小样本条件下的深水海底扇岩石相智能分类

doi:10.11885/j.issn.1674-5086.2024.09.15.02

摘要/Abstract

摘要： 针对深水海底扇油气藏的勘探开发作业成本高、施工难度大、技术要求高，岩石取芯数量有限，因此获取的岩石相样本数量较少这一问题，提出了基于小样本的深水海底扇岩石相智能分类方法。首先，利用经验模态分解方法(EMD)和滑动窗口为每个井点构建多层图像样式作为输入；其次，利用长短时记忆网络(LSTM)和卷积神经网络(CNN)算法构建岩石相识别模型；再利用生成对抗网络(GAN)模型扩充少数类样本；最后，引入遗传算法(GA)优化模型参数。以西非尼日尔三角洲盆地Akpo油田为研究工区，利用此方法开展岩石相智能识别研究。研究表明，本文提出的GAN-GA-CNN模型对岩石相的分类准确率可达94.22%，相比原始的CNN模型的预测精度有很大提升，证明了本文所提方法的可行性。

关键词: 岩石相识别, 海底扇储层, 小样本, 深度学习, 遗传算法

Abstract: Aiming at the problems of high cost, difficulty in operation, high technical requirements, limited number of rock core-taking and small number of lithofacies samples obtained in the exploration and development of deep-water submarine fan reservoir, an intelligent classification method of deep-water submarine fan lithofacies based on small samples was proposed. First, the Empirical Mode Decomposition (EMD) and sliding window are used to construct multi-layer image styles input for each well point as inputs. Secondly, the lithofacies recognition model is constructed by using Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN) algorithms. The Generative Adversarial Networks (GAN) model was used to expand a few class samples. Finally, a Genetic Algorithm (GA) was introduced to optimize the model parameters. Taking Akpo Oilfield in the Niger Delta Basin of West Africa as the research area, this method is used to carry out the intelligent identification of lithofacies. Research shows that the lithofacies classification accuracy of the GAN-GA-CNN model proposed in this paper can reach 94.22%, which greatly improves the prediction accuracy compared with the original CNN model, proving the feasibility of the proposed method.

Key words: lithofacies identification, submarine fan reservoir, small sample, deep learning, genetic algorithm

中图分类号:

TE319

甄艳, 刘小伟, 张艺昊, 赵珍, 肖逸菲, 赵晓明. 小样本条件下的深水海底扇岩石相智能分类[J]. 西南石油大学学报(自然科学版), 2025, 47(6): 60-71.

ZHEN Yan, LIU Xiaowei, ZHANG Yihao, ZHAO Zhen, XIAO Yifei, ZHAO Xiaoming. Intelligent Classification of Deep-water Submarine Fan Lithofacies Based on Small Sample[J]. Journal of Southwest Petroleum University(Science & Technology Edition), 2025, 47(6): 60-71.

0
/ / 推荐

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://journal15.magtechjournal.com/Jwk_xnzk/CN/10.11885/j.issn.1674-5086.2024.09.15.02

