提出一种基于卷积神经网络的径向复合油藏自动试井解释方法,并利用现场实测数据验证其有效性和准确性。采用对数函数进行数据变换,采用均方误差作为损失函数,利用“dropout”方法避免过拟合,通过不断减小损失函数进行网络优化,得到最优的卷积神经网络。训练好的最优网络可直接用于解释径向复合油藏中井的压力恢复或压力降落数据,将给定的实测压力变化及其导数数据的双对数图输入到网络中,即可输出对应的油藏参数(流度比、储容比、无因次复合半径以及表征井储和表皮效应的无因次组),从而实现了试井参数解释的自动初拟合。利用大庆油田现场实测数据对该方法进行了验证,研究表明,该方法具有很高的解释精度,且优于解析法和最小二乘法。图12表6参29
An automatic well test interpretation method for radial composite reservoirs based on convolutional neural network (CNN) is proposed, and its effectiveness and accuracy are verified using actual field data. In this paper, Based on the data transformed by logarithm function and the loss function of mean square error (MSE), the optimal CNN is obtained by reducing the loss function to optimize the network with "dropout" method to avoid over fitting. The trained optimal network can be directly used to interpret the buildup or drawdown pressure data of the well in the radial composite reservoir, that is, the log-log plot of the given measured pressure variation and its derivative data are input into the network, the outputs are corresponding reservoir parameters (mobility ratio, storativity ratio, dimensionless composite radius, and dimensionless group characterizing well storage and skin effects), which realizes the automatic initial fitting of well test interpretation parameters. The method is verified with field measured data of Daqing Oilfield. The research shows that the method has high interpretation accuracy, and it is superior to the analytical method and the least square method.
[1] MOHAGHEGH S D.Reservoir simulation and modeling based on artificial intelligence and data mining (AI & DM)[J]. Journal of Natural Gas Science and Engineering, 2011, 3(6): 697-705.
[2] ESMAILI S, MOHAGHEGH S D.Full field reservoir modeling of shale assets using advanced data-driven analytics[J]. Geoscience Frontiers, 2016, 7(1): 11-20.
[3] AKIN S, KOK M V, URAZ I.Optimization of well placement geothermal reservoirs using artificial intelligence[J]. Computers & Geosciences, 2010, 36(6): 776-785.
[4] PANJA P, VELASCO R, PATHAK M, et al.Application of artificial intelligence to forecast hydrocarbon production from shales[J]. Petroleum, 2018, 4(1): 75-89.
[5] 李道伦, 卢德唐, 孔祥言, 等. BP神经网络隐式法在测井数据中的应用研究[J]. 石油学报, 2007, 28(3): 105-108.
LI Daolun, LU Detang, KONG Xiangyan, et al.Processing of well log data based on backpropagation neural network implicit approximation[J]. Acta Petrolei Sinica, 2007, 28(3): 105-108.
[6] LI D, LU D, ZHA W.Implicit approximation of neural network and applications[J]. SPE Reservoir Evaluation & Engineering, 2009, 12(6): 921-928.
[7] ASADISAGHANDI J, TAHMASEBI P.Comparative evaluation of back-propagation neural network learning algorithms and empirical correlations for prediction of oil PVT properties in Iran oilfields[J]. Journal of Petroleum Science & Engineering, 2011, 78(2): 464-475.
[8] ENAB K, ERTEKIN T.Artificial neural network based design for dual lateral well applications[J]. Journal of Petroleum Science & Engineering, 2014, 123: 84-95.
[9] SINGH S, KANLI A I, SEVGEN S.A general approach for porosity estimation using artificial neural network method: A case study from Kansas gas field[J]. Studia Geophysica et Geodaetica, 2015, 60(1): 1-11.
[10] MEMON P Q, YONG S P, PAO W, et al.Dynamic well bottom-hole flowing pressure prediction based on radial basis neural network[J]. Studies in Computational Intelligence, 2015, 591: 279-292.
