OAKTrust :: Browsing by Author "Gildin, Eduardo"

Browsing by Author "Gildin, Eduardo"

Now showing 1 - 20 of 202

3D Reservoir Simulation of a Hydraulically Fractured Vertical Gas Well Using the Embedded Discrete Fracture Model (EDFM)
(2017-07-27) Orta, Samuel Rene; Killough, John; Gildin, Eduardo; Pistikopoulos, Stratos
According to the 2017 outlook on energy conducted by the EIA, the production of natural gas from shale reservoirs is expected to increase by 2040. This increase is dependent on multiple factors like technology, resources, and market conditions. In order to meet production increases, improvements in industry practices must happen in order to achieve better economic development of these reservoirs. One of the primary tools used in economic development is reservoir simulation. However, modeling and simulating low permeability reservoirs, like shale, can be complex due to the presence of natural fractures. Knowing that fractures have a significant role in the total recovery of a field, modeling the influence of these fractures is of utmost importance. There are two common methods used to model naturally fractured reservoirs, the dual continuum model and the discrete fracture model. In this study, an embedded discrete fracture model (EDFM) was used to simulate and match a hydraulically fractured vertical gas well in the Barnett Shale in 3D. EDFM was proposed by Li and Lee (2008) to couple the dual continuum model with the discrete fracture model to take advantage of both methods. The results show that EDFM can be validated and can be an efficient tool for future reservoir simulation. In addition, a parametric study was conducted to visualize how fracture orientation and fracture density impact fluid flow. It was concluded that fracture orientation plays an important part in fracture connectivity, which is a result of fracture orientation. If the fractures are highly connective, the better the well performance. The same was seen with fracture density, an increased fracture density leads to better well performance.
Accounting for Nonuniform Induced Properties in Production Analysis of Unconventional Reservoirs
(2014-10-31) Fuentes Cruz, Gorgonio; Gildin, Eduardo; Valkó, Peter P.; Blasingame, Thomas A.; Medina-Cetina, Zenon
Several conceptual models for unconventional reservoirs have been proposed in recent years based on extensions of well-studied analytical/semi-analytical models for conventional reservoirs. The standard semi-analytical approaches assume uniform properties in the reservoir. In this study, we develop new models for production data analysis of hydraulically fractured wells based on the concept of nonuniform induced properties, in particular, the induced permeability field and the induced interporosity flow field. In the induced permeability field approach, we consider the case when the hydraulic fracturing operation alters the ability of the formation to conduct fluids throughout, but in varying degrees depending on the distance from the main hydraulic fracture plane. In the induced interporosity flow field, we assume that, as a result of the hydraulic fracturing treatment, the density of micro-fractures (natural and induced) is high near the hydraulic fracture face, but gradually decreases away from it. We also address common issues related to variations in the wellbore pressure, desorption effects, and non-linearity caused by gas flow, with the intent to provide a simple, yet clear understanding of their effects on the production performance. The methods used in this work include mostly semi-analytical techniques (Laplace transform and numerical inversion to the time domain). Analytical (formulae in the time domain) and numerical simulation (finite difference) techniques are also used to validate the results from the new models. The results indicate that the maximum and minimum induced permeabilities may provide the key to evaluate the overall completion efficiency in unconventional formations, where the extent and quality of the stimulated volume are equally significant. Also, the closely spaced micro-fractures have a strong impact on well performance, even when their density is diminishing toward the far parts of the stimulated reservoir. We conclude that the new models preserve the typical linear-flow signature of commonly observed well performance of unconventional shale reservoirs; however, the extrapolation of the production behavior departs from the standard models significantly. This research contributes to the understanding of the production behavior of unconventional reservoirs to characterize the quality of the stimulated reservoir and to consider often neglected factors effecting forecasts of well performance.
Accounting for Remaining Injected Fracturing Fluid
(2013-12-06) Zhang, Yannan; Ehlig-Economides, Christine; Sun, Yuefeng; Gildin, Eduardo
The technology of multi-stage fracturing of horizontal wells made the development of shale gas reservoirs become greatly successful during the past decades. A large amount of fracturing fluid, usually from 53,000 bbls to 81,400 bbls, is injected into the reservoir to create the fractures. However, only a small fraction of injected fracturing fluid from 10% to 40% has been recovered during the flowback process and the long term shale gas well production period. Possible mechanisms for low load recovery include ineffective dewatering of the propped fractures, matrix pore scale water retention related to imbibition, capillary fluid retention, relative permeability, and water held up in a fracture network (complexity) opened or reopened during fracture treatments. This work is critical both to understand existing shale gas well performance and to improve shale gas well designs. Current treatment practices that promote fracture complexity as an objective may be misplaced in some shale formations. As well, the number of fractures seemingly created from so many perforation clusters per fracture stage may be undermining the ability to dewater created fractures. The insights derived from this research reveal important differences in load recovery behavior that may impact well performance in different shale formations and highlight how effectively the wells are draining the stimulated shale volume.