http://journal15.magtechjournal.com/Jwk_xnzk/CN/Y2025/V47/I6/60

参考文献

[1] PICKERING K T, HISCOTT R N. Deep marine systems: Processes, deposits, environments, tectonics and sedimentation[M]. Hoboken: American Geophysical Union, 2015.
[2] 刘小兵，窦立荣. 国际大油公司深水油气勘探实践及启示——以圭亚那斯塔布鲁克区块为例[J]. 中国石油勘探， 2023， 28（3）: 78-89. doi: 10.3969/j.issn.1672-7703.2023.03.007 LIU Xiaobing, DOU Lirong. Practice and enlightenment of deepwater petroleum exploration of international major oil companies: A case study of Guiyana Stabroek Block[J]. China Petroleum Exploration, 2023, 28(3): 78-89. doi: 10.3969/j.issn.1672-7703.2023.03.007
[3] 李大伟，李德生，陈长民，等. 深海扇油气勘探综述[J]. 中国海上油气， 2007， 19（1）: 18-24. doi: 10.3969/j.issn.1673-1506.2007.01.004 LI Dawei, LI Desheng, CHEN Changmin, et al. An overview of hydrocarbon exploration in deep submarine fans[J]. China Offshore Oil and Gas, 2007, 19(1): 18-24. doi: 10.3969/j.issn.1673-1506.2007.01.004
[4] 赵晓明，吴胜和，岳大力，等. 西非某油田深水海底扇岩石相类型及其识别方法研究[J]. 测井技术， 2010， 34（5）: 505-510. doi: 10.3969/j.issn.1004-1338.2010.05.023 ZHAO Xiaoming, WU Shenghe, YUE Dali, et al. Research on Litho faces types and identification method of deep-water submarine fan—Taking one oilfield of West African as a case[J]. Well Logging Technology, 2010, 34(5): 505-510. doi: 0.3969/j.issn.1004-1338.2010.05.023
[5] GUO Shuwen, YANG Naxia, GUO Chunxiang, et al. Intelligent sedimentary lithofacies identification with integrated well logging features[J]. IEEE Geoscience and Remote Sensing Letters, 2024, 21: 7500805. doi: 10.1109/LGRS.2023.3347565
[6] JIANG Chunbi, ZHANG Dongxiao, CHEN Shifeng. Lithology identification from well-log curves via neural networks with additional geologic constraint[J]. Geophysics, 2021, 86(5): 85-100. doi: 10.1190/geo2020-0676.1
[7] 孙玉洁，李晓彦，张超. 基于机器学习的黑云母成分判别花岗岩成因类型方法研究[J]. 现代地质， 2025， 39（3）: 523-540. doi: 10.19657/j.geoscience.1000-8527.2025.001 SUN Yujie, LI Xiaoyan, ZHANG Chao. Machine-learning based discrimination of granite type using biotite composition[J]. Geoscience, 2025, 39(3): 523-540. doi: 10.19657/j.geoscience.1000-8527.2025.001
[8] SANTOS D T D, ROISENBERG M, NASCIMENTO M D S. Deep recurrent neural networks approach to sedimentary facies classification using well logs[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 19: 1-5. doi: 10.1109/LGRS.2021.3053383
[9] 王婷婷，黄志贤，王洪涛，等. 基于MobileNetV2的岩石薄片岩性识别[J]. 吉林大学学报（地球科学版）， 2024， 54（4）: 1432-1442. doi: 10.13278/j.cnki.jjuese.20230106 WANG Tingting, HUANG Zhixian, WANG Hongtao, et al. Rock thin slice lithology identification based on MobileNetV2[J]. Journal of Jilin University (Earth Science Edition), 2024, 54(4): 1432-1442. doi: 10.13278/j.cnki.jjuese.20230106
[10] 马陇飞，萧汉敏，陶敬伟，等. 基于梯度提升决策树算法的岩性智能分类方法[J]. 油气地质与采收率， 2022， 29（1）: 21-29. doi: 10.13673/j.cnki.cn37-1359/te.2022.01.003 MA Longfei, XIAO Hanmin, TAO Jingwei, et al. Intelligent lithology classification method based on GBDT algorithm[J]. Petroleum Geology and Recovery Efficiency, 2022, 29(1): 21-29. doi: 10.13673/j.cnki.cn37-1359/te.2022.01.003
[11] XU Zhenhao, MA Wen, LIN Peng, et al. Deep learning of rock microscopic images for intelligent lithology identification: Neural network comparison and selection[J]. Journal of Rock Mechanics and Geotechnical Engineering, 2022, 14(4): 1140-1152. doi: 10.1016/j.jrmge.2022.05.009
[12] 张凤博，马雪玲，董珍珍，等. 基于CNN和LSTM的机器学习模型在测井岩性识别的应用[J]. 西安石油大学学报（自然科学版）， 2024， 39（5）: 96-103， 133. doi:10.3969/j.issn.1673-064X.2024.05.012 ZHANG Fengbo, MA Xueling, DONG Zhenzhen, et al. Application of machine learning model based on CNN and LSTM in well logging lithology identification[J]. Journal of Xi'an Shiyou University (Natural Science Edition), 2024, 39(5): 96-103, 133. doi: 10.3969/j.issn.1673-064X.2024.05.012
