针对传统储层参数预测方法常忽略渗流机理和参数相关性的问题,提出了一种融合渗流理论与迁移学习的储层参数预测模型。通过SMOTE过采样算法,有效处理了样本不均衡问题;利用随机森林算法建立了岩性及渗流能力判别模型,为储层参数预测提供渗流机理信息;结合参数相关性,运用迁移学习技术搭建储层参数预测模型。结果表明,通过引入岩性和渗流能力判别技术,能够进行储层参数之间的相关性分析,可以有效提升储层参数的预测精度。该模型的储层孔隙度和渗透率的误差为3.51%和15.17%,预测准确率较传统的储层参数预测方法有显著提升。该方法可有效解决非均质储层参数预测难题,为人工智能技术与物理模型结合研究提供参考。
To address the issue of traditional methods neglecting flow mechanisms and parameter correlations,a prediction model of reservoir parametert hat integrates seepage theory with transfer learning is proposed.By using the oversampling algorithm of SMOTE,the issue of imbalanced samples is effectively addressed.The discriminative model of lithology andseepage capacity is established by using random forest,which provides the information of seepage mechanism for reservoir parameter predicting.Combined with parameter correlation,the technology of transfer learningis used to build a reservoir parameter prediction model.The results show that the correlation analysis between reservoir parameters can be conducted by introducing lithology and seepage capacity discrimination technology,which can effectively improve the prediction accuracy of reservoir parameters.The prediction error of the model in porosity and permeability parameters is 3.51% and 15.17%,respectively,and the prediction accuracy is significantly improved.This method can effectively address the issues of parameters prediction in heterogeneous reservoir, and provide reference for the research combining artificial intelligence technology with physical models.
[1] 王斌,魏柳斌,于小伟,等.不同孔隙结构碳酸盐岩的岩石物理响应特征及储层预测新方法——以鄂尔多斯盆地奥陶系马家沟组四段为例[J].天然气工业,2023,43(3):46-58.
WANG Bin,WEI Liubin,YU Xiaowei,et al.Petrophysical response characteristics of carbonate rocks with different pore structures and new reservoir prediction method:a case study of the fourth member of Ordovician Majiagou Formation in the Ordos Basin[J].Natural Gas Industry,2023,43(3):46-58.
[2] KHAKSAR A,GRIFFITHS C M.Porosity form sonic log in gas-bearing shaly sandstones:field data versus empirical equations[J].Exploration Geophysics,1998,29(4):440-446.
[3] KENYON W E,DAY P I,STRALEYC,et al.A three-part study of NMR longitudinal relaxation properties of water-saturated sandstones[J].SPE formation evaluation,1988,3(3): 622-636.
[4] COATES G R,GALFORD J,MARDOND,et al.A new characterization of bulk-volume irreducible using magnetic resonance[J].The log analyst,1998,39(1):51-63.
[5] 宋辉,陈伟,李谋杰,等.基于卷积门控循环单元网络的储层参数预测方法[J].油气地质与采收率,2019,26(5):73-78.
SONG Hui,CHEN Wei,LI Moujie,et al.A method to predict reservoir parameters based on convolutional neural network-gated recurrent unit(CNN-GRU)[J].Petroleum Geology & Recovery Efficiency,2019,26(5):73-78.
[6] 谷宇峰,张道勇,鲍志东.测井资料PSO-XGBoost渗透率预测[J].石油地球物理勘探,2021,56(1):26-37.
GU Yufeng,ZHANG Daoyong,BAO Zhidong.Permeability prediction using PSO-XGBoost based on logging data[J].Oil Geophysical Prospecting,2021,56(1):26-37.
[7] 魏国华,韩宏伟,刘浩杰,等.基于半监督高斯混合模型与梯度提升树的砂岩储层相控孔隙度预测[J].石油地球物理勘探,2023,58(1):46-55.
WEI Guohua,HAN Hongwei,LIU Haojie,et al.Facies-controlled porosity prediction of sandstone reservoirs based on semi-supervised Gaussian mixture model and gradient boosting tree[J].Oil Geophysical Prospecting,2023,58(1):46-55.
[8] MAO W,CHEN Z,ZHANG Y,et al.Tensor-DAAD:when tensor meets online early fault detection with transfer learning[J].Measurement,2023,208:112478.
[9] 邵蓉波,肖立志,廖广志,等.基于迁移学习的地球物理测井储层参数预测方法研究[J].地球物理学报,2022,65(2):796-808.
SHAO Rongbo,XIAO Lizhi,LIAO Guangzhi,et al.A reservoir parameters prediction method for geophysical logs based on transfer learning[J].Chinese Journal of Geophysics,2022,65(2):796-808.
[10] 刘美成.致密储层测井评价技术及发展方向[J].特种油气藏,2022,29(4):12-20.
LIU Meicheng.Logging evaluation technology and further development of tight reservoirs[J].Special Oil & Gas Reservoirs,2022,29(4):12-20.
[11] 中华人民共和国自然资源部.石油天然气储量估算规范:DZ/T 0217—2020[S].北京:中国地质出版社,2020:22.
Ministry of Natural Resources of the People′s Republic of China.Regulation of petroleum reserves estimation:DZ/T 0217-2020[S].Beijing:China university of geosciences press,2020:22.
[12] CHAWLA N V,BOWYER K W,HALL L O,et al.SMOTE:synthetic minority over-sampling technique[J].Journal of Artificial Intelligence Research,2002,16:321-357.
[13] BREIMANL.Randomforests[J].Machine learning,2001,45(1):5-32.
[14] 郭伟东,李咸森,王鑫,等.基于迁移学习的隧道凿岩台车钻孔震源实测地震数据去噪方法及应用[J].应用基础与工程科学学报,2023,31(6):1552-1570.
GUO Weidong,LI Xiansen,WANG Xin,et al.Noise reduction method and application of drilling source seismic data for tunnel rock drilling rig based on transfer learning[J].Journal of Basic Science and Engineering,2023,31(6):1552-1570.
[15] 唐晓敏,殷雪松,吕亚娟,等.基于孔隙结构储层分类的中低孔特低渗储层渗透率确定——以B区块S油层为例[J].地球物理学进展,2023,38(1):271-284.
TANG Xiaomin,YIN Xuesong,LYU Yajuan,et al.Determination of permeability of medium-low porosity and extra-low permeability reservoirs based on pore structure reservoir classification:a case study of S Reservoir in Block B[J].Progress in Geophysics,2023,38(1):271-284.
[16] HUANG Q,DOU Q,JIANG Y,et al.An integrated approach to quantify geologic controls on carbonate pore types and permeability,Puguang Gas Field,China[J].Interpretation,2017,5(4): 545-561.
[17] 朱宇.利用测井数据研究泥质含量和孔隙度之间关系[J].价值工程,2011,30(30):94-95.
ZHU Yu.Research of clay volume and porosity based on well logging data[J].Value Engineering,2011,30(30):94-95.
[18] 崔俊峰,杨金路,王民,等.基于随机森林算法的泥页岩孔隙度预测[J].油气地质与采收率,2023,30(6):13-21.
CUI Junfeng,YANG Jinlu,WANG Min,et al.Shale porosity prediction based on random forest algorithm[J].Petroleum Geology & Recovery Efficiency,2023,30(6):13-21.
辽公网安备 21110002000031号 辽ICP备18011648号-1
