A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z  

  Subjects -> GEOGRAPHY (Total: 493 journals)
The end of the list has been reached or no journals were found for your choice.
   Most Followed Journals

   Top Publishers

   Top Subjects
Similar Journals
Journal Cover
Computational Geosciences
Journal Prestige (SJR): 0.985
Citation Impact (citeScore): 3
Number of Followers: 17  
 
  Hybrid Journal Hybrid journal (It can contain Open Access articles)
ISSN (Print) 1573-1499 - ISSN (Online) 1420-0597
Published by Springer-Verlag Homepage  [2468 journals]
  • Neural spline flow multi-constraint NURBS method for three-dimensional
           automatic geological modeling with multiple constraints

    • Free pre-print version: Loading...

      Abstract: The strike and the dip angle are vital for describing the geometry of the rock formations. However, in the interpolation and geological modeling, only the coordinates are considered; the strike and the dip angle are ignored. To this end, the neural spline flow (NSF) multi-constraint non-uniform rational B-splines (McNURBS) method is proposed in this study. Any complex high-dimensional joint distribution can be learned using the deep generative model, NSF; thus, the NSF model was used to perform the exact maximum likelihood estimation and joint sampling of three-dimensional (3D) geological point coordinates, strike, and dip angle, overcoming the shortcomings of conventional statistical models, which are difficult to extend to high-dimensional problems. In addition, the conventional single-constraint NURBS modeling method based on geological point coordinates was improved to obtain the McNURBS modeling method, which considers the geological point coordinates, strike, and dip angle during the modeling process. The practical application results show that by using the proposed method, a 3D geological model can be flexibly and automatically established considering both geological point coordinates and strike and dip angle constraints. Moreover, the fitting relative error (RE) of the proposed method was reduced by 52.4% compared to the conventional NURBS method. This study provides a convenient and efficient means to automatically build a reliable 3D geological model.
      PubDate: 2023-06-01
       
  • Sequential fully implicit newton method for flow and transport with
           natural black-oil formulation

    • Free pre-print version: Loading...

      Abstract: There is a rising interest in developing robust and efficient sequential methods for reservoir simulation due to potential benefits from specialized nonlinear and linear solvers, flexible discretizations, and adaptivity. The recently proposed sequential fully implicit Newton (SFIN) method addresses a major bottleneck of the sequential strategies: the slow sequential coupling convergence when flow and transport problems are strongly coupled. However, the original SFIN algorithm requires fixed primary variables during the simulation. For the natural formulation widely used in reservoir simulation, primary variables will switch when a phase change happens. In this work, we proposed strategies to address the issue of inconsistent primary variables and extended the SFIN method to the natural black-oil formulation. Several challenging numerical cases are presented to demonstrate that the extended SFIN method can achieve significant sequential convergence acceleration and improved overall performance for natural formulation when phase changes and primary variable switch happen frequently.
      PubDate: 2023-06-01
       
  • A generic workflow of projection-based embedded discrete fracture model
           for flow simulation in porous media

    • Free pre-print version: Loading...

      Abstract: Numerical modeling of multiphase flow phenomena in fractured porous media has always been a hot topic in recent years. Projection-based embedded discrete fracture model (pEDFM) is a recently developed modeling framework that has been shown to resolve some limitations of the commonly used embedded discrete fracture model (EDFM). However, determining the continuous projection path of fracture networks and adding needed fracture-fracture (f-f) connections in the previous pEDFM workflows is too complicated to construct a generic and easy-to-program algorithm. Although there are several studies about applications of pEDFM to numerically model the flow problems in fractured porous media, a generic pEDFM workflow has not been truly formed. Therefore, this theoretically advanced numerical modeling framework of flow problems in fractured porous media has not been widely noticed and applied. To construct a generic pEDFM workflow, this paper first illustrates that the fracture projection path obtained by the micro-translation method can achieve acceptable simulation accuracy for flow across a high-conductivity fracture, then a generic and easy-to-program non-projective method is developed to model low-conductivity fractures. Numerical examples verify that the developed pEDFM workflow can effectively handle low-conductivity fractures (e.g. flow barriers) in general cases, thus avoiding the difficulty of developing a generic algorithm for constructing a continuous projection path for the low-conductivity fracture. It is also verified that the new pEDFM workflow with the addition of simple f-f connections can solve the cases that classical pEDFM solves with significant errors. Furthermore, it achieves similar or better computational accuracy than the pEDFM modified by adding complex f-f connections, thus avoiding the difficulty of developing a generic algorithm for constructing these additional f-f connections. Overall, the first generic pEDFM workflow is developed in this paper, which significantly improves the practicability of pEDFM and may induce more applications of pEDFM in numerical modeling of flow problems in fractured porous media.
      PubDate: 2023-05-24
       
