丁益民, 阎立诚, 陈念贻. 自组织神经树用于芬太尼衍生物构效关系的研究[J]. 应用科学学报, 1997, 15(2): 217-222.
[9]	陆文聪, 阎立诚, 陈念贻. 人工神经网络法用于V-PTC材料诸性能的预报[J]. 应用科学学报, 1996, 14(3): 375-378.
[10]	沈涛, 甘骏人, 姚林声. 二分联结模型在集成电路分级总体布线中的应用[J]. 应用科学学报, 1994, 12(2): 127-132.