[11] KIM C, KIM Y, SHIN C, et al.A comprehensive approach to select completion and fracturing fluid in shale gas reservoirs using the artificial neural network[J]. Environmental Earth Sciences, 2017, 76(20): 671.
[12] CHOUBINEH A, HELALIZADEH A, WOOD D A.Estimation of minimum miscibility pressure of varied gas compositions and reservoir crude oil over a wide range of conditions using an artificial neural network model[J]. Advances in Geo-Energy Research, 2019, 3(1): 52-66.
[13] ATHICHANAGORN S, HORNE R N.Automatic parameter estimation from well test data using artificial neural network[R]. SPE 30556, 1995.
[14] 邓远忠, 陈钦雷. 试井解释图版拟合分析的神经网络方法[J]. 石油勘探与开发, 2000, 27(1): 64-66.
DENG Yuanzhong, CHEN Qinlei.Using artificial neural network to realize type curve match analysis of well test data[J]. Petroleum Exploration and Development, 2000, 27(1): 64-66.
[15] JEIRANI Z, MOHEBBI A.Estimating the initial pressure, permeability and skin factor of oil reservoirs using artificial neural networks[J]. Journal of Petroleum Science & Engineering, 2006, 50(1): 11-20.
[16] ADIBIFARD M, TABATABAEI-NEJAD S A R, KHODAPANAH E. Artificial Neural Network (ANN) to estimate reservoir parameters in Naturally Fractured Reservoirs using well test data[J]. Journal of Petroleum Science and Engineering, 2014, 122: 585-594.
[17] GHAFFARIAN N, ESLAMLOEYAN R, VAFERI B.Model identification for gas condensate reservoirs by using ANN method based on well test data[J]. Journal of Petroleum Science and Engineering, 2014, 123: 20-29.
[18] TIAN C, HORNE R N.Recurrent neural networks for permanent downhole gauge data analysis[R]. SPE 187181-MS, 2017.
[19] SUDAKOV O, BURNAEV E, KOROTEEV D.Driving digital rock towards machine learning: Predicting permeability with gradient boosting and deep neural networks[J]. Computers & Geosciences, 2019, 127: 91-98.
[20] ZHA W, LI X, XING Y, et al.Reconstruction of shale image based on Wasserstein Generative Adversarial Networks with gradient penalty[J]. Advances in Geo-Energy Research, 2020, 4(1): 107-114.
[21] 张东晓, 陈云天, 孟晋. 基于循环神经网络的测井曲线生成方法[J]. 石油勘探与开发, 2018, 45(4): 598-607.
ZHANG Dongxiao, CHEN Yuntian, MENG Jin.Synthetic well logs generation via Recurrent Neural Networks[J]. Petroleum Exploration and Development, 2018, 45(4): 598-607.
[22] WAIBEL A.Phoneme recognition using time-delay neural networks[R]. Tokyo, Japan: Meeting of the Institute of Electrical, Information and Communication Engineers (IEICE), 1987.
[23] 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.
[24] KRIZHEVSKY A, SUTSKEVER I, HINTON G E.ImageNet classification with deep convolutional neural networks[C]// Proceedings of the 25th International Conference on Neural Information Processing Systems. New York: Curran Associates Inc., 2012: 1097-1105.
[25] ALQAHTANI N, ARMSTRONG R T, MOSTAGHIMI P.Deep learning convolutional neural networks to predict porous media properties[R]. SPE 191906-MS, 2018.
[26] BOURDET D, GRINGARTEN A C.Determination of fissure volume and block size in fractured reservoirs by type-curve analysis[R]. SPE 9293, 1980.
[27] BOURDET D, ALAGOA A, AYOUB J A, et al.New type curves aid analysis of fissured zone well tests[J]. World Oil, 1984, 198(5): 111-124.
[28] GOODFELLOW I, BENGIO Y, COURVILLE A.Deep learning[M]. Cambridge, MA, USA: MIT Press, 2016.
[29] BOURDET D.Well test analysis: The use of advanced interpretation models[M]. New York: Elsevier, 2002.