  • A novel prediction method for coalbed methane production capacity combined
           extreme gradient boosting with bayesian optimization

    • Free pre-print version: Loading...

      Abstract: Coalbed methane plays a significant role for the sustainable utilizing of resources and ecological environment. Production capacity forecasting of coalbed methane wells can effectively guide the optimization of development schemes directly affecting the economic benefits. To overcome the inefficiency of traditional theory-based numerical simulators and their weak adaptability to observational data, we explore a potential and efficient alternative for modeling of production capacity in a data-driven approach. This study makes full use of dynamic production data and geological static data from 530 CBM wells. We develop a production capacity prediction model utilizing the extreme gradient boosting algorithm and incorporated bayesian optimization to implement an automated search for hyperparameters. The results demonstrate that the prediction model developed by extreme gradient boosting has a more powerful prediction performance with an R2 close to 0.9 compared to other machine learning or even deep learning. Moreover, the coupled framework of extreme gradient boosting and bayesian optimization can notably upgrade the prediction power of the production capacity model by about 8%. The analysis of influencing factors also illustrates that dynamic production data during the first three years of development can well characterize the coalbed methane adsorption–desorption-seepage features, which contribute to the construction of the production capacity model.
      PubDate: 2023-05-19
       
  • Improved Petrophysical Property Evaluation of Shaly Sand Reservoirs Using
           Modified Grey Wolf Intelligence Algorithm

    • Free pre-print version: Loading...

      Abstract: Multi-frequency dielectric scanning logging is an advanced method that plays a critical role in evaluating unconventional oil and gas reserves and residual oil distribution. This method provides higher accuracy compared to conventional logging and can obtain essential formation parameters such as formation water salinity, pore textural index, and dispersive phase volume fraction. Despite its advantages, the inversion of permittivity and conductivity measurements at multiple frequencies into formation properties remains a "black box" problem. This challenge makes it challenging to understand specific implementation methods and inversion techniques without purchasing expensive software and hardware from oil field service companies. To address this issue, this work proposes a publicly available and advanced intelligent optimization algorithm. Our approach reduces calculation complexity and achieves high accuracy in evaluating petrophysical properties, including shaly sand reservoirs, without relying on costly software services from oil field companies. Our method offers distinct advantages over traditional approaches, such as the ability to derive formation properties from dielectric logging information with confidence, without specialized equipment or software from commercial providers. Additionally, the open-source code is readily available in various programming languages, including Python, R, and Matlab, making our approach accessible and easy to implement.
      PubDate: 2023-05-17
       
  • Correlation analysis: application of DFA and DCCA in well log profiles

    • Free pre-print version: Loading...

      Abstract: Detrended fluctuation analysis and detrended cross correlation analysis are used in this work to identify and characterize correlated well log data. This is performed by first separating the different fluctuations due to external trends, and evaluating the autocorrelation and cross-correlation exponents to determine whether scale properties persist as the size of the series changes. Two new methodologies were developed to identify optimal values of the cross-correlation coefficients and graphically display them, which we call the automatic search procedure and correlation map. The methodologies were applied to well logs from the Jequitinhonha Basin, Brazil, to verify the existence of scale property in these data. For practical purposes, our goal is to use a local analysis framework to detect all points of high cross-correlation among different physical parameters in the same well, and among one same physical parameter in different wells. The correlated events suggested the continuity of geological features, including the vertical displacements of rock layers. In particular, it was possible to identify layers of calcilutites in a specific depth range. These rocks are of particular importance to the study of stratigraphic correlations due to their great regional extent and regular layering.
      PubDate: 2023-05-17
       
  • Stochastic reconstruction of porous media based on attention mechanisms
           and multi-stage generative adversarial network

    • Free pre-print version: Loading...

