Automatic well test interpretation based on convolutional neural network for infinite reservoir

Liu, Xuliang; Li, Daolun; Yang, Jinghai; Zha, Wenshu; Zhou, Ziqi; Gao, Liping; Han, Jiahang

doi:10.1016/j.petrol.2020.107618

Cited by 34 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…XGBoost is an optimally distributed decision gradient boosting model with low computational complexity, flexibility, high accuracy, and portability, which is 10 times faster than the existing popular solutions on a single machine (Liu X. et al, 2020). To reduce the complexity of the model and prevent overfitting, a regularization term is added.…”

Section: Brief Description Of the Four Algorithmsmentioning

confidence: 99%

XGBoost Formation Thickness Identification Based on Logging Data

et al. 2022

View full text Add to dashboard Cite

Based on research on the response mechanism of formation and reservoir response to logging curves, 12 logging curves were selected in combination with formation depth characteristics, and 4 algorithms were used to identify the formation and reservoir: logistic regression (LR), support vector machine (SVM), random forest (RF), and XGBoost. In the study block, 57 wells out of 60 wells were selected for training and learning, and the remaining three wells were used as prediction samples. The recognition of formation thickness and reservoirs is performed by each of these four machine learning algorithms, and predictive knowledge is obtained separately. It was found that the accuracy of the four algorithms for formation thickness and reservoir layer identification reached over 90%, but the XGBoost algorithm was found to be the best in terms of the four scoring criteria of F1-score, precision, recall, and accuracy. The accuracy of formation thickness identification could reach over 95%, and the correlation analysis between the logging curve and formation thickness could be performed on this basis. The results show that RMN, RLLD, and RLLS have the most obvious response to the sandstone layer, off-surface reservoir, and effective thickness layer, while CAL has the least effect on formation and reservoir identification, which can provide an effective reference for the selection and downscaling of subsequent logging curves.

show abstract

Section: Brief Description Of the Four Algorithmsmentioning

confidence: 99%

XGBoost Formation Thickness Identification Based on Logging Data

et al. 2022

View full text Add to dashboard Cite

show abstract

“…In recent years, convolutional neural networks have been widely used in the field of image recognition, including many typical applications in the petroleum industry, such as offshore oil slick detection (Kubat et al, 1998;Corucci et al, 2010), reservoir physical property detection (Ahmadi 2015), using CNN as an automatic well test interpretation approach for infinite acting reservoirs (Liu et al, 2020), and pipeline network internal image detection (Loskutov et al, 2006;Smola et al, 2004). For the problem of dynamometer card classification, the use of convolutional neural networks does not require artificially designed feature extraction methods, and the performance is generally better than that of models such as SVM and BP.…”

Section: Cnn-based Diagnosis Modelmentioning

confidence: 99%

Multi-Scale Normalization Method Combined With a Deep CNN Diagnosis Model of Dynamometer Card in SRP Well

et al. 2022

View full text Add to dashboard Cite

There are more than 20 types of dynamometer card measured of sucker rod pumping (SRP) wells in oil fields, and some working conditions are very complicated. The common diagnosis model of SRP well based on dynamometer card recognition has low accuracy and recall rate of complicated working conditions. In order to improve the accuracy and recall rate of multi-condition diagnosis of SRP well and solve the problem of inseparable data attributes caused by traditional dynamometer card normalization methods, a new dynamometer card preprocessing method is proposed, which uses a clustering analysis algorithm to obtain multiple normalized dynamometer cards of the original dynamometer card and at the same time, adds a set of time-series dynamometer cards to enhance the separability of data. The dynamometer card preprocessing method combined with four deep convolutional neural networks are used to build a diagnosis model. Experiments are conducted under 24 different working conditions, the accuracy of our method is up to 95.8%, and the average recall rate of complicated working conditions is up to 93.1%, which is 13.6 and 35.3% higher than that of the model (AlexNet) built by the traditional preprocessing method. In addition, the preprocessing method of dynamometer card proposed is applicable to all deep learning models and machine learning models. Field applications show that our method is very effective for recalling abnormal working conditions, which is of great significance to the real demand for intelligent diagnosis of SRP well.

show abstract

“…Chen et al [31] presented an efficient semianalytical model for pressure-transient analysis in fractured wells by considering arbitrarily distributed fracture networks. Liu et al [32,33] proposed a discrete fracture-matrix method based on a numerical well testing model to study the pressure transient behavior of discretely distributed natural fractures in a 2D reservoir. Additionally, in their sensitivity analysis, the "dip" on the pressure derivative is an important signal to identify the properties and the impacts of natural fractures.…”

Section: Introductionmentioning

confidence: 99%

A Numerical Model for Pressure Analysis of a Well in Unconventional Fractured Reservoirs

Li²,

Jin

et al. 2023

Energies

View full text Add to dashboard Cite

Fractured reservoirs are highly heterogeneous in both matrix and fracture properties, which results in significant variations in well production. Assessing and quantifying the influence of fractures on fluid flow is essential for developing unconventional reservoirs. The complicated effects of fractures in unconventional fractured reservoirs on fluid flow highly depend on fracture geometry, fracture distribution, and fracture properties, which can be reflected in pressure transient testing. The biggest challenge lies in delineating the pre-existing natural fracture distribution pattern, density, azimuth, and connectivity. Using the advanced finite element method, this paper builds a finely characterized near-wellbore model to numerically simulate the pressure transient testing process in naturally fractured reservoirs and further evaluates fracture-related effects to obtain a more accurate solution. First, the numerical program is benchmarked by the analytical solutions and numerical results of Eclipse. Next, different fracture models with single fractures or fracture networks are set up to investigate the effects of fracture parameters numerically (e.g., fracture location, fracture dip angle, fracture spacing, the ratio of fracture permeability to matrix permeability, fracture network orientation, horizontal fracture distribution, etc.) on pressure transient behaviors in naturally fractured reservoirs. Velocity and pressure profiles are presented to visualize and analyze their effects, and new features in the flow regimes of the derivative plots of the bottom-hole pressure are identified and discussed. Finally, based on geological and geophysical data, including image logs, core descriptions, wireline logs, and seismic and well test data, a practical fractured model of the Dalwogan 2 well in the Surat basin is built, analyzed, and compared with homogenous and measured data. The results show significance in characterizing the complex fracture networks in near-wellbore models of unconventional fractured reservoirs.

show abstract

Automatic well test interpretation based on convolutional neural network for infinite reservoir

Cited by 34 publications

References 12 publications

XGBoost Formation Thickness Identification Based on Logging Data

XGBoost Formation Thickness Identification Based on Logging Data

Multi-Scale Normalization Method Combined With a Deep CNN Diagnosis Model of Dynamometer Card in SRP Well

A Numerical Model for Pressure Analysis of a Well in Unconventional Fractured Reservoirs

Contact Info

Product

Resources

About