[13] 管耀，王清辉，冯进，等. 基于机器学习的蚀变火成岩测录井综合岩性识别: 以南海北部珠江口盆地惠州266 井区为例[J]. 吉林大学学报（地球科学版）, 2024, 54（1）: 345-358. doi: 10.13278/j.cnki.jjuese.20220310 GUAN Yao, WANG Qinghui, FENG Jin, et al. Comprehensive lithology recognition of altered igneous reservoirs based on machine learning for wireline and cutting logs in Huizhou Depression, Pearl River Mouth Basin, northern South China Sea[J]. Journal of Jilin University (Earth Science Edition), 2024, 54(1): 345-358. doi: 10.13278/j.cnki.jjuese.20220310
[14] 郑阳. 基于深度学习的岩性识别研究[D]. 北京:中国石油大学（北京）， 2017. ZHENG Yang. Research on lithology recognition based on deep learning[D]. Beijing: China University of Petroleum (Beijing), 2017.
[15] 武中原，张欣，张春雷，等. 基于LSTM循环神经网络的岩性识别方法[J]. 岩性油气藏， 2021， 33（3）: 120-128. doi: 10.12108/yxyqc.20210312 WU Zhongyuan, ZHANG Xin, ZHANG Chunlei, et al. Lithology identification based on LSTM recurrent neural network[J]. Lithologic Reservoirs, 2021, 33(3): 120-128. doi: 10.12108/yxyqc.20210312
[16] SHAN Liqun, LIU Yanchang, TANG Min, et al. CNNBiLSTM hybrid neural networks with attention mechanism for well log prediction[J]. Journal of Petroleum Science & Engineering, 2021, 205: 108838. doi: 10.1016/j.petrol.2021.108838
[17] 冯雅兴，龚希，徐永洋，等. 基于岩石新鲜面图像与孪生卷积神经网络的岩性识别方法研究[J]. 地理与地理信息科学， 2019， 35（5）: 89-94. doi: 10.3969/j.issn.1672-0504.2019.05.015 FENG Yaxing, GONG Xi, XU Yongyang, et al. Lithology recognition based on fresh rock images and twins convolution neural network[J]. Geography and Geo-Information Science, 2019, 35(5): 89-94. doi: 10.3969/j.issn.1672-0504.2019.05.015
[18] ZHU Liping, LI Hongqi, YANG Zhongguo, et al. Intelligent logging lithological interpretation with convolution neural networks[J]. Petrophysics, 2018, 59(6): 799-810. doi: 10.30632/PJV59N6-2018a5
[19] CHAWLA N V, BOWYER K W, HALL L O. SMOTE: Synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16: 321-357.
[20] GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139-144. doi: 10.1145/3422622
[21] HSIN Y Y, DAI T S, TI Y W, et al. Feature engineering and resampling strategies for fund transfer fraud with limited transaction data and a time-inhomogeneous modi operandi[J]. IEEE Access, 2022, 10: 86101-86116. doi: 10.1109/ACCESS.2022.3199425
[22] HARSHINI S V, NIVEDHIDHA M, RAMKUMAR M P, et al. Copula GAN boosted random forest based network intrusion detection system for hospital network infrastructure[C]. Delhi: 14th International Conference on Computing Communication and Networking Technologies, 2023.
[23] QADDOURA R, BILTAWI M M. Improving fraud detection in an imbalanced class distribution using different oversampling techniques[C]. Zarqa: 2022 International Engineering Conference on Electrical, Energy, and Artificial Intelligence, 2022.
[24] PATIL A P, JERE S, RAM R, et al. T5W: A paraphrasing approach to oversampling for imbalanced text classification[C]. Bangalore: 2022 IEEE International Conference on Electronics, Computing and Communication Technologies, 2022.
[25] 王泽尚. 基于聚类回归的油气井套损图像智能检测研究[D]. 西安:西安石油大学， 2021. WANG Zeshang. Intelligent detection of casing damage image of oil and gas well based on cluster and regression[D]. Xi'an: Xi'an Shiyou University, 2021.
[26] 王光宇，宋建国，徐飞，等. 不平衡样本集随机森林岩性预测方法[J]. 石油地球物理勘探， 2021， 56（4）: 679-687， 669. doi: 10.13810/j.cnki.issn.1000-7210.2021.04.001 WANG Guangyu, SONG Jianguo, XU Fei, et al. Random forests lithology prediction method for imbalanced data sets[J]. Oil Geophysical Prospecting, 2021, 56(4): 679-687, 669. doi: 10.13810/j.cnki.issn.1000-7210.2021.04.001
[27] HUANG N E, SHEN Zheng, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Mathematical, Physical and Engineering Sciences, 1998, 454(1971): 903-995.