      Abstract: The accurate reconstruction of porous media is difficult to accomplish in practical research due to the intricacy of the internal structure of porous media, which cannot be described only using some equations or languages. The emerging deep learning methods open up a new research field for the reconstruction of porous media. One of the classic deep learning methods is generative adversarial network (GAN), which has been applied to the reconstruction of porous media, but also requires sufficient training samples as well as a long simulation process. Some GAN’s variants, such as the single-image GAN (SinGAN), are proposed to achieve desired results with only one training image (TI), but its serial structure still necessitates lengthy training time. Based on SinGAN, a stochastic reconstruction method of porous media combining attention mechanisms with multiple-stage GAN is proposed, focusing on important features of porous media with a single TI to achieve favorable simulation quality and using two discriminators to maintain enough diversity (one discriminator prefers real data, and the other favors fake data). Experiments prove that our method outperforms some numerical reconstruction methods and SinGAN in terms of practicality and efficiency.
      PubDate: 2023-05-06
       
  • A machine-learning-accelerated distributed LBFGS method for field
           development optimization: algorithm, validation, and applications

    • Free pre-print version: Loading...

      Abstract: We have developed a support vector regression (SVR) accelerated variant of the distributed derivative-free optimization (DFO) method using the limited-memory BFGS Hessian updating formulation (LBFGS) for subsurface field-development optimization problems. The SVR-enhanced distributed LBFGS (D-LBFGS) optimizer is designed to effectively locate multiple local optima of highly nonlinear optimization problems subject to numerical noise. It operates both on single- and multiple-objective field-development optimization problems. The basic D-LBFGS DFO optimizer runs multiple optimization threads in parallel and uses the linear interpolation method to approximate the sensitivity matrix of simulated responses with respect to optimized model parameters. However, this approach is less accurate and slows down convergence. In this paper, we implement an effective variant of the SVR method, namely ε-SVR, and integrate it into the D-LBFGS engine in synchronous mode within the framework of a versatile optimization library inside a next-generation reservoir simulation platform. Because ε-SVR has a closed-form of predictive formulation, we analytically calculate the approximated objective function and its gradients with respect to input model variables subject to optimization. We investigate two different methods to propose a new search point for each optimization thread in each iteration through seamless integration of ε-SVR with the D-LBFGS optimizer. The first method estimates the sensitivity matrix and the gradients directly using the analytical ε-SVR surrogate and then solves a LBFGS trust-region subproblem (TRS). The second method applies a trust-region search LBFGS method to optimize the approximated objective function using the analytical ε-SVR surrogate within a box-shaped trust region. We first show that ε-SVR provides accurate estimates of gradient vectors on a set of nonlinear analytical test problems. We then report the results of numerical experiments conducted using the newly proposed SVR-enhanced D-LBFGS algorithms on both synthetic and realistic field-development optimization problems. We demonstrate that these algorithms operate effectively on realistic nonlinear optimization problems subject to numerical noise. We show that both SVR-enhanced D-LBFGS variants converge faster and thereby provide a significant acceleration over the basic implementation of D-LBFGS with linear interpolation.
      PubDate: 2023-05-04
       
  • A hybrid stability analysis bypassing method for compositional simulation

    • Free pre-print version: Loading...

      Abstract: Stability analysis has become the standard method to test the phase appearance in compositional reservoir simulation. This method is very computationally intensive and consumes most of the CPU time in simulation because every single-phase grid block needs to perform this test at every outer iteration in solving global equations (the iterations to solve pressure, saturation and composition). To reduce this part of simulation time, some stability analysis bypassing methods are proposed. Two main methods are the neighborhood bypassing (NB) method and the shadow region bypassing (SB) method. Based on these two methods, a hybrid bypassing (HB) method is presented to improve the efficiency and accuracy of the current two-phase stability analysis bypassing methods. The NB method has limited applications and a comprehensive comparison of these two methods has not been given, so the most effective method is yet unknown. We implement these three bypassing methods in a parallel isothermal compositional simulator and provide a detailed comparison of the performance of these three methods for different reservoir grid models, hydrocarbon fluids and field operations. The results show that the performance of these methods is very case-dependent, and the drawbacks of the NB and SB methods are clearly demonstrated. It is also shown that the HB method always performs better than the SB method and better than or equally compared to the NB method in most cases. Based on the performance of these methods, we give the recommendation that which method should be used in a certain condition.
      PubDate: 2023-05-04
       
