Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Ultradeep carbonate reservoir lithofacies classification based on a deep convolutional neural network — A case study in the Tarim Basin, China
Lu, Shengyu (China University of Geosciences) | Cai, Chuyang (Monash University) | Zhong, Zhi (China University of Geosciences, China University of Geosciences) | Cai, Zhongxian (China University of Geosciences, China University of Geosciences) | Guo, Xu (China University of Geosciences) | Zhang, Heng (China University of Geosciences, China University of Geosciences) | Li, Jie (China University of Geosciences, China University of Geosciences)
Abstract Lithofacies identification is essential in reservoir evaluation, especially in ultradeep carbonate reservoirs. In general, coring samples are the best sources to identify carbonate lithofacies because they are taken directly from reservoirs. However, the core is expensive to obtain, and generally its availability is greatly limited. In recent years, deep learning has attracted enormous attention because of its robust nonlinear regression and classification ability. This study applies a deep-learning algorithm to identify the lithofacies using geophysical well-log data. Six types of well-log data, such as natural gamma ray, density (DEN), neutron porosity (CNL), acoustic (AC), and shallow and deep lateral resistivity well logs (RT/RXO), are smoothed by the average sliding method and converted to 2D data. Then, the 2D data are treated as inputs to predict the carbonate lithofacies through the convolutional neural network (CNN). The results indicate that the prediction accuracy rate is 90.2%. This indicates that the CNN can identify different carbonate lithofacies well.
- Asia > China (1.00)
- North America > United States > Colorado > Garfield County (0.28)
- South America > Brazil > Brazil > South Atlantic Ocean > Santos Basin (0.99)
- North America > United States > Colorado > Piceance Basin > Williams Fork Formation (0.99)
- North America > United States > Colorado > Piceance Basin > Greater Grand Valley Field Complex Field > Williams Fork Formation (0.99)
- (20 more...)
- Reservoir Description and Dynamics > Reservoir Characterization > Exploration, development, structural geology (1.00)
- Reservoir Description and Dynamics > Formation Evaluation & Management > Open hole/cased hole log analysis (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Neural networks (1.00)
Ultradeep carbonate reservoir lithofacies classification based on a deep convolutional neural network — A case study in the Tarim Basin, China
Lu, Shengyu (China University of Geosciences) | Cai, Chuyang (Monash University) | Zhong, Zhi (China University of Geosciences, China University of Geosciences) | Cai, Zhongxian (China University of Geosciences, China University of Geosciences) | Guo, Xu (China University of Geosciences) | Zhang, Heng (China University of Geosciences, China University of Geosciences) | Li, Jie (China University of Geosciences, China University of Geosciences)
Abstract Lithofacies identification is essential in reservoir evaluation, especially in ultradeep carbonate reservoirs. In general, coring samples are the best sources to identify carbonate lithofacies because they are taken directly from reservoirs. However, the core is expensive to obtain, and generally its availability is greatly limited. In recent years, deep learning has attracted enormous attention because of its robust nonlinear regression and classification ability. This study applies a deep-learning algorithm to identify the lithofacies using geophysical well-log data. Six types of well-log data, such as natural gamma ray, density (DEN), neutron porosity (CNL), acoustic (AC), and shallow and deep lateral resistivity well logs (RT/RXO), are smoothed by the average sliding method and converted to 2D data. Then, the 2D data are treated as inputs to predict the carbonate lithofacies through the convolutional neural network (CNN). The results indicate that the prediction accuracy rate is 90.2%. This indicates that the CNN can identify different carbonate lithofacies well.
- Asia > China (1.00)
- North America > United States > Colorado > Garfield County (0.28)
- South America > Brazil > Brazil > South Atlantic Ocean > Santos Basin (0.99)
- North America > United States > Colorado > Piceance Basin > Williams Fork Formation (0.99)
- North America > United States > Colorado > Piceance Basin > Greater Grand Valley Field Complex Field > Williams Fork Formation (0.99)
- (20 more...)
