Determination of oil well placement using convolutional neural network coupled with robust optimization under geological uncertainty

Kwon, Seoyoon; Park, Gayoung; Jang, Young Ho; Cho, Jinhyung; Chu, M. G.; Min, Baehyun

doi:10.1016/j.petrol.2020.108118

Cited by 35 publications

(3 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Being different from traditional ANN, CNN can directly capture the spatial features from images to improve both prediction accuracy and efficiency [31]. The CNN has been widely employed in diverse fields such as computer vision including image classification [32], object tracking, visual salience detection, action recognition; natural language processing [33] and time series classification and forecasting [22], [30], [34].…”

Section: B Convolutional Neural Network (Cnn)mentioning

confidence: 99%

“…In the convolutional layer, a neuron is connected to only a portion of the neighboring neurons, which reduces the complexity of the network and the number of parameters [30]. This layer is used to extract the features of input images by applying a set of filters to the input data, producing a set of output feature maps [34]. As observed in Equation 6, XVI Brazilian Conference on Computational Intelligence (CBIC 2023), Salvador, October 8th to 11th which represents the convolutional layer equation and filters are applied to the input matrix.…”

Section: B Convolutional Neural Network (Cnn)mentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of the nominal thickness of pipelines using neural networks through the dispersion curves drawn with the SAFE method

Boechat,

Lavor,

Sesini

et al. 2023

Anais Do XVI Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

Inspection through non-destructive testing is important for the analysis of oil well structures in different phases of the project, whether during execution, during the production period, or at closure. In this context, the propagation of ultrasonic waves is an important ally to assessing the integrity of structural components, through the tracing of dispersion curves, which show the pair frequency x wavelength for a specific waveguide. This study presents a neural network framework for estimating the percentage of nominal pipeline thickness. The neural network inputs are dispersion curves obtained by simulation, using the semi-analytical finite element method (SAFE), that presents a computational cost reduced compared with the totally numerical simulation. In this way, 100 samples were generated. The case studied consists of a hollow cylinder with different thickness sizes. From the results, it is concluded that the adopted methodology is efficient to predict the percentage of nominal pipeline thickness.

show abstract

Section: B Convolutional Neural Network (Cnn)mentioning

confidence: 99%

Section: B Convolutional Neural Network (Cnn)mentioning

confidence: 99%

Evaluation of the nominal thickness of pipelines using neural networks through the dispersion curves drawn with the SAFE method

Boechat,

Lavor,

Sesini

et al. 2023

Anais Do XVI Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

show abstract

“…All outputs showed a small error with less than 0.7 for both validation and testing sets. Kwon et al [29] applied 2D-CNN to determine the location of an oil well under geological uncertainty by predicting the cumulative oil production of reservoir simulation. The performance of the 2D-CNN model was compared with a shallow machine learning model, ANN, and the results showed that the 2D-CNN model achieved 88% accuracy with a relative error of 0.035.…”

Section: Introductionmentioning

confidence: 99%

One-Dimensional Convolutional Neural Network with Adaptive Moment Estimation for Modelling of the Sand Retention Test

et al. 2021

View full text Add to dashboard Cite

Stand-alone screens (SASs) are active sand control methods where compatible screens and slot sizes are selected through the sand retention test (SRT) to filter an unacceptable amount of sand produced from oil and gas wells. SRTs have been modelled in the laboratory using computer simulation to replicate experimental conditions and ensure that the selected screens are suitable for selected reservoirs. However, the SRT experimental setups and result analyses are not standardized. A few changes made to the experimental setup can cause a huge variation in results, leading to different plugging performance and sand retention analysis. Besides, conducting many laboratory experiments is expensive and time-consuming. Since the application of CNN in the petroleum industry attained promising results for both classification and regression problems, this method is proposed on SRT to reduce the time, cost, and effort to run the laboratory test by predicting the plugging performance and sand production. The application of deep learning has yet to be imposed in SRT. Therefore, in this study, a deep learning model using a one-dimensional convolutional neural network (1D-CNN) with adaptive moment estimation is developed to model the SRT with the aim of classifying plugging sign (screen plug, the screen does not plug) as well as to predict sand production and retained permeability using a varying sand distribution, SAS, screen slot size, and sand concentration as inputs. The performance of the proposed 1D-CNN model for the slurry test shows that the prediction of retained permeability and the classification of plugging sign achieved robust accuracy with more than a 90% value of R2, while the prediction of sand production achieved 77% accuracy. In addition, the model for the sand pack test achieved 84% accuracy in predicting sand production. For comparative model performance, gradient boosting (GB), K-nearest neighbor (KNN), random forest (RF), and support vector machine (SVM) were also modelled on the same datasets. The results showed that the proposed 1D-CNN model outperforms the other four machine learning models for both SRT tests in terms of prediction accuracy.

show abstract

Dimensionality Reduction Methods Used in History Matching

Yousefzadeh¹,

Kazemi²,

Ahmadi³

et al. 2023

SpringerBriefs in Petroleum Geoscience &Amp; Engineering

View full text Add to dashboard Cite

Determination of oil well placement using convolutional neural network coupled with robust optimization under geological uncertainty

Cited by 35 publications

References 41 publications

Evaluation of the nominal thickness of pipelines using neural networks through the dispersion curves drawn with the SAFE method

Evaluation of the nominal thickness of pipelines using neural networks through the dispersion curves drawn with the SAFE method

One-Dimensional Convolutional Neural Network with Adaptive Moment Estimation for Modelling of the Sand Retention Test

Dimensionality Reduction Methods Used in History Matching

Contact Info

Product

Resources

About