  • Impact of heterogeneity upon the accuracy of the Eikonal solution using
           the Fast Marching Method

    • Free pre-print version: Loading...

      Abstract: The Fast Marching Method (FMM) is an efficient tool for the solution of the Eikonal equation to characterize frontal propagation in heterogeneous and anisotropic media. Previous studies have applied the FMM in solving the Eikonal equation to obtain the “diffusive time of flight” (DTOF), which is used to characterize the pressure front propagation in subsurface porous media. In the DTOF (τ) based one-dimensional flow simulation, the accuracy of the pressure solution relies significantly upon the drainage volume characterization and the DTOF calculation, both of which are influenced by the reservoir heterogeneity. We study first order discretization schemes of the Eikonal equation in porous media with different levels of heterogeneity and determine the impact of the corresponding DTOF solutions upon the one-dimensional flow simulation. The local solution of the Eikonal equation is formulated based on a piecewise uniform approximation of the DTOF gradient that satisfies a causality requirement within each simplex that provides the available nodal DTOF values. A homogenization approach is developed for use within the local solution for corner point grid cells, in which effective properties are obtained across adjacent grid cells. Comparison of the computational cost and error from different discretizations of the Eikonal equation and the corresponding pressure solution with increasing heterogeneity, demonstrates the need for increased angular resolution compared to the most usual cell centered 5-point discretization. The study also demonstrates the need for more accurate solutions in the near well region, suggesting a multi-stencil approach with higher angular and linear resolution near the well, and a less accurate and less costly calculation in the bulk of the computational grid.
      PubDate: 2023-04-22
       
  • Digital core image reconstruction based on residual self-attention
           generative adversarial networks

    • Free pre-print version: Loading...

      Abstract: In order to perform accurate physical analysis of digital core, the reconstruction of high-quality digital core image has become a problem to be resolved at present. In this paper, a digital core image reconstruction method based on the residual self-attention generative adversarial networks is proposed. In the process of digital core image reconstruction, the traditional generative adversarial networks (GANs) can obtain high resolution detail features only by the spatial local point generation in low resolution details, and the far away dependency can only be processed by multiple convolution operations. In view of this, in this paper the residual self-attention block is introduced in the traditional GANs, which can strengthen the correlation learning between features and extract more features. In order to analyze the quality of generated shale images, in this paper the Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) are used to evaluate the consistency of Gaussian distribution between reconstructed shale images and original ones, and the two-point covariance function is used to evaluate the structural similarity between reconstructed shale images and original ones. Plenty experiments show that the reconstructed shale images by the proposed method in the paper are closer to the original images and have better effect, compared to those of the state-of-art methods.
      PubDate: 2023-04-21
       
  • A 2D numerical model for simulation of cohesive sediment transport

    • Free pre-print version: Loading...

      Abstract: The study of the characteristics of fine-grained and cohesive sediments due to its filling characteristics in harbors and ports is an important subject in coastal studies. Fine-grained cohesive sediments have special complicated characteristics as compared to other sediments regarding their behavior. Hence, numerous research works have been carried out to establish well validated physical and mathematical descriptions of the behavior and outcome of concentrated near-bed cohesive sediment suspensions and their interaction with the water column and the bed as well as the turbulence characteristics of sediment laden flow. A two-dimensional model is developed in this study that includes: a cohesive sediment simulator module, processes such as advection and diffusion of cohesive sediment, flocculation and its effect on the settling velocity of cohesive sediment particles, consolidation of bed layers and sediment transport between layers, substrate shear stress variations affecting the simultaneous presence of wave and flow, bed morphology, deposition and erosion. In all these processes, the depth is considered according to the actual topography of the bed. For verification of model performance the model results have been compared with the MIKE 21 model results against the field data reported during the construction phase and at the simulation stage of Ho Bay, a case study presented by DHI (MIKE, 15). The comparisons indicate a favorable accuracy of the present model performance in simulation of cohesive sediment transport.
      PubDate: 2023-04-20
       