- Reservoir Description and Dynamics > Reservoir Characterization > Exploration, development, structural geology (1.00)
- Reservoir Description and Dynamics > Formation Evaluation & Management > Open hole/cased hole log analysis (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Neural networks (1.00)
Integrated Carbonate Lithofacies Modeling Based on the Deep Learning and Seismic Inversion and its Application
Xin, Chen (BGP, CNPC, Zhuozhou, Hebei, China) | Deshuang, Chang (BGP, CNPC, Zhuozhou, Hebei, China) | Qing, Liu (Dagang Oilfield, CNPC, Dagang, Tianjing, China) | Qian, Sun (BGP, CNPC, Zhuozhou, Hebei, China) | Weixiang, Zhong (BGP, CNPC, Zhuozhou, Hebei, China) | Wenwen, Yu (BGP, CNPC, Zhuozhou, Hebei, China) | Xiaoliang, Li (BGP, CNPC, Zhuozhou, Hebei, China) | Hanzhou, Fan (BGP, CNPC, Zhuozhou, Hebei, China) | Qingning, Yang (BGP, CNPC, Zhuozhou, Hebei, China) | Dengyi, Xiao (BGP, CNPC, Zhuozhou, Hebei, China) | Fuli, An (BGP, CNPC, Zhuozhou, Hebei, China) | bo, Wang (BGP, CNPC, Zhuozhou, Hebei, China) | Lu, Lv (BGP, CNPC, Zhuozhou, Hebei, China) | Yu, Peng (BGP, CNPC, Zhuozhou, Hebei, China) | Qiang, Liu (BGP, CNPC, Zhuozhou, Hebei, China) | Kongzhi, Huang (BGP, CNPC, Zhuozhou, Hebei, China)
Abstract To improve the accuracy of carbonate lithofacies modeling, mainly well data such as core, thin section and well logging data had been adopted in conventional methods. Although the reservoir types classification is very detailed, but it is usually difficult to integrate with seismic data to make 3D lithofacies model. To address the issue of carbonate lithofacies modeling, a new integrated carbonate lithofacies modeling technique was summarized based on thin section, core, well logging, 3D seismic data and production performance data. The integrated carbonate lithofacies modeling workflow mainly contains 5 steps. 1) Integrated lithofacies classification based on the core, thin section, well logging, FMI, CMR and production performance data. 2) Petrophysics lithofacies classification based on the cross-plot analysis between sensitive well log curves. 3) Petrophysics lithofacies prediction based on the sensitive well log curve by deep learning method, and verification by core lithofacies analysis. 4) Seismic inversion volume optimization by well lithofacies calibration. 5) Lithofacies modelling based seismic inversion based on the seismic inversion cut-off analysis (Fig.2). This workflow integrated seismic impedance (continuous variable) with lithofacies (discrete variable), and converts seismic inversion into lithofacies directly. According to the certification of new wells, this technique had been applied successfully in carbonate reservoir of M oil field in Middle East, it not only improves the accuracy of 1D lithofacies prediction for wells by deep learning method, but also improves the accuracy of 3D lithofacies modeling for the whole oilfield by well and seismic inversion integrated. The lithofacies modeling not only matched with lithofacies from core analysis and petrophysics lithofacies prediction from well log analysis, but also matched with seismic inversion data in no well area. The integrated carbonate lithofacies modeling workflow integrated thin section, core, well logging, 3D seismic data and production performance data, and improved the improves the accuracy of 3D lithofacies modeling for no well area. It’s useful for new wells optimization and high efficiency development with lower cost. The integrated carbonate lithofacies modeling workflow not only suit for carbonate reservoir, but also suit for clastic reservoir.
- Asia > China (0.70)
- Asia > Middle East (0.69)
- Asia > China > Tianjin > Bohai Basin > Huanghua Basin > Dagang Field (0.99)
- Asia > Middle East > Iraq > Basra Governorate > Arabian Basin > Widyan Basin > Mesopotamian Basin > West Quma Field > Yamama Formation (0.98)
Petrophysical characteristics and log identification of lacustrine shale lithofacies: A case study of the first member of Qingshankou Formation in the Songliao Basin, Northeast China
Liu, Bo (Northeast Petroleum University) | Zhao, Xiaoqing (Northeast Petroleum University) | Fu, Xiaofei (Northeast Petroleum University) | Yuan, Baiyan (Daqing Drilling & Exploration Engineering Corporation) | Bai, Longhui (Northeast Petroleum University) | Zhang, Yuwei (Northeast Petroleum University) | Ostadhassan, Mehdi (Northeast Petroleum University)
Abstract As an unconventional resource, shale reservoirs recently have attracted considerable attention in the petroleum industry. Shale plays are highly heterogenous vertically and laterally and are characterized by rapid changes in mineral composition. Thus, identification of dominant lithofacies is a key issue in the development of shale oil and gas reservoirs. In this study, various existing lithofacies in a shale section as a target unit in the Qingshankou Formation are divided based on organic matter content, sedimentary structure, and mineral composition. To delineate the electrofacies from the log, the multiresolution graph-based clustering (MRGC) is used to optimize the conventional logs that are sensitive to the electrofacies clustering analyses. Based on the principle of lithofacies identification, the electrofacies are artificially related to the lithofacies as well. This was done by analyzing the petrophysical characteristics of various shale lithofacies, to enable obtaining the main log parameters for the facies of the lacustrine shale section understudy. The results showed that by considering the underlying geologic criterion of each lithofacies, the MRGC method is able to correlate geophysical characteristics of each identified electrofacies for an optimal selection of six lithofacies.
