计算机应用 ›› 2021, Vol. 41 ›› Issue (6): 1805-1811.DOI: 10.11772/j.issn.1001-9081.2020091367

所属专题: 前沿与综合应用

• 前沿与综合应用 • 上一篇    下一篇

基于带梯度惩罚深度卷积生成对抗网络的页岩三维数字岩心重构方法

王先武1, 张挺1, 吉欣1, 杜奕2   

  1. 1. 上海电力大学 计算机科学与技术学院, 上海 200090;
    2. 上海第二工业大学 工学部, 上海 201209
  • 收稿日期:2020-09-07 修回日期:2020-11-13 出版日期:2021-06-10 发布日期:2020-11-26
  • 通讯作者: 张挺
  • 作者简介:王先武(1997-),男,江苏宿迁人,硕士研究生,主要研究方向:深度学习、图像重构;张挺(1979-),男,安徽安庆人,教授,博士,主要研究方向:图像处理、机器学习;吉欣(1997-),男,江苏南通人,硕士研究生,主要研究方向:深度学习、图像重构;杜奕(1977-),女,江苏吴江人,副教授,博士,CCF会员,主要研究方向:数据挖掘。
  • 基金资助:
    国家自然科学基金面上项目(41672114);国家自然科学基金青年基金资助项目(41702148)。

3D shale digital core reconstruction method based on deep convolutional generative adversarial network with gradient penalty

WANG Xianwu1, ZHANG Ting1, JI Xin1, DU Yi2   

  1. 1. College of Computer Science and Technology, Shanghai University of Electric Power, Shanghai 200090, China;
    2. College of Engineering, Shanghai Polytechnic University, Shanghai 201209, China
  • Received:2020-09-07 Revised:2020-11-13 Online:2021-06-10 Published:2020-11-26
  • Supported by:
    This work is partially supported by the Surface Program of National Natural Science Foundation of China (41672114), the Youth Program of National Natural Science Foundation of China (41702148).

摘要: 针对传统数字岩心重构技术存在的成本高昂、复用性差和重构质量低等问题,提出了一种基于带梯度惩罚深度卷积生成对抗网络(DCGAN-GP)的三维页岩数字岩心重构方法。首先,利用神经网络参数来描述页岩训练图像的分布概率,并完成训练图像的特征提取;其次,保存训练后的网络参数;最后,利用生成器重构出页岩三维数字岩心。实验结果表明,相较于经典的数字岩心重构技术得到的图像,DCGAN-GP得到的图像在孔隙度、变差函数和孔隙大小及分布特征上都更接近训练图像,而且DCGAN-GP的CPU使用率不到经典算法的一半,内存峰值仅有7.1 GB,重构时间达到了每次42 s,体现出模型重构质量高、效率高的特点。

关键词: 重构, 数字岩心, 生成对抗网络, 深度卷积, 梯度惩罚

Abstract: Aiming at the problems of high cost, poor reusability and low reconstruction quality in traditional digital core reconstruction technology, a 3D shale digital core reconstruction method based on Deep Convolutional Generation Adversarial Network with Gradient Penalty (DCGAN-GP) was proposed. Firstly, the neural network parameters were used to describe the distribution probability of the shale training image, and the feature extraction of the training image was completed. Secondly, the trained network parameters were saved. Finally, the 3D shale digital core was constructed by using the generator. The experimental results show that, compared to the classic digital core reconstruction technologies, the proposed DCGAN-GP obtains the image closer to the training image in porosity, variogram, as well as pore size and distribution characteristics. Moreover, DCGAN-GP has the CPU usage less than half of the classic algorithms, the memory peak usage only 7.1 GB, and the reconstruction time reached 42 s per time, reflecting the characteristics of high quality and high efficiency of model reconstruction.

Key words: reconstruction, digital core, Generative Adversarial Network (GAN), deep convolution, gradient penalty

中图分类号: