摘要
传统的横波速度预测方法包括经验公式法和岩石物理模型法。前者适用于岩石矿物组分相对单一的储层,且受区域限制等因素的影响,不具有普适性,预测精度较低。后者需要根据不同的实际情况选择合适的岩石物理模型,才能达到预期的目的。大多数机器学习横波速度预测方法基于纯数据驱动,数据集的质量和数量将直接决定横波预测模型精度,并缺乏充分的物理内涵。为此,基于深度神经网络(DNN)的方法,假设研究区储层波传播方程的数学形式已知,通过测井数据训练DNN得到未知的弹性参数,以确立目的层的波传播方程。利用平面波分析法得到相应的纵波、横波速度,实现神经网络与理论模型的结合。此外,针对传统Xu-White模型的不足,考虑随深度变化的孔隙纵横比,提出了改进横波速度预测岩石物理模型。利用研究区较丰富的测井数据,分别采用构建的DNN模型和改进横波速度预测岩石物理模型预测横波速度,并与传统的Xu-White模型预测结果进行对比、分析。结果表明,由DNN模型和改进岩石物理模型均可获得较高精度的横波速度预测结果,且前者的预测效果更好。
Conventional shear wave(S-wave)velocity prediction methods include empirical formulas and rock physics model methods.The former is suitable for reservoirs with relatively simple rock mineral compositions,and it is affected by areas and some other factors.Therefore,it is difficult to be widely applied for different for-mations and has low prediction accuracy.The latter requires selecting appropriate rock physics models based on different situations,so as to achieve the expected goals.Most machine learning methods for S-wave velocity prediction aredriven by pure data,and the quality and quantity of the dataset directly determine the accuracy of the S-wave velocity prediction model,which are in lack of sufficient physical insights.Therefore,based on the deep neural network(DNN)methods,this paper assumes that the mathematical form of wave propagation equa-tions for the reservoir in the study area is known,but the elastic parameters are unknown and are learned through a DNN training on the basis of well logging data,so as to establish the wave propagation equations of the target layer.The corresponding compressional wave(P-wave)and S-wave velocities are obtained with the plane wave analysis method to connect the neural networks and the theoretical model.In addition,to address the shortcomings of the conventional Xu-White model,an improved rock physicsmodel for S-wave velocity pre-diction is proposed by considering the pore aspect ratio varying with depth.By using the adequate well logging data in the study area,the established DNN model and the improved rock physics model for S-wave velocity prediction are used to predict the S-wave velocity,and the results are compared with the conventional Xu-White model.It shows that both the DNN model and the improved rock physics model can help obtain high-precision S-wave velocity prediction results,and the former has better prediction performances.
作者
方志坚
巴晶
熊繁升
杨志芳
晏信飞
阮传同
FANG Zhijian;BA Jing;XIONG Fansheng;YANG Zhifang;YAN Xinfei;RUAN Chuantong(School of Earth Sciences and Engineering,Hohai University,Nanjing,Jiangsu 211100,China;Yanqi Lake Beijing Institute of Mathematical Sciences and Applications,Beijing 101408,China;China National Petroleum Corporation Exploration and Development Research Institute,Beijing 100083,China;China National Petroleum Corporation Key Laboratory of Geophysics,Beijing 100083,China)
出处
《石油地球物理勘探》
EI
CSCD
北大核心
2024年第3期381-391,共11页
Oil Geophysical Prospecting
基金
国家自然科学基金项目“页岩油储层多尺度岩石物理模型及参数预测方法研究”(42174161)和“基于微观孔隙结构特征构建致密砂岩衰减岩石物理模型”(41974123)
中国石油天然气集团有限公司科技项目“油气藏精细描述与剩余油分布地球物理预测方法”(2023ZZ0504)
江苏省科技计划青年基金项目“基于多尺度衰减岩石物理模型的页岩油储层孔裂隙特征和黏土含量定量预测研究”(BK20220995)联合资助。
关键词
深度神经网络
岩石物理模型
页岩油层系
储层参数
横波速度
孔隙纵横比
deep neural network
rock physics model
shale oil formations
reservoir parameters
S-wave velocity
pore aspect ratio
