[1]张可佳,方佳佳,高楷程,等.基于神经网络的岩性物性参数计算方法研究[J].计算机技术与发展,2022,32(06):186-191.[doi:10. 3969 / j. issn. 1673-629X. 2022. 06. 031]
　ZHANG Ke-jia,FANG Jia-jia*,GAO Kai-cheng,et al.Study on Calculation Method of Lithologic Physical Property Parameters Based on Neural Network[J].,2022,32(06):186-191.[doi:10. 3969 / j. issn. 1673-629X. 2022. 06. 031]
点击复制

基于神经网络的岩性物性参数计算方法研究()
分享到：

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

卷:
32
期数:
2022年06期
页码:
186-191
栏目:
应用前沿与综合
出版日期:
2022-06-10

文章信息/Info

Title:
Study on Calculation Method of Lithologic Physical Property Parameters Based on Neural Network
文章编号:
1673-629X(2022)06-0186-06
作者:
张可佳; 方佳佳; 高楷程; 刘 涛
东北石油大学,黑龙江 大庆 163318
Author(s):
ZHANG Ke-jia; FANG Jia-jia* ; GAO Kai-cheng; LIU Tao
Northeast Petroleum University,Daqing 163318,China
关键词:
岩性物性参数; 孔隙度; 渗透率; BP 神经网络; 测井曲线; 岩性识别
Keywords:
lithology physical property parameters; porosity; permeability; BP neural network; logging curves; lithology recognition
分类号:
TP183
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 06. 031
摘要:
岩性物性参数的计算会关系到后续油气田开发生产,因而备受录井解释、地质分析等部门的高度重视。 为了解决传统数理计算方法计算岩性物性参数需要依赖人工经验,计算量大,造成人力成本和时间成本较高且计算准确率不高的问题,提出一种基于神经网络的岩性物性参数计算方法。 通过对研究的储层地质特征进行分析,选取孔隙度和渗透率作为反映岩性的物性参数;阐明应用测井曲线反映岩性物性参数的合理性,选取合适的测井曲线,应用不同的测井曲线特征提取方法提取测井曲线特征,作为神经网络的输入参数;设计岩性物性参数计算的技术路线,构建 BP 神经网络模型,选取样本数据集对神经网络模型完成模型训练,最终实现对选区岩性物性参数的计算,为储层岩性识别提供依据。 实验结果表明,基于 BP 神经网络的岩性物性参数计算方法能够较为准确地计算岩性物性参数。
Abstract:
The calculation of lithologic physical property parameters is related to the subsequent development and production of oil and gasfields,so it is highly valued by departments of logging interpretation and geological analysis. In order to solve the problem that thecalculation of lithologic physical property parameters by traditional mathematical calculation method depends on manual experience,thecalculation amount is large,resulting in high labor cost and time cost,and the calculation accuracy is not high,a calculation method of lithologic physical property parameters based on neural network is proposed. Through the analysis of the geological characteristics of thestudied reservoir,porosity and permeability are selected as the physical property parameters reflecting the lithology. We expound therationality of using logging curves to reflect lithologic physical property parameters,select appropriate logging curves and use differentlogging curves feature extraction methods to extract the characteristics of logging curves as the input parameters of neural network. Thetechnical route of lithology physical property parameter calculation is designed,the BP neural network model is constructed,the sampledata set is selected to complete the model training of the neural network model,and finally the calculation of lithology physical propertyparameters in the selection area is realized,which provides a basis for reservoir lithology recognition. The experimental results show thatthe proposed method can accurately calculate the parameters of lithology physical properties.

相似文献/References:

[1]陈国军,李 胜.基于空间标记的岩心孔隙并行分割[J].计算机技术与发展,2020,30(12):142.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 025]
　CHEN Guo-jun,LI Sheng.Parallel Segmentation of Core Pores Based on Spatial Markers[J].,2020,30(06):142.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 025]

更新日期/Last Update: 2022-06-10