Adaptive Generalized Multiscale Model Reduction Techniques for Problems in Perforated Domains
(2018-05-24) Wang, Yating; Efendiev, Yalchin; Chung, Eric; Gildin, Eduardo; Lazarov, Raytcho; Titi, Edriss
Multiscale modeling of complex physical phenomena in many areas, including hydrogeology, material science, chemistry and biology, consists of solving problems in highly heterogeneous porous media. In many of these applications, differential equations are formulated in perforated domains which can be considered as the region outside of inclusions or connected bodies of various sizes. Due to complicated geometries of these inclusions, solutions to these problems have multiscale features. Taking into account the uncertainties, one needs to solve these problems extensively many times. Model reduction techniques are significant for problems in perforated domains in order to improve the computational efficiency. There are some existing approaches for model reduction in perforated domains including homogenization, heterogeneous multiscale methods and multiscale finite element methods. These techniques typically consider the case when there is a scale separation or the perforation distribution is periodic, and assume that the solution space can be approximated by the solutions of directional cell problems and the effective equations contain a limited number of effective parameters. For more complicated problems where the effective properties may be richer, we are interested in developing systematic local multiscale model reduction techniques to obtain accurate macroscale representations of the underlying fine-scale problem in highly heterogeneous perforated domains. In this dissertation, based on the framework of Generalized Multiscale Finite Element Method, we develop novel methods and algorithms including (1) development of systematic local model reduction techniques for computing multiscale basis in perforated domains, (2) numerical analysis and exhaustive simulation utilizing the proposed basis functions, (3) design of different applicable global coupling frameworks and (4) applications to various problems with challenging engineering backgrounds. Our proposed methods can significantly advance the computational efficiency and accuracy for multiscale problems in perforated media.
Adaptive Mesh Refinement and Coarsening for Compositional Reservoir Simulation
(2016-11-28) Gonzalez Abad, Karin Gabriela; Barrufet, Maria A; Blasingame, Thomas A; King, Michael; Gildin, Eduardo
This dissertation introduces a new method to create adaptive mesh refinement and coarsening in compositional reservoir simulation. The methodology targets individual cells for refinement based on forecasted compositional fronts calculated using streamlines and the analytical convection-dispersion transport equation. Quadtree decomposition determines the optimal spatial discretization across the simulation grid using dynamic and static reservoir properties. Application of the new approach results in improved computational performance without compromising the accuracy of phase behavior. Current dynamic gridding implementations have rigid schemes, posing two major limitations: cell refinement size is a pre-determined input value and compositional maps from the previous time step define the refinement region. This solution leads to suboptimal modeling due to time-lagging refinement and lack of grid adaptability in heterogeneous reservoirs and/or fast-moving compositional fronts. The new methodology overcomes these limitations by combining streamline and particle trajectory to forecast the injection front location and adapt grid sizes in advance. Tracking compositional variations starts by calculating fluxes for all cells using the finite-difference solution. Next, Pollock’s tracing method allows reducing the 3-dimensional model into a series of 1-dimensional streamlines, while the convection-dispersion equation forecasts future compositions, shape, and location of injection front along each streamline trajectory. Finally, quadtree decomposition analyzes the homogeneity of dynamic and/or static properties (e.g., composition, pressure, permeability, facies) to determine if a volume can be represented by a single gridblock or if it requires refinement to preserve spatial details. Grid discretization is dynamic over time, refining cells requiring high-resolution and/or coarsening those with low variation. A mechanistic model with CO2 injection served to evaluate the methodology. The fluid was modeled with five pseudo-components and the Peng-Robison equation of state with volume translation to improve volumetric predictions. The new approach reduced the total number of cells required to model miscible injection by continuously creating adaptive grids that represent the advancement and shape of the injection front. Results showed a reduction in computational cost between 30-63% over a static fine grid without compromising the representation of compositional mixing phenomena and production forecast.