  • Gaussian active learning on multi-resolution arbitrary polynomial chaos
           emulator: concept for bias correction, assessment of surrogate reliability
           and its application to the carbon dioxide benchmark

    • Free pre-print version: Loading...

      Abstract: Surrogate models are widely used to improve the computational efficiency in various geophysical simulation problems by reducing the number of model runs. Conventional one-layer surrogate representations are based on global (e.g. polynomial chaos expansion, PCE) or on local kernels (e.g., Gaussian process emulator, GPE). Global representations omit some details, while local kernels require more model runs. The existing multi-resolution PCE is a promising hybrid: it is a global representation with local refinement. However, it can not (yet) estimate the uncertainty of the resulting surrogate, which techniques like the GPE can do. We propose to join multi-resolution PCE and GPE s into a joint surrogate framework to get the best out of both worlds. By doing so, we correct the surrogate bias and assess the remaining uncertainty of the surrogate itself. The resulting multi-resolution emulator offers a pathway for several active learning strategies to improve the surrogate at acceptable computational costs, compared to the existing PCE-kriging approach it adds the multi-resolution aspect. We analyze the performance of a multi-resolution emulator and a plain GPE using didactic test cases and a CO2 benchmark, that is representative of many alike problems in the geosciences. Both approaches show similar improvements during the active learning, but our multi-resolution emulator leads to much more stable results than the GPE. Overall, our suggested emulator can be seen as a generalization of multi-resolution PCE and GPE concepts that offers the possibility for active learning.
      PubDate: 2023-04-14
       
  • The use of stochastic geomechanical properties of potential failure plane
           and fracture networks for realistic modelling of rock mass behaviour: A
           synthetic rock mass modelling (SRM) study

    • Free pre-print version: Loading...

      Abstract: Almost all conventional discrete fracture network (DFN) models embedded within rock masses are discontinuities with zero tensile strength and mean values of geomechanical parameters. However, the spatial variability and networks of weak and strong potential failure planes and discontinuities have a significant effect on rock mass behaviour. Therefore, the geomechanical heterogeneous nature of potential failure planes and fractures, along with their geometrical parameters, is crucial for understanding rock mass behaviour. To bridge this gap in research, this paper provides a methodology for stochastic modelling of potential failure plane and discontinuity geomechanical properties effects on rock mass behaviour using a combination of DFN and discrete element modelling approach. Due to the uncertainties and distributed geomechanical characteristics of DFN, fracturing may occur through an intact part, DFN, or a combination of intact and DFN. A parametric study was performed to investigate the influence of friction angle, normal and shear stiffness, cohesion, and tensile strength of potential failure planes and discontinuities. The results indicate that the proposed stochastic geomechanical DFN model gives more realistic rock mass strengths and failure patterns compared to the conventional DFN framework.
      PubDate: 2023-04-12
       
  • Correction to: Estimating permeability of 3D micro-CT images by
           physics-informed CNNs based on DNS

    • Free pre-print version: Loading...

      PubDate: 2023-03-23
       
  • Correction to: Sequential fully implicit newton method for flow and
           transport with natural black-oil formulation

    • Free pre-print version: Loading...

      PubDate: 2023-03-21
       
  • Three-dimensional numerical simulation of hydraulically driven cohesive
           fracture propagation in deformable reservoir rock using enriched EFG
           method

    • Free pre-print version: Loading...

