Reservoir Prediction Based on Improved U-Net Convolutional Neural Network
-
摘要: 传统的U-Net卷积神经网络大多存在深层网络梯度消失的问题。本文在U-Net卷积神经网络中加入残差模块，提出了一种改进U-Net卷积神经网络。残差模块保证了U-Net卷积神经网络在误差反向传播过程中梯度的存在，在一定程度上可以缓解梯度消失的问题。最后将改进U-Net卷积神经网络应用于实际储层预测中，实际数据测试结果表明基于改进U-Net卷积神经网络在岩性识别以及“甜点”预测上均能取得较好的效果。Abstract: Most of the traditional U-Net convolutional neural networks have the problem that the gradient of the deep network disappears. In this paper, a residual module is added to the U-Net convolutional neural network, and an improved U-Net convolutional neural network is proposed. The residual module guarantees the existence of the gradient of the U-Net convolutional neural network in the process of error back-propagation, which can alleviate the problem of gradient disappearance to a certain extent. Finally, the improved U-Net convolutional neural network is applied to the actual reservoir prediction. The actual data measurement shows that the improved U-Net convolutional neural network can achieve better results in lithology identification and "Sweet Point" prediction.
-
Keywords:
- convolutional neural network /
- U-Net /
- deep learning /
- lithology recognition
-
-
[1] MARFURT K.J, KIRLIN R L, FARMER S H, et al. 3-D seismic attributes using a semblance-based coherency algorithm[J]. Geophysics, 1998, 63(4):1150-1165.
[2] GERSZTENKORN A, MARFURT K J. Eigenstructure-based coherence computations as an aid to 3-D structural and stratigraphic mapping[J]. Geophysics, 1999, 64(5):1468-1479.
[3] GOODFELLOW I, BENGIO Y, COURVILLE A. Deep learning (Vol.1)[M]. Cambridge:MIT press, 2016:326-366.
[4] HINTON G E, OSINDERO S, TEH Y W. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18(7):1527-1554.
[5] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016:770-778.
[6] HOFFMAN J, WANG D, YU F, et al. Fcns in the wild:Pixel-level adversarial and constraint-based adaptation[J]. arXiv 2016, arXiv:1612.02649.
[7] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11):2278-2324.
[8] 印兴耀, 葸晓宇, 张繁昌. Kohonen自组织神经网络的改进及其在地震多属性分析中的应用[C]//CPS/SEG国际地球物理会议, 2004. [9] 丁峰, 尹成, 朱振宇, 等. 利用改进的自组织网络进行地震属性分析[J]. 西南石油大学学报(自然科学版), 2009, 31(4):47-51, 200. DING F, YING C, ZHU Z Y, et al. Seismic attribute analysis by using improved self-organizing network[J]. Journal of Southwest Petroleum University (Seience & Technology Edition), 2009, 31(4):47-51, 200. (in Chinese). [10] 张繁昌, 刘汉卿, 钮学民. 褶积神经网络高分辨率地震反演[J]. 石油地球物理勘探, 2014, 49(6):1165-1169. [11] OYEDELE O A, DUPRÉ W R, MARFURT K J. Seismic facies analysis and age dating of mid-pleistocene channel-lobe deposits, mad dog field, gulf of mexico[J]. Gulf Coast Association of Geological Societies Transactions, 2015, 65:301-312.
[12] ARAYA-POLO M, DAHLKE T, FROGNER C, et al. Automated fault detection without seismic processing[J]. The Leading Edge, 2017, 36(3):208-214.
[13] ZHAO T, MUKHOPADHYAY P. A fault detection workflow using deep learning and image processing[C]//SEG Technical Program Expanded Abstracts, 2018:1966-1970.
[14] GRAMSTAD O, NICKEL M. Automated interpretation of top and base salt using deep convolutional networks[C]//SEG Technical Program Expanded Abstracts, 2018:1956-1960.
[15] MOSSER L, KIMMAN W, DRAMSCH J, et al. Rapid seismic domain transfer:Seismic velocity inversion and modeling using deep generative neural networks[C]//EAGE Conference and Exhibition, 2018:1-5.
[16] JIN L. Machine learning approaches for seismic facies prediction and reservoir property inversion[C]//SEG Technical Program Expanded Abstracts, 2018:2147-2151.
[17] 周游, 张广智, 高刚, 等. 核主成分分析法在测井浊积岩岩性识别中的应用[J]. 石油地球物理勘探, 2019, 54(3):667-675 , 490.
[18] CHEVITARESE D S, SZWARCMAN D, e Silva R M G, et al. Deep learning applied to seismic facies classification:A methodology for training[C]//Saint Petersburg 2018.
[19] HINTON G E. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786):504-507.
[20] IOFFE S, SZEGEDY C. Batch normalization:accelerating deep network training by reducing internal covariate shift[C]//International Conference on International Conference on Machine Learning, 2015.
[21] SUSSILLO D, ABBOTT L F. Random walk initialization for training very deep feed forward networks[J]. arXiv 2015, arXiv:1412.6558.
[22] RONNEBERGER O, FISCHER P, BROX T. U-net:Convolutional networks for biomedical image segmentation[C]//International Conference on Medical image computing and computer-assisted intervention, 2015:234-241.
-
期刊类型引用(8)
1. 周雨晴，张佳佳，张广智，蔺营. 致密砂岩储层岩石物理建模及“甜点”预测. 地球物理学进展. 2025(01): 255-265 . 
百度学术
2. 汪锐，李芳，刘仕友，孙万元，李松龄，黄晟. 岩石物理建模引导的低渗储层参数预测方法. 煤田地质与勘探. 2024(08): 187-197 . 百度学术
3. 钱佩璋，乔志伟，杜聪聪. 基于3MNet去噪网络的快速EPRI成像. CT理论与应用研究. 2023(01): 55-66 . 本站查看
4. 王迪，张益明，张繁昌，丁继才，牛聪. 利用先验信息约束的深度学习方法定量预测致密砂岩“甜点”. 石油地球物理勘探. 2023(01): 65-74 . 百度学术
5. 张傲，李宗杰，刘军，闫星宇，李伟，顾汉明. 基于Yolox算法的碳酸盐岩储层溶洞“串珠状”异常反射智能检测. 石油地球物理勘探. 2023(03): 540-549 . 百度学术
6. 杜聪聪，乔志伟. 基于对抗式残差密集深度神经网络的CT稀疏重建. CT理论与应用研究. 2022(02): 163-172 . 本站查看
7. 刘仕友，曲福良，周凡，邓利峰. 基于地震属性约简的深度学习储层物性参数预测:以莺歌海盆地乐东区为例. CT理论与应用研究. 2022(05): 577-586 . 本站查看
8. 郑琳. 基于物联网边缘计算的数据挖掘方法研究. 无线互联科技. 2022(15): 140-142 . 百度学术
其他类型引用(23)
计量
- 文章访问数: 451
- HTML全文浏览量: 68
- PDF下载量: 58
- 被引次数: 31
下载:
