Constructing Dual-Porosity Models from High-Resolution Discrete-Fracture Models Using Deep Neural Networks

2021 ◽  
Author(s):  
Xupeng He ◽  
Ryan Santoso ◽  
Marwa Alsinan ◽  
Hyung Kwak ◽  
Hussein Hoteit
Keyword(s):  
Simulation Model ◽  
Oil Recovery ◽  
Reservoir Simulation ◽  
Water Saturation ◽  
Dual Porosity ◽  
Discrete Fracture Model ◽  
Discrete Fracture ◽  
Dual Porosity Model ◽  
Reservoir Simulation Model ◽  
Porosity Model

Abstract Detailed geological description of fractured reservoirs is typically characterized by the discrete-fracture model (DFM), in which the rock matrix and fractures are explicitly represented in the form of unstructured grids. Its high computation cost makes it infeasible for field-scale applications. Traditional flow-based and static-based methods used to upscale detailed geological DFM to reservoir simulation model suffer from, to some extent, high computation cost and low accuracy, respectively. In this paper, we present a novel deep learning-based upscaling method as an alternative to traditional methods. This work aims to build an image-to-value model based on convolutional neural network to model the nonlinear mapping between the high-resolution image of detailed DFM as input and the upscaled reservoir simulation model as output. The reservoir simulation model (herein refers to the dual-porosity model) includes the predicted fracture-fracture transmissibility linking two adjacent grid blocks and fracture-matrix transmissibility within each coarse block. The proposed upscaling workflow comprises the train-validation samples generation, convolutional neural network training-validating process, and model evaluation. We apply a two-point flux approximation (TPFA) scheme based on embedded discrete-fracture model (EDFM) to generate the datasets. We perform trial-error analysis on the coupling training-validating process to update the ratio of train-validation samples, optimize the learning rate and the network architecture. This process is applied until the trained model obtains an accuracy above 90 % for both train-validation samples. We then demonstrate its performance with the two-phase reference solutions obtained from the fine model in terms of water saturation profile and oil recovery versus PVI. Results show that the DL-based approach provides a good match with the reference solutions for both water saturation distribution and oil recovery curve. This work manifests the value of the DL-based method for the upscaling of detailed DFM to the dual-porosity model and can be extended to construct generalized dual-porosity, dual-permeability models or include more complex physics, such as capillary and gravity effects.

General Transfer Functions for Multiphase Flow in Fractured Reservoirs

SPE Journal ◽  
2008 ◽  
Vol 13 (03) ◽  
pp. 289-297 ◽  
Author(s):  
Huiyun Lu ◽  
Ginevra Di Donato ◽  
Martin J. Blunt
Keyword(s):  
Transfer Function ◽  
Reservoir Simulation ◽  
Transfer Rate ◽  
Unit Volume ◽  
Fractured Reservoirs ◽  
Dual Porosity ◽  
Permeability Model ◽  
Dual Porosity Model ◽  
The Matrix ◽  
Porosity Model

Summary We propose a physically motivated formulation for the matrix/fracture transfer function in dual-porosity and dual-permeability reservoir simulation. The approach currently applied in commercial simulators (Barenblatt et al. 1960; Kazemi et al. 1976) uses a Darcy-like flux from matrix to fracture, assuming a quasisteady state between the two domains that does not correctly represent the average transfer rate in a dynamic displacement. On the basis of 1D analyses in the literature, we find expressions for the transfer rate accounting for both displacement and fluid expansion at early and late times. The resultant transfer function is a sum of two terms: a saturation-dependent term representing displacement and a pressure-dependent term to model fluid expansion. The transfer function is validated through comparison with 1D and 2D fine-grid simulations and is compared to predictions using the traditional Kazemi et al. (1976) formulation. Our method captures the dynamics of expansion and displacement more accurately. Introduction The conventional macroscopic treatment of flow in fractured reservoirs assumes that there are two communicating domains: a flowing region containing connected fractures and high permeability matrix and a stagnant region of low-permeability matrix (Barenblatt et al. 1960; Warren and Root 1963). Conventionally, these are referred to as fracture and matrix, respectively. Transfer between fracture and matrix is mediated by gravitational and capillary forces. In a dual-porosity model, it is assumed that there is no viscous flow in the matrix; a dual-permeability model allows flow in both fracture and matrix. In a general compositional model (where black-oil and incompressible flow are special cases) we can write[Equation 1], where where Gc is a transfer term with units of mass per unit volume per unit time--it is a rate (units of inverse time) times a density (mass per unit volume). c is a component density (concentration) with units of mass of component per unit volume. The subscript p labels the phase, and c labels the component. Gc represents the transfer of component c from fracture to matrix. The subscript f refers to the flowing or fractured domain. The first term is accumulation, and the second term represents flow--this is the same as in standard (nonfractured) reservoir simulation. We can write a corresponding equation for the matrix, m,[Equation 2] where we have assumed a dual-porosity model (no flow in the matrix); for a dual-permeability model, a flow term is added to Eq. 2.

History Matching and Performance Prediction of a Polymer Flood Pilot in Heavy Oil Reservoir on Alaska North Slope

2021 ◽  
Author(s):  
Xindan Wang ◽  
Cody Keith ◽  
Yin Zhang ◽  
Abhijit Dandekar ◽  
Samson Ning ◽  
...  
Keyword(s):  
Simulation Model ◽  
Oil Recovery ◽  
Reservoir Simulation ◽  
Heavy Oil ◽  
History Matching ◽  
Polymer Flooding ◽  
North Slope ◽  
Polymer Flood ◽  
Reservoir Simulation Model ◽  
Water Cut

Abstract The first-ever polymer flood pilot to enhance heavy oil recovery on Alaska North Slope (ANS) is ongoing. After more than 2.5 years of polymer injection, significant benefit has been observed from the decrease in water cut from 65% to less than 15% in the project producers. The primary objective of this study is to develop a robust history-matched reservoir simulation model capable of predicting future polymer flood performance. In this work, the reservoir simulation model has been developed based on the geological model and available reservoir and fluid data. In particular, four high transmissibility strips were introduced to connect the injector-producer well pairs, simulating short-circuiting flow behavior that can be explained by viscous fingering and reproducing the water cut history. The strip transmissibilities were manually tuned to improve the history matching results during the waterflooding and polymer flooding periods, respectively. It has been found that higher strip transmissibilities match the sharp water cut increase very well in the waterflooding period. Then the strip transmissibilities need to be reduced with time to match the significant water cut reduction. The viscous fingering effect in the reservoir during waterflooding and the restoration of injection conformance during polymer flooding have been effectively represented. Based on the validated simulation model, numerical simulation tests have been conducted to investigate the oil recovery performance under different development strategies, with consideration for sensitivity to polymer parameter uncertainties. The oil recovery factor with polymer flooding can reach about 39% in 30 years, twice as much as forecasted with continued waterflooding. Besides, the updated reservoir model has been successfully employed to forecast polymer utilization, a valuable parameter to evaluate the pilot test’s economic efficiency. All the investigated development strategies indicate polymer utilization lower than 3.5 lbs/bbl in 30 years, which is economically attractive.

Robust Life-Cycle Production Optimization With a Support-Vector-Regression Proxy

SPE Journal ◽  
2018 ◽  
Vol 23 (06) ◽  
pp. 2409-2427 ◽  
Author(s):  
Zhenyu Guo ◽  
Albert C. Reynolds
Keyword(s):  
Life Cycle ◽  
Simulation Model ◽  
Support Vector Regression ◽  
Reservoir Simulation ◽  
Production Optimization ◽  
Forward Model ◽  
Support Vector ◽  
Well Control ◽  
Steepest Ascent ◽  
Reservoir Simulation Model

Summary We design a new and general work flow for efficient estimation of the optimal well controls for the robust production-optimization problem using support-vector regression (SVR), where the cost function is the net present value (NPV). Given a set of simulation results, an SVR model is built as a proxy to approximate a reservoir-simulation model, and then the estimated optimal controls are found by maximizing NPV using the SVR proxy as the forward model. The gradient of the SVR model can be computed analytically so the steepest-ascent algorithm can easily and efficiently be applied to maximize NPV. Then, the well-control optimization is performed using an SVR model as the forward model with a steepest-ascent algorithm. To the best of our knowledge, this is the first SVR application to the optimal well-control problem. We provide insight and information on proper training of the SVR proxy for life-cycle production optimization. In particular, we develop and implement a new iterative-sampling-refinement algorithm that is designed specifically to promote the accuracy of the SVR model for robust production optimization. One key observation that is important for reservoir optimization is that SVR produces a high-fidelity model near an optimal point, but at points far away, we only need SVR to produce reasonable approximations of the predicting output from the reservoir-simulation model. Because running an SVR model is computationally more efficient than running a full-scale reservoir-simulation model, the large computational cost spent on multiple forward-reservoir-simulation runs for robust optimization is significantly reduced by applying the proposed method. We compare the performance of the proposed method using the SVR runs with the popular stochastic simplex approximate gradient (StoSAG) and reservoir-simulations runs for three synthetic examples, including one field-scale example. We also compare the optimization performance of our proposed method with that obtained from a linear-response-surface model and multiple SVR proxies that are built for each of the geological models.

Connecting the in vitro and in vivo experiments in electrochemotherapy - a feasibility study modeling cisplatin transport in mouse melanoma using the dual-porosity model

2018 ◽  
Vol 286 ◽  
pp. 33-45 ◽  
Author(s):  
Janja Dermol-Černe ◽  
Janja Vidmar ◽  
Janez Ščančar ◽  
Katja Uršič ◽  
Gregor Serša ◽  
...  
Keyword(s):  
Feasibility Study ◽  
Dual Porosity ◽  
Mouse Melanoma ◽  
In Vivo Experiments ◽  
Dual Porosity Model ◽  
Porosity Model

Conceptualization of Karstic Aquifer with Multiple Outlets Using a Dual Porosity Model

Ground Water ◽  
2017 ◽  
Vol 55 (4) ◽  
pp. 558-564 ◽  
Author(s):  
Seiyed Mossa Hosseini ◽  
Behzad Ataie-Ashtiani
Keyword(s):  
Dual Porosity ◽  
Karstic Aquifer ◽  
Dual Porosity Model ◽  
Porosity Model
Sign in / Sign up

Export Citation Format

Share Document