[28] 徐晓刚，徐冠雷，王孝通，等. 经验模式分解（EMD）及其应用[J]. 电子学报， 2009， 37（3）: 581-585. doi: 10.3321/j.issn:0372-2112.2009.03.028 XU Xiaogang, XU Guanlei, WANG Xiaotong, et al. Empirical mode decomposition and its application[J]. Acta Electronica Sinica, 2009, 37(3): 581-585. doi: 10.3321/j.issn:0372-2112.2009.03.028
[29] 李海涛，邓少贵，王跃祥，等. 基于经验模态分解的核磁共振去噪方法研究[J]. 西南石油大学学报（自然科学版）， 2020， 42（3）: 51-59. doi: 10.11885/j.issn.1674-5086.2019.09.16.32 LI Haitao, DENG Shaogui, WANG Yuexiang, et al. Research on NMR denoising method based on empirical mode decomposition[J]. Journal of Southwest Petroleum University (Science & Technology Edition), 2020, 42(3): 51-59. doi: 10.11885/j.issn.1674-5086.2019.09.16.32
[30] 杨勇，王观军，孙东，等. 基于金属磁记忆效应的管道泄漏定位技术研究[J]. 西南石油大学学报（自然科学版）， 2013， 35（5）: 165-171. doi: 10.3863/j.issn.1674-5086.2013.05.024 YANG Yong, WANG Guanjun, SUN Dong, et al. Study on leakage detection and localization of underground pipelines based on metal magnetic memory[J]. Journal of Southwest Petroleum University (Science & Technology Edition), 2013, 35(5): 165-171. doi: 10.3863/j.issn.1674-5086.2013.05.024
[31] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780. doi: 10.1162/neco.1997.9.8.1735
[32] 钟原，刘小溶，王杰，等. 基于NER的石油非结构化信息抽取研究[J]. 西南石油大学学报（自然科学版）， 2020， 42（6）: 165-173. doi: 10.11885/j.issn.1674-5086.2020.05.12.01 ZHONG Yuan, LIU Xiaorong, WANG Jie, et al. Research of extraction on petroleum unstructured information based on named entity recognition[J]. Journal of Southwest Petroleum University (Science & Technology Edition), 2020, 42(6): 165-173. doi: 10.11885/j.issn.1674-5086.2020.05.12.01
[33] 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.
[34] LECUN Y A, BOSER B, DENKER J S, et al. Backpropagation applied to handwritten zip code recognition[J]. Neural Computation, 1989, 1(4): 541-551. doi: 10.1162/neco.1989.1.4.541
[35] 钟仪华，王淑宁，罗兰，等. 用深度学习挖掘油田开发指标预测模型的知识[J]. 西南石油大学学报（自然科学版）， 2020， 42（6）: 63-74. doi: 10.11885/j.issn.1674-5086.2020.05.11.02 ZHONG Yihua, WANG Shuning, LUO Lan, et al. Knowledge mining for oilfield development index prediction model using deep learning[J]. Journal of Southwest Petroleum University (Science & Technology Edition), 2020, 42(6): 63-74. doi: 10.11885/j.issn.1674-5086.2020.05.11.02
[36] 杨永飞，刘夫贵，姚军，等. 基于生成对抗网络的页岩三维数字岩芯构建[J]. 西南石油大学学报(自然科学版)， 2021， 43（5）: 73-83. doi: 10.11885/j.issn.1674-5086.2021.01.15.02 YANG Yongfei, LIU Fugui, YAO Jun, et al. Reconstruction of 3D shale digital rock based on generative adversarial network[J]. Journal of Southwest Petroleum University (Science & Technology Edition), 2021, 43(5): 73-83. doi: 10.11885/j.issn.1674-5086.2021.01.15.02
[37] THOMAS B. Genetic algorithms+data structures=evolution programs[J]. Computational Statistics and Data Analysis, 1997, 24(3): 372-373. doi: 10.1016/S0167-947-3(97)87028-4
[38] STEPHAN A G, ADEDAYO A A, TIMOTHY R M. Transient fan architecture and depositional controls from near-surface 3-D seismic data, Niger Delta continental slope[J]. AAPG Bulletin, 2005, 89(5): 627-643. doi: 10.1306/11200404025
[39] DAMUTH J E. Neogene gravity tectonics and depositional processes on the deep Niger Delta continental margin[J]. Marine and Petroleum Geology, 1994, 11(3): 321-346. doi: 10.1016/0264-8172(94)90053-1
[40] HARVEY A, COHEN K M. Sedimentation and shale tectonics of the northwestern Niger Delta front[J]. Marine and Petroleum Geology, 1996, 13(3): 313-328. doi: 10.1016/0264-8172(95)00067-4
[41] JOHN H S, FRANK B, FREDDY C. Structural styles in the deep-water fold and thrust belts of the Niger Delta[J]. AAPG Bulletin, 2005, 89(6): 753-780. doi: 10.1306/02170504074