Adaptivity and Online Basis Construction for Generalized Multiscale Finite Element Methods
(2017-07-18) Leung, Wing Tat; Efendeiv, Yalchin; Chung, Eric; Gildin, Eduardo; Howard, Peter; Lazarov, Raytcho
Many problems in application involve media with multiple scale, for example, in composite materials, porous media. These problems are usually computationally challenging since fine grid computation is extremely expensive. Therefore, one may need to develop a coarse grid model reduction for this type of problems. In this dissertation, we will consider a multiscale method called generalized multiscale finite element method (GMsFEM). GMsFEM follows the framework of multiscale finite element method. Instead of using one basis function per coarse grid node, GMsFEM uses several basis functions for one coarse grid node. Since the media is highly heterogeneous and may involves high contrast, having more than one basis function per node is important to reduce the error significantly. Due to the varying heterogeneity in the domain, we may require different numbers of basis functions in different regions. Then the question is how to determine the number of basis functions in each region. In this dissertation, we will discuss an adaptive enrichment algorithm for enriching basis functions for the regions with large error. We will consider two different types of basis function for enrichment. One is using the pre-computed offline basis functions. We call this method offline adaptive enrichment. The other method uses online constructed basis functions called online adaptive enrichment. In applications, non-conforming basis functions can give us more flexibility on gridding. The discontinuous Galerkin method also makes the mass matrix block diagonal, which enhances the computation speed in solving time-dependent problem with an explicit scheme. In this dissertation, we will discuss offline and online adaptive methods for the generalized multiscale discontinuous Galerkin method (GMsDGM). We will also discuss using GMsDGM for simulating wave propagation in heterogeneous media.
Advances in Data-Driven Modeling and Global Optimization of Constrained Grey-Box Computational Systems
(2020-03-24) Beykal, Burcu; Pistikopoulos, Efstratios N; El-Halwagi, Mahmoud M; Gildin, Eduardo; Hasan, M. M. Faruque
The effort to mimic a chemical plant’s operations or to design and operate a completely new technology in silico is a highly studied research field under process systems engineering. As the rising computation power allows us to simulate and model systems in greater detail through careful consideration of the underlying phenomena, the increasing use of complex simulation software and generation of multi-scale models that spans over multiple length and time scales calls for computationally efficient solution strategies that can handle problems with different complexities and characteristics. This work presents theoretical and algorithmic advancements for a range of challenging classes of mathematical programming problems through introducing new data-driven hybrid modeling and optimization strategies. First, theoretical and algorithmic advances for bi-level programming, multi-objective optimization, problems containing stiff differential algebraic equations, and nonlinear programming problems are presented. Each advancement is accompanied with an application from the grand challenges faced in the engineering domain including, food-energy-water nexus considerations, energy systems design with economic and environmental considerations, thermal cracking of natural gas liquids, and oil production optimization. Second, key modeling challenges in environmental and biomedical systems are addressed through employing advanced data analysis techniques. Chemical contaminants created during environmental emergencies, such as hurricanes, pose environmental and health related risks for exposure. The goal of this work is to alleviate challenges associated with understanding contaminant characteristics, their redistribution, and their biological potential through the use of data analytics.
Advancing Multiparametric Programming for Model Predictive Control
(2020-02-21) Katz, Justin; Pistikopoulos, Efstratios N; Kravaris, Costas; Gildin, Eduardo; Kwon, Joseph S
Model predictive control provides the optimal operation for chemical processes by explicitly accounting for the system, constraints, and costs. In an online setting, developing the implicit optimal control action under time consideration is non-trivial. Over a decade ago, it was demonstrated through multiparametric programming that the implicit control law defining the model predictive controller can be determined explicitly, once and offline. The benefit of such an approach is the (i) improved online computational time, (ii) the development of the offline map of solution \textit{a priori}, and (iii) the derivation of the optimal control laws under any state variation. In recent years there has been a significant push for the development of novel algorithms and theoretical advancements for multiparametric model predictive control. These algorithms and theoretical underpinnings have expanded the problem classes that are solvable and improved the computational efficiency. However, there is still a need to provide analysis for formulations based on different surrogate models, and to tackle large scale multiparametric model predictive control problems. In this dissertation, the research focus is (i) the inclusion of a new surrogate modeling technique from the machine learning community, (ii) developing a criterion to compare multiparametric model predictive control formulations based on different surrogate models, (iii) the development of an algorithm to solve large scale multiparametric optimization problems, and (iv) improving the online computational performance of online solvers via multiparametric programming. To this end, tools from data science, computational geometry, and the operations research community contributed greatly to the results presented in this work. This research is verified via the optimal operation of chemical engineering processes and the efficacy of the developed algorithms is demonstrated on computational studies.
