S-wave velocity prediction method for sand-shale formation based on quadratic optimization network
SHAN Bo1, ZHANG Fanchang1, DING Jicai2
1. School of Geosciences, China University of Petroleum(East China), Qingdao, Shandong 266580, China; 2. CNOOC Research Institute Co., Ltd., Beijing 100027, China
Abstract:Shear wave (S-wave) velocity is an important parameter that reflects the lithological characteristics of a reservoir. However, it is often absent in actual logging data. In this paper, an end-to-end quadratic optimization network is constructed according to the relationships of S-wave velocity with other parameters. Instead of solving intermediate parameters, it uses the gamma ray, porosity, and compressional wave (P-wave) velocity to predict the S-wave velocity directly. The quadratic optimization algorithm is applied in the network training process to replace the Adam algorithm and achieves higher accuracy and efficiency. In addition, the orthogonal experiment is used to analyze the influences of different parameters and training strategies (including optimization algorithm, number of network layers, and number of training wells) on the prediction of the S-wave velocity. The results show that the optimization algorithm has the greatest impact on the prediction. The quadratic optimization algorithm has a better prediction effect and higher efficiency than those of the Adam optimization algorithm. A suitable activation function has a positive effect on the prediction. According to the experiment results, the optimal network parameters and training strategy are selected for S-wave velocity prediction. The prediction results on the test set show that the method can predict the S-wave velocity accurately and effectively.
郭栋, 印兴耀, 吴国忱.横波速度计算方法与应用[J].石油地球物理勘探, 2007, 42(5):535-538.GUO Dong, YIN Xingyao, WU Guochen.Computational approach of S-wave velocity and application[J].Oil Geophysical Prospecting, 2007, 42(5):535-538.
[2]
孙福利, 杨长春, 麻三怀, 等.横波速度预测方法[J].地球物理学进展, 2008, 23(2):470-474.SUN Fuli, YANG Changchun, MA Sanhuai, et al.An S-wave velocity predicted method[J].Progress in Geophysics, 2008, 23(2):470-474.
[3]
GREENBERG M L, CASTAGNA J P.Shear-wave velocity estimation in porous rocks:theoretical formulation, preliminary verification, and applications[J].Geophysical Prospecting, 1992, 40(2):195-209.
[4]
CASTAGNA J P, BATALE M L, EASTWOOD R L.Relationships between compressional-wave and shear-wave velocities in clastic silicate rocks[J].Geophy-sics, 1985, 50(4):571-581.
[5]
XU S Y, WHITE R E.A new velocity model for clay-sand mixtures[J].Geophysical Prospecting, 1995, 43(1):91-118.
[6]
XU S Y, WHITE R E.A physical model for shear-wave velocity prediction[J].Geophysical Prospecting, 1996, 44(4):687-717.
[7]
XU S Y, PAYNE M A.Modeling elastic properties in carbonate rocks[J].The Leading Edge, 2009, 28(1):66-74.
[8]
PAYNE S S, WILD P, LUBBE R.An integrated solution to rock physics modeling in fractured carbonate reservoirs[C].SEG Technical Program Expanded Abstracts, 2010, 29:358-362.
[9]
张秉铭, 刘致水, 刘俊州, 等.鄂尔多斯盆地北部复杂碳酸盐岩横波速度预测研究[J].石油物探, 2017, 56(3):328-337.ZHANG Bingming, LIU Zhishui, LIU Junzhou, et al.An improved S-wave velocity prediction method for complex carbonate reservoir in North Ordos Basin, China[J].Geophysical Prospecting for Petroleum, 2017, 56(3):328-337.
[10]
张元中, 周开金, 赵建斌, 等.砂泥岩地层横波测井曲线预测方法研究[J].石油物探, 2012, 51(5):508-514.ZHANG Yuanzhong, ZHOU Kaijin, ZHAO Jianbin, et al.Shear-wave logging curve prediction method for shaly sand formation[J].Geophysical Prospecting for Petroleum, 2012, 51(5):508-514.
