摘要
储层参数是储层评价的一项重要内容。针对传统储层预测方法难以摆脱线性方程的束缚及预测精度不高的问题,将卷积神经网络与门控循环单元网络相结合,提出了卷积门控循环单元网络模型。该模型不仅具备卷积神经网络局部感知的特性,还具备门控循环单元网络长期记忆的功能,从而具有表达数据时空特征的能力。基于某井区A井已知井段测井资料建立卷积门控循环单元网络孔隙度预测模型,预测该井区未知深度段的孔隙度,并提出变学习率训练方法。实验证明,与单一的卷积神经网络模型、门控循环单元网络模型相比,卷积门控循环单元网络模型能够更有效地提取数据特征,预测精度更高,可为储层参数的预测提供新的思路。
Reservoir parameters are important for reservoir evaluation. Aiming at the difficulties of the traditional reservoir parameters prediction method to get rid of the constraint of a linear equation and the low prediction accuracy,a model com bined with convolutional neural network( CNN) and gated recurrent uni(t GRU) is proposed. The model not only has the lo cal perception characteristics of CNN but also has the long-term memory function of GRU,thus having the ability to ex press the spatio-temporal features of data. The CNN-GRU porosity prediction model is established based on the well log ging data of Well A to predict the porosity of unknown depth segment in this well area,and to further propose a variable learning rate training method. Compared with CNN or GRU models,experimental results show that CNN-GRU model can extract data features more effectively and can improve the reservoir parameters prediction accuracy,which provides a new idea to predict reservoir parameters.
作者
宋辉
陈伟
李谋杰
王浩懿
SONG Hui;CHEN Wei;LI Moujie;WANG Haoyi(Key Laboratory of Exploration Technologies for Oil and Gas Resources,Ministry of Education,Yangtze University,Wuhan City,Hubei Province,430100,China;Institute of Geophysics and Petroleum Resources,Yangtze University,Wuhan City,Hubei Province,430100,China;Hubei Cooperative Innovation Center of Unconventional Oil and Gas,Yangtze University,Wuhan City,Hubei Province,430100,China;School of Materials Science and Engineering,Taiyuan University of Technology,Taiyuan City,Shanxi Province,030024,China)
出处
《油气地质与采收率》
CAS
CSCD
北大核心
2019年第5期73-78,共6页
Petroleum Geology and Recovery Efficiency
基金
国家自然科学基金项目“基于经验模态分解的自由表面多次波衰减方法研究”(41804140)
湖北省教育厅指导性项目“基于地震数据结构的高分辨率油藏识别方法研究”(B2018556)
油气资源与勘探技术教育部重点实验室(长江大学)“地球物理信息探测方法与技术”(PI2018-02)
关键词
储层参数预测
孔隙度
深度学习
卷积神经网络
循环神经网络
门控循环单元网络
reservoir parameter prediction
porosity
deep learning
convolutional neural network
recurrent neural net work
gated recurrent unit