AIMR (Azimuth and Inclination Modeling in Realtime): A Method for Prediction of Dog-Leg Severity based on Mechanical Specific Energy
(2013-08-13) Noynaert, Samuel F; Holditch, Stephen A.; Schubert, Jerome J.; Barrufet, Maria A.; Gildin, Eduardo
Since the 1980’s horizontal drilling has been a game-changing technology as it allowed the oil and gas industry to produce from reservoirs previously considered marginal or uneconomic. However, while it is considered a mature technology, directional drilling is still done in a reactive fashion. Although many directional drillers are quite adept at predicting the directional response of the bottomhole assembly (BHA) in a given well, the ability to manage all of the drilling parameters on a foot by foot basis while accurately predicting the effects of each parameter is impossible for the human brain alone. Given current rig rates, any amount of increased slide time and its reduced ROP which occurred due to poorly predicted directional response can result in a significant economic impact. There exist many measured parameters or system inputs which have been proven to affect the directional response of a drilling system. One parameter whose effect has not been investigated is mechanical specific energy or MSE. MSE is measure of how efficient the drilling process is in relation to rate of penetration. To date, MSE has primarily been used with for vibration analysis and rate of penetration optimization. The following dissertation covers research into the effect of MSE on the overall wellbore direction change or dog-leg severity. Using published experimental data, a correlation was developed which shows a clear relationship between the dog-leg severity, rate of penetration (ROP) and MSE. The correlation requires only a few hundred feet of drilling before it is able to be tuned to match an individual well’s results. With minimal tuning throughout the drilling of a well, very good results can be obtained with regards to forecasting dog-leg severity as the wellbores were drilled ahead. The correlation was tested using data from multiple, geo-steered wells drilled in a shale reservoir. The analysis of the correlation using real-world data proved it to be a robust and accurate method of predicting the magnitude of dog-leg severity. The use of this correlation results in a smoother wellbore, drilled with a faster overall ROP with a better chance of staying within the geologic targets.
Analysis of Gas Production from Hydraulically Fractured Wells in Naturally Fractured Reservoir Using Source Function Method
(2017-12-07) Hwang, Yun Suk; Zhu, Ding; Schechter, David; Gildin, Eduardo; Battle, Guy
According to the 2014 EIA statistics, natural gas production from shale and tight oil plays accounted for 48% of US natural gas production and this number is expected to grow to 69% in 2040. Natural fractures are commonly observed in these unconventional reservoirs. Multi-stage hydraulic fracturing in horizontal wells has been applied to develop these shale/tight sands. Natural fractures could be open during treatment or conductive even before treatment, providing a larger drainage by creating a complex network. It still remains a challenge to reasonably predict well performance in such a complex system, especially by honoring the distribution of natural fractures explicitly. This study presents a methodology based on Green’s source function and Fractal discrete fracture network (FDFN) model. Slab source is a plane source with finite thickness, which is a novel approach of classic source function by reducing the erroneous integration. The hydraulic and natural fractures together are represented by independent slab sources, and their influence on each other is considered, which is more realistic than summing the flow from each fracture as total flow. FDFN model was used to generate realistic natural fracture maps. Production from adsorbed gas, common in shale reservoirs, is also modeled using modified material balance equation. I applied our model to estimate the multi-stage hydraulic fractured horizontal gas well performance in synthetically generated naturally fractured reservoirs. An extended number of natural fractures were handled by introducing several approaches to speed up the calculation. A parametric study was conducted to delineate important parameters affecting well performance. Simulation results indicated that conductive natural fracture largely influence gas production in unconventional reservoirs. The characteristics of natural fractures, such as density, length and interaction with hydraulic fractures were found to be controlling parameters. It was also found that the inclusion of adsorbed gas could result in the total gas production increase up to 25%. Also, comparisons are provided with published or commercially available numerical and analytical approaches to verify the methodology of this study. The novelty of the method is in the ability to respect the previously identified fracture distribution explicitly, either hydraulic or natural, even if the fractures are non-orthogonal to the horizontal wellbore. Since the approach is semi-analytical, it is easy to use and solves the problem in reasonable time using standard computers.