      Abstract: In this paper, a fully coupled 3D. numerical simulation of hydraulic fracture propagation in saturated deformable porous media is presented in the context of the extrinsically enriched element free Galerkin (EFG) method. By exploiting the partition of unity property of moving least square shape functions, weak and strong discontinuities are simulated using the Ridge and the Heaviside enrichment functions, respectively. The cohesive crack model is used to describe the nonlinear fracture processes developing in the area in front of the crack tip where the energy dissipation takes place. The fracturing fluid flow within the fracture is modeled using Darcy’s law and the fracture permeability is considered to follow the cubic law. The developed fully coupled numerical framework can simulate the fluid leak-off phenomenon and formation of the fluid-lag zone. For verification of the developed computational algorithm, a problem with an analytical solution was simulated and a good agreement was seen between numerical and analytical results. The numerical simulations and the parametric studies results show that the proposed numerical framework can successfully simulate various aspects of the complicated process of the hydraulic fracturing treatment.
      PubDate: 2023-03-15
      DOI: 10.1007/s10596-023-10198-2
       
  • Comparison of nonlinear field-split preconditioners for two-phase flow in
           heterogeneous porous media

    • Free pre-print version: Loading...

      Abstract: This work focuses on the development of a two-step field-split nonlinear preconditioner to accelerate the convergence of two-phase flow and transport in heterogeneous porous media. We propose a field-split algorithm named Field-Split Multiplicative Schwarz Newton (FSMSN), consisting in two steps: first, we apply a preconditioning step to update pressure and saturations nonlinearly by solving approximately two subproblems in a sequential fashion; then, we apply a global step relying on a Newton update obtained by linearizing the system at the preconditioned state. Using challenging test cases, FSMSN is compared to an existing field-split preconditioner, Multiplicative Schwarz Preconditioned for Inexact Newton (MSPIN), and to standard solution strategies such as the Sequential Fully Implicit (SFI) method or the Fully Implicit Method (FIM). The comparison highlights the impact of the upwinding scheme in the algorithmic performance of the preconditioners and the importance of the dynamic adaptation of the subproblem tolerance in the preconditioning step. Our results demonstrate that the two-step nonlinear preconditioning approach—and in particular, FSMSN—results in a faster outer-loop convergence than with the SFI and FIM methods. The impact of the preconditioners on computational performance–i.e., measured by wall-clock time–will be studied in a subsequent publication.
      PubDate: 2023-03-15
      DOI: 10.1007/s10596-023-10200-x
       
  • Acknowledgement for Reviewers for 2022

    • Free pre-print version: Loading...

      PubDate: 2023-02-16
      DOI: 10.1007/s10596-023-10193-7
       
  • Estimating permeability of 3D micro-CT images by physics-informed CNNs
           based on DNS

    • Free pre-print version: Loading...

      Abstract: In recent years, convolutional neural networks (CNNs) have experienced an increasing interest in their ability to perform a fast approximation of effective hydrodynamic parameters in porous media research and applications. This paper presents a novel methodology for permeability prediction from micro-CT scans of geological rock samples. The training data set for CNNs dedicated to permeability prediction consists of permeability labels that are typically generated by classical lattice Boltzmann methods (LBM) that simulate the flow through the pore space of the segmented image data. We instead perform direct numerical simulation (DNS) by solving the stationary Stokes equation in an efficient and distributed-parallel manner. As such, we circumvent the convergence issues of LBM that frequently are observed on complex pore geometries, and therefore, improve the generality and accuracy of our training data set. Using the DNS-computed permeabilities, a physics-informed CNN (PhyCNN) is trained by additionally providing a tailored characteristic quantity of the pore space. More precisely, by exploiting the connection to flow problems on a graph representation of the pore space, additional information about confined structures is provided to the network in terms of the maximum flow value, which is the key innovative component of our workflow. The robustness of this approach is reflected by very high prediction accuracy, which is observed for a variety of sandstone samples from archetypal rock formations.
      PubDate: 2023-01-31
      DOI: 10.1007/s10596-022-10184-0
       
 
JournalTOCs
School of Mathematical and Computer Sciences
Heriot-Watt University
Edinburgh, EH14 4AS, UK
Email: journaltocs@hw.ac.uk
Tel: +00 44 (0)131 4513762
 


Your IP address: 34.229.131.158
 
Home (Search)
API
About JournalTOCs
News (blog, publications)
JournalTOCs on Twitter   JournalTOCs on Facebook

JournalTOCs © 2009-