Advanced Visualization for Reservoir Simulation

Fracturing modeling methods developed for conventional hydraulic fracturing are now being used for unconventional fracturing in waterfracs, water or steam flooding, produced water reinjection, etc. A common feature of these unconventional fracturing processes is the strong interaction among fracture propagation (often with high 3D fluid leakoff), reservoir flow, changes in stresses (poroelastic and thermoelastic effects), and permeability and porosity changes (geomechanical effects) around the fracture. Conventional fracturing models are inadequate under such conditions; moreover, they are also disconnected from well performance forecasting, which makes integrated data analysis difficult. Therefore, it is necessary to seek a new modeling concept including all these mechanisms and their mutual influences. This paper describes a method to model hydraulic fracturing with dynamic transmissibility multipliers based on coupled reservoir and geomechanics simulation. The method is the first step in developing a fracturing model fully coupled into reservoir and geomechanics simulation, where the fracture geometry will be also internally calculated from the fracture face displacements in the coupled FEM geomechanical module. The method described here ignores fracture volume but focuses on the effect of fracture on fluid flow and geomechanics in reservoir by introducing pressure/stress dependent dynamic transmissibility multipliers and treating them as a property of the matrix. This approach allows modeling fracture propagation, dynamical multiphase fracture conductivity, clean-up, and pre- and post-frac well performance in a changing stress, pressure and temperature environment, all in a unified manner. This paper also discusses the strategy of coupling hydraulic fracture propagation, reservoir and geomchanics simulation, resulting in a method to improve the stability of the dynamic hydraulic fracture propagation in coupled reservoir and geomechanics simulation. The case studies in this paper confirm that the strategy and the method to model dynamic hydraulic fracture propagation coupled with reservoir and geomechanics simulation is feasible, flexible and reliable. It is easy and convenient to implement in conventional reservoir simulators and coupled reservoir and geomechanics simulators (such as GEOSIM).

Development of a reservoir simulation model involves collection and analysis of an enormous amount of reservoir and well data for initial input. Computer graphics is the most effective tool for geologists and engineers to visualize, verify, and analyze the input data. After each simulation run, simulation engineers then have to depend heavily on computer graphics to analyze an even more intimidating amount of simulation results data. This is especially true at Saudi Aramco because of the large simulation models used to simulate the world's largest hydrocarbon reservoirs. Saudi Aramco's EXPEC Computer Center supports a wide spectrum of computer graphics applications to aid the exploration and engineering staffs through the entire cycle of reservoir characterization. The graphics applications help during the pre-simulation model data analysis as well as in the post-simulation results analysis. Larger and more accurate simulation models are currently being developed In Saudi Aramco. Computer technology (both software and hardware) needs to advance continually such that computer graphics can be utilized as an effective and efficient engineering tool by reservoir simulation professionals. This paper addresses Saudi Aramco's experience in dealing with very large simulation models and using computer graphics as an engineering tool. The program efficiency, device dependence, and data management problems will be discussed. The computer speed, data storage and access methods, data transmission speed, and graphics devices will also be detailed.

▪ Abstract Efficiently and accurately solving the equations governing fluid flow in oil reservoirs is very challenging because of the complex geological environment and the intricate properties of crude oil and gas at high pressure. We present these challenges and review successful and promising solution approaches. We discuss in detail the modeling of fluid flow in reservoirs with strongly varying rock properties. This requires subgrid-scale models that accurately represent the flow physics due to fine-scale fluctuations. A second focus is on the complex multiphase, multicomponent systems that describe miscible gas injection processes for enhanced oil recovery and CO2 sequestration.

Low permeability reservoirs occupy a large proportion in the remaining reserves all over the world. So, it is urgent to research the seepage law of fluid flow in low permeability reservoirs intensively. Fluid