Analysis of Hydraulic Fracture Propagation and Well Performance using Geomechanical Models and Fast Marching Method
(2017-05-01) Huang, Jixiang; Datta-Gupta, Akhil; King, Michael J; Gildin, Eduardo; Efendiev, Yalchin
Successful exploitation of unconventional resource plays relies on the massive hydraulic fractures which provide high conductive paths and large contact area between formation and wellbore. The pursuit of efficiency and cost savings drives the industry to implement the strategies that utilize more closely spaced hydraulic fractures, as well as multiple horizontal wells with reduced spacing, to maximize the production from unconventional reservoirs with ultra-low permeability. One rising challenge from this trend is to find the optimized spacing between fracture clusters, fracture stages, and fractured horizontal wells so that the potential fracture interference could be minimized. This interference could occur at different scales within lifecycle of exploration, from stress interference in completion stage to pressure interference in production stage. Thus, to systematically study these issues, both geomechanical model and reservoir model are required. In this dissertation, a finite element based geomechanical model and a fast marching based reservoir model are customized to address these emerging problems in unconventional reservoir development. First, we present a comprehensive study of various factors that affect the performance of refracturing operation, such as fracturing spacing, permeability, proppants and refracturing time, by using a cohesive zone finite element based model that can capture the effect of depletion on fracture propagation. The well performance are evaluated under two different refracturing designs: refracturing new or existing perforations. Based on the simulation results, their respective suitability have been concluded. Second, we integrate fracture propagation, reservoir flow and wellbore hydraulics to evaluate the stress shadow effect and efficiency of limited entry perforations during multiple simultaneous fracture propagation within a single fracture stage. Simulation results provide insights to the selection of operational parameters such as cluster spacing, number of clusters and perforations, which can be modified accordingly to deal with the fracture interference and thus promote the uniform stimulation in the formation. Last, to study the production interference between wells, on top of current fast marching based reservoir simulation workflow, we proposed an approach to extend its applicability from transient to boundary-dominated flow regime, as well as a new partition method to identify the respective drainage volume of individual well. This partition criterion utilizes asymptotic pressure solution and results in a good approximation to the conventional streamline tracing method. The supremacy of numerical efficiency has been further demonstrated with numerical experiments.
Analysis of Mud Motor Stalls and Its Impact on Performance in High Temperature Unconventional Reservoirs
(2018-07-31) Hopkins, Zachary Ira; Noynaert, Samuel F; Gildin, Eduardo; Kuttolamadom, Matthew
The objective of this study was to perform analysis in to the mechanisms of motor failure in the curve lateral portion of an operator’s high temperature Eagleford shale wells. This was achieved through multiple high frequency downhole sensors that collected drilling dynamics and vibration data. The high frequency downhole sensors were able to provide evidence of 21 motor stalls which began in the last 1,200 ft of the lateral section. The motor stall discussed in this paper appeared to be caused by a sudden increase in weight applied downhole, which caused the torque required to rotate the bit to exceed the torque that was supplied to the bit. The stall was only released once the string shortened from the continued top drive rotation which allowed the weight applied and subsequent indentation depth of the bit to be reduced and allow the bit to rotate. Additionally the pressure required to stall the motor was decreased as the motor experienced stalls and became damaged. Surface measurements during the stall did not reflect the true conditions downhole due to the nature of the 1 Hz recording capability. Surface measured differential pressures were hundreds of psi below actual downhole differential pressure, and did not demonstrate the extent of the damage that the motor was seeing. Motor damage and fatigue was correlated through the use of MSEbit and pressure normalized rate of penetration (ROP). These metrics were able to provide the approximate depth of the onset of motor stalls and show the progression of performance loses throughout the lateral. Motor stalls cannot be eliminated completely but design changes can be made to lessen the frequency, and improve motor life. Design changes including bottom hole assembly (BHA) design, motor configurations, as well as a real-time deration practice are presented in a workflow to manage motor stalls
Analysis of Twitter Hashtags' Geographic Propagation
(2016-01-04) Ibrahim Abdelhalim, Ashraf A.; Caverlee, James; Furuta, Richard; Gildin, Eduardo
The goal of this work is to study the geographic propagation patterns of Twitters' hashtags. In order to analyze the hashtags' diffusion patterns, we look at the globe as a graph consists of a large grid of locations and use two different approaches to study the hashtags' behaviour. The first approach is to consider the locations on the global grid as variables (or features) and the individual hashtags as the examples. This viewpoint of our dataset allows us to perform dimensionality reduction techniques to reduce the size of the dataset without much loss of information and to identify the more influential locations. The second methodology is to transform the global grid into an undirected weighted graph and compute the influence curves associated with the hashtags propagation and their properties. We show that the influence curves of different classes of hashtags have similar patterns. In addition, we show that the influence curve can be approximated adequately using only six Chebychev polynomials.