- Asia > China (1.00)
- North America > United States > Texas (0.93)
Heterogeneous Domain Adaptation Framework for Logging Lithofacies Identification
Ren, Quan (School of Earth Sciences and Engineering, Hohai University) | Zhang, Hongbing (School of Earth Sciences and Engineering, Hohai University (Corresponding author)) | Yu, Xiang (Design & Consulting Corp., Nanjing Hydraulic Research Institute) | Zhang, Dailu (School of Earth Sciences and Engineering, Hohai University) | Zhao, Xiang (School of Earth Sciences and Engineering, Hohai University) | Zhu, Xinyi (School of Earth Sciences and Engineering, Hohai University) | Hu, Xun (College of Geosciences, China University of Petroleum (Beijing))
Summary Reservoir lithofacies type is an important indicator of reservoir quality and oiliness, and understanding lithofacies type can help geologists and engineers make informed decisions about exploration and development activities. The use of well-log data to establish machine learning models for lithofacies identification has gained popularity; however, the assumption that data are independent identical distribution followed by these models is often unrealistic. Additionally, there is a possible incompatibility between the training and test data in terms of feature space dimensions. We propose the heterogeneous domain adaptation framework for logging lithofacies identification (HDAFLI) to address these problems. The framework comprises three main contributions: (i) The denoising autoencoder feature mapping (DAFM) module is adopted to resolve the incompatibility issue in feature space between training and test data. The connection between training and test data can be effectively established to improve the performance and generalization ability. (ii) The transferability and discriminative joint probability distribution adaptive (TDJPDA) module addresses the issue of data distribution differences. It improves the transferability of training and test data by minimizing the maximum mean difference (MMD) of the joint probabilities of the source and target domains and enhances their discriminative ability by maximizing the joint probability MMD of different lithofacies categories. (iii) Bayesian optimization is used to optimize hyperparameters in the light gradient boosting machine (LightGBM) model for high computational efficiency in determining the best accuracy. We selected well-logging data from eight wells in the Pearl River Mouth Basin of the South China Sea to design four tasks and compared HDAFLI with various baseline machine learning algorithms and baseline domain adaptive algorithms. The results show that HDAFLI has the highest average accuracy among the four tasks. It is 19.76% and 8.94% higher than the best-performing baseline machine learning algorithm and baseline domain adaptive method among the comparison algorithms, respectively. For HDAFLI, we also conducted ablation experiments, time cost and convergence performance analysis, parameter sensitivity experiments, and feature visualization experiments. The results of ablation experiments show that the three modules of HDAFLI all play an active role, working together to achieve the best results. In addition, HDAFLI has a reasonable time cost, can become stable after several iterations, and has good convergence performance. The results of parameter sensitivity experiments confirm that the accuracy of HDAFLI does not change significantly with changes in hyperparameters, which is robust. The results of feature visualization experiments show that the data of the training set and the test set are concentrated together to a certain extent, which indicates that HDAFLI has completed the task of data distribution alignment very well. The findings of this study can help for a better understanding of how to address the challenge of reservoir lithofacies identification through a heterogeneous domain adaptation framework. By solving the problem of feature space incompatibility and data distribution difference between training data and test data, the application of HDAFLI provides geologists and engineers with more accurate lithofacies classification tools. This study has practical application value for reservoir quality assessment, oiliness prediction, and exploration and development decision-making.
- North America > United States (1.00)
- Europe (0.93)
- Asia > China > South China Sea (0.24)
- Geology > Sedimentary Geology > Depositional Environment (1.00)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock (1.00)
- Asia > China > South China Sea > Zhujiangkou Basin (0.99)
- North America > United States > Louisiana > China Field (0.95)
- North America > United States > New Mexico > Permian Basin > Pearl Field > Queen Formation (0.94)
- Europe > Norway > North Sea > Central North Sea > South Viking Graben > PL 046 > Paleocene Formation (0.94)