[11]
罗水亮, 杨培杰, 胡光明, 等.基于变形P-L模型的矩阵方程迭代精细横波预测[J].地球物理学报, 2016, 59(5):1839-1848.LUO Shuiliang, YANG Peijie, HU Guangming, et al.S-wave velocity prediction based on the modified P-L model and matrix equation iteration[J].Chinese Journal of Geophysics, 2016, 59(5):1839-1848.
[12]
唐杰, 王浩, 姚振岸, 等.基于岩石物理诊断的横波速度估算方法[J].石油地球物理勘探, 2016, 51(3):537-543.TANG Jie, WANG Hao, YAO Zhen'an, et al.Shear wave velocity estimation based on rock physics diagnosis[J].Oil Geophysical Prospecting, 2016, 51(3):537-543.
[13]
林凯, 贺振华, 熊晓军, 等.基于基质矿物模量自适应提取横波速度反演方法[J].石油地球物理勘探, 2013, 48(2):262-267.LIN Kai, HE Zhenhua, XIONG Xiaojun, et al.S-wave velocity inversion based on adaptive extraction of matrix mineral modulus[J].Oil Geophysical Prospecting, 2013, 48(2):262-267.
[14]
张繁昌, 刘汉卿, 钮学民, 等.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.ZHANG Fanchang, LIU Hanqing, NIU Xuemin, et al.High resolution seismic inversion by convolutional neural network[J].Oil Geophysical Prospecting, 2014, 49(6):1165-1169.
[15]
桑凯恒, 张繁昌.基于模糊粗糙集的机器学习储层参数预测[J].CT理论与应用研究, 2018, 27(4):455-464.SANG Kaiheng, ZHANG Fanchang.Prediction of re-servoir parameters of machine learning based on fuzzy rough set[J].CT Theory and Applications, 2018, 27(4):455-464.
[16]
徐旺林, 张繁昌, 郭彦如, 等.贝叶斯全频信息同时反演在鄂尔多斯盆地碳酸盐岩储层预测中的应用[J].石油地球物理勘探, 2014, 49(4):723-728.XU Wanglin, ZHANG Fanchang, GUO Yanru, et al.Bayesian full frequency simultaneous inversion for carbonate reservoir in Erdos Basin[J].Oil Geophysical Prospecting, 2014, 49(4):723-728.
[17]
ZHANG Z D, ALKHALIFAH T.Regularized elastic full-waveform inversion using deep learning[J].Geo-physics, 2019, 84(5):741-751.
[18]
YANG F S and MA J W, Deep-learning inversion:A next-generation seismic velocity model building me-thod[J].Geophysics, 2019, 84(4):583-599.
[19]
RUMELHART D, HINTON G, WILLIAMS R.Lear-ning representations by back-propagating errors[J].Nature, 1986, 323(9):533-536.
[20]
吕晶, 谢润成, 周文, 等.LM-BP神经网络在泥页岩地层横波波速拟合中的应用[J].中国石油大学学报(自然科学版), 2017, 41(3):75-83.LYU Jing, XIE Runcheng, ZHOU Wen, et al.Application of LM-BP neural network in simulation of shear wave velocity of shale formation[J].Journal of China University of Petroleum (Edition of Natural Science), 2017, 41(3):75-83.
[21]
孙宇航, 刘洋.利用GRU神经网络预测横波速度[J].石油地球物理勘探, 2020, 55(3):484-492.SUN Yuhang, LIU Yang.Prediction of S-wave velo-city based on GRU neural network[J].Oil Geophysical Prospecting, 2020, 55(3):484-492.
[22]
KINGMA D P, BA J L.Adam:A method for stochastic optimization[C].International Conference on Learning Representations, 2015, 1-13.
[23]
常兆光, 王清河, 杜彩凤.应用统计方法[M].北京:石油工业出版社,2009.CHANG Zhaoguang, WANG Qinghe, DU Caifeng.Applied Statistical Methods[M].Petroleum Industry Press, Beijing, 2009.
[24]
SHEWCHUK J R.An Introduction to the Conjugate Gradient Method Without the Agonizing Pain[R].School of Computer Science, Carnegie Mellon University, PA United States, 1994.