Analytical and Numerical Solutions for the Case of a Horizontal Well with a Radial Power-Law Permeability Distribution--Comparison to the Multi-Fracture Horizontal Case
(2013-02-08) Broussard, Ryan Sawyer; Blasingame, Thomas A; Moridis, George J; Gildin, Eduardo; Kronenberg, Andreas; Valko, Peter P
In this work, I present the development of analytical solutions in the Laplace domain for a fully-penetrating, horizontal well producing at a constant flow rate or constant wellbore pressure in the center of a composite, cylindrical reservoir system with an impermeable outer boundary. The composite reservoir consists of two regions. The cylindrical region closest to the wellbore is stimulated, and the permeability within this region follows a power-law function of the radial distance from the wellbore. The unstimulated outer region has homogeneous reservoir properties. The current norm for successful stimulation of low permeability reservoir rocks is multi-stage hydraulic fracturing. The process of hydraulic fracturing creates thin, high permeability fractures that propagate deep into the reservoir, increasing the area of the rock matrix that is exposed to this low-resistance flow pathway. The large surface area of the high conductivity fracture is what makes hydraulic fracturing so successful. Unfortunately, hydraulic fracturing is often encumbered by problems such as high capital costs and a need for large volumes of water. Therefore, I investigate a new stimulation concept based upon the alteration of the permeability of a large volume around the producing well assembly from its original regime to that following a power-law function. I evaluate the effectiveness of the new concept by comparing it to conventional multi-stage hydraulic fracturing. The results of this investigation show that the power-law permeability reservoir (PPR) has a performance advantage over the multi-fractured horizontal treatment (MFH) only when the fracture conductivity and fracture half-length are small. Most importantly, the results demonstrate that the PPR can provide respectable flow rates and recovery factors, thus making it a viable stimulation concept for ultra-low permeability reservoirs, especially under conditions that may not be conducive to a conventional MHF treatment.
Application of a Lattice Boltzmann - Molecular Dynamics Simulation to Pore-Scale Modeling of Fluid Flow in Shale
(2017-12-11) Madiebo, Ijeoma Kingsley; Nasrabadi, Hadi; Gildin, Eduardo; Barrufet, Maria; King, Michael
In this work, a modified workflow for incorporating molecular effects into a macroscopic fluid flow model via a mesoscopic transition model to more uniformly ascertain transport properties during pore scale analysis, is presented and validated. A combined lattice Boltzmann-molecular dynamics (LBMD) simulation approach to address this issue is employed. The hydrocarbon and shale system taken under consideration here were modeled in molecular form as n-octane and silica respectively. The n-octane was set up using the united atom (UA) model. The interaction forcefields primarily employed for the MD system included the standard Lennard-Jones potential, the transferable potentials for phase equilibria (TRAPPE) and the Buckingham potential. The properties studied here were the volumetric flux per unit area, apparent permeability and general fluid dynamics for hydrocarbon flow in the system. Results from the MD showed a non-linear relationship between the force and the noctane density. This force was then incorporated into the LB system which already had a Peng-Robinson equation of state embedded into a fluid-fluid particle interaction forcing function. With the variation of the Knudsen number which accounts for slip effect (or gradual deviation from continuum), the fluid dynamics of the system was then modeled. Analysis showed that the slip effect as a function of the Knudsen regime was non-linearly proportional to the volumetric flux per unit area, and thus the deduced permeability of the fluid. The LBMD prediction of apparent permeability showed good agreement with established apparent permeability correlations for shale found in literature. Good qualitative agreement with flow dynamics was also achieved when compared to lab-on-a-chip experiment, representative of nanoscopic shale media and with all results obtained without parameter fitting. This work aims to extend current understanding of fluid flow behaviour below the continuum regime and improve the accuracy of apparent permeability computation on tight rock geometric imagery, typical of shale rock physics when producing hydrocarbons from shale gas reservoirs. This will be fundamental in the development of a more robust and complex pore-scale modeling framework for simulating more accurate subsurface flow dynamics.
Application of Ensemble-based Optimization on UNISIM-I
(2017-04-27) Plukmonton, Pattanapong; Gildin, Eduardo; Nasrabadi, Hadi; Medina-Cetina, Zenon
In reservoir management, it is challenging to obtain an efficient production schedule and maximize the profits. An optimization workflow is usually used in maximizing/minimizing the production objective. However, production optimization is not an easy task and could be time-consuming since the reservoir and the production optimization itself consist of complex subsystems and uncertainties. Thus, many studies have been done to propose optimization methods that are efficient and yet practical in finding the optimal strategy. Most of these methods usually focus on the gradient-based approaches, where the information from gradients of the objective function with respect to control parameters is used in finding the optimal solutions. One of the gradient-based methods that recently has gained popularity in petroleum production optimization is Ensemble-based Optimization (EnOpt). In EnOpt, the gradient is approximated using a linear regression between an ensemble of control vectors and their corresponding objective function values. Thus, the computational cost of the method relies on the number of realizations in the control ensemble and is nearly independent of the number of control parameters. Moreover, the EnOpt can be used with any reservoir simulator without modification to the simulator. Many publications have demonstrated that EnOpt gave a good optimized-result on different reservoir models and recovery techniques. In this thesis, we study the benefits of the EnOpt applied to waterflooding optimization problems using realistic reservoir data. In particular, the EnOpt is used to optimize the waterflooding process in a benchmark reservoir, namely UNISIM-I. The objective of this optimization is to maximize the expected net present value (NPV) over 20 years of production. The control parameters are injection and production rates in injector and producer wells. We consider two optimization problems: random initial control settings and extended production from the production history. The EnOpt was successful in finding optimal solutions in both cases with significantly cheaper computational cost required in gradient calculations. In addition, we study the effect of discount rate use in calculating the NPV: the short-term EnOpt uses high discount rate, whereas the long-term EnOpt sets discount rate equal to zero. The different discount rates result in different optimal solutions. The high discount rate results in an increase of cash flow in the early stages of the production time while low to no discount rate results in an increase of cumulative cash flow throughout the production time.
Application of Non-intrusive Reduced Order Modelling on Reservoir Simulation Using Radial Basis Function
(2021-04-26) Dong, Xiaomu; Gildin, Eduardo; Medina-Cetina, Zenon; Moridis, George
The aim of this thesis is to assess the efficiency of a new data-based method of Non-Intrusive Reduced Order Modeling (NIROM) which can be applied on commercial reservoir simulators and discuss the performance of the NIROM method on reducing the computational running time within an acceptable compromise in accuracy. NIROM methods are usually constructed by a combination of machine learning techniques and projection-based ROM methods. As opposed to projection-based ROM, such as the proper orthogonal decomposition (POD), NIROM methods treats the non-linear equation as a black box and its approximation is performed by the Reduced-based Function (RBF). In this case, NIROM manifests as a data-driven methodology, whereby the only information the users need is the snapshots of states, e.g., pressures, saturations, temperature, which are output by the high-fidelity model. This means NIROM can be applied to any commercial reservoir simulator. In this work, two cases are used to investigate the abilities and limitations of the NIROM method. The first case is the UNISIM-I-D three-phase isotropic heterogeneous model without considering heat flow. In this case, the proposed method is applied to obtain an approximation of the pressure and three-phase saturation field within 20 years. The second case is a 3-dimensional geothermal reservoir with injection and production wells working at the same time. In this model, cold water is injected from the injection wells and hot water is produced from the production wells. Pressure field and temperature field are predicted within 5 years. In both cases, the NIROM method is proved to be able to significantly reduce the computational time of running the simulation after an offline training and an offline optional validation process while remain the accuracy of the simulation within certain range. The accuracy of the proxy obtained is case dependent and varies when different timestep interval of the training data set is applied.
Application of Radial Basis Functions in History Matching of Production Data
(2020-03-11) Zubarev, Denis Igorevich; Datta-Gupta, Akhil; Lee, William J; Gildin, Eduardo; Mallick, Bani K
A successful development of oil and gas reservoirs requires a deep understanding of key subsurface complexities, possible development outcomes, and execution of optimal development actions. This is achieved through the utilization of detailed reservoir models that integrate all available data and modern reservoir simulation, which incorporates the whole spectrum of physical and chemical processes associated with development activities. To further improve reservoir models, we condition them to the collected production data through the history matching process. This process employs sequences of reservoir simulation runs to searches for changes in reservoir description that can improve reservoir model prediction accuracy. Over the past decades, multiple assisted history matching algorithms were proposed in order to speed up the process and extend it to handling multiple alternative realizations. However, the history matching process is still highly computationally expensive and associated with compromises in the robustness of the algorithms that have to be made to keep it practical. In this work, we investigate the application of fast Radial Basis Function (RBF) proxy models as a substitute for reservoir simulation in history matching. First, we review the general background in proxy modeling and state-of-the-art developments in the RBF application in engineering optimization problems. We also review the background and recent progress in the development of some popular ensemble-based and multi-objective algorithms and discuss outstanding challenges in their practical implementation. Then we introduce a modified ensemble Kalman filter method (RBF-EnKF) that utilizes RBF proxy models as a partial substitute to numerical simulation. This method improves the accuracy of the cross-covariance estimation between the model parameters and model dynamic responses for small ensemble size. To achieve higher accuracy of RBF proxy models, an improved heterogeneous and anisotropic formulation for basis functions scaling was implemented. The proposed method was tested using a synthetic case and showed significant improvement in cross-covariance estimation and good history matching results. Next, we extended the proposed RBF-EnKF algorithm for application to practical size history matching problems. This was achieved by introducing Grid Connectivity Transforms (GCT) parametrization to convert spatial variables to a set of discrete inputs for proxy models and iterative sensitivity-based GCT coefficients selection to further reduce the number of proxy model input parameters. An extended version of the algorithm outperformed EnKF with localization and showed a close match to conventional EnKF with significantly larger ensemble size in history matching of Brugge and Norne field models. Finally, we proposed a modified of Multi-objective Evolutionary Algorithm based on decomposition and dominance (MOEA/DD) that utilizes RBF proxy models to reduce computational requirements. The proposed RBF-MOEA/DD algorithm adopted the GCT parametrization and Gradual Deformation approach to enable history matching of model spatial variables such as absolute permeability. Optimization workflow was modified to incorporate RBF proxy modeling with adaptive and sparsity-based iterative improvement. The proposed method was applied to the Brugge case history matching and showed the quality of the match and computational improvement similar to the RBF-EnKF method.
Application of Simple Smart Logic for Waterflooding Reservoir Management
(2016-12-06) Dall'Aqua, Marcelo Jacques; Gildin, Eduardo; Killough, John E; Barrufet, Maria A
A simple smart logic for controlling inflow control valves (ICV) in waterflooding reservoir management is implemented and analyzed, with the final objective of improving the long term financial return of a petroleum reservoir. Such a control is based in a reactive simple logic that responds to the watercut measured in the ICV. Basically, when the watercut increases, the ICV is set to close proportionally. For comparison purposes, four strategies are presented: base case scenario with conventional control, the best completion configuration found by trial-and-error, the reactive control, and a deterministic optimal control based on Nonlinear Gradient Method with adjoint-gradient formulation is shown for comparison purposes. Finally, all four strategies are tested again in different reservoir realizations in order to mimic the geological uncertainties. Two different synthetic reservoir models were studied. First, a simple cube with a five-spot well configuration, in which the permeability field has a horizontal pattern defined by lognormal distributions. The second model is a benchmark proposed by the Dutch university, TU delft, with 101 channelized permeability fields representing river patterns. For the first model, no significant relative gain is found neither in the variable control nor in the optimal control. Manly because of the high homogeneity of the reservoir models. Therefore, no intelligent completion is recommended. On the other hand, for the second and more complex case, the results indicate an expressive relative gain in the use of simple reactive logic. Besides, this type of control achieves results nearly as good as the optimal control. The test in different realizations, however, shows that reservoir characterization is still a key part of any attempt to improve production. Although the variable reactive control is semi-independent, with action being taken based on measurements, some parameters need a priori model to be tuned.
Application of the Discrete Element Method to Study the Effects of Occlusion Interfaces in Shale
(2016-12-06) Xie, Antu; Blasingame, Thomas A; Moridis, George J; Gildin, Eduardo; Medina-Cetina, Zenon
The multiscale heterogeneity of ultratight shale rocks leads to the interesting yet pragmatic question of whether or not its microscale features may be used to predict macroscopic fracture behavior. Understanding the dominant parameters of microfracturing in these structures may help both with understanding the evolution of macroscopic fractures as well as permeability changes in shales. While many macroscopic analogs exist to correlate shale composition with fracture properties, few studies have examined the role that shale microstructure has on fracturing. In this thesis, I first describe the method I developed to use SEM/EDS data from shale images to set up discrete element method simulations. I then explore the role of shale microstructures under standard uniaxial fracturing and what effect it may have on macroscopic material properties and if there is a special role that interfaces between different materials may play during fracturing in shales. Using the data provided and the simulation results, I demonstrate the qualitative role that the interfaces between different materials play during both compressive and tensile fracturing.

Browsing by Author "Gildin, Eduardo"

Results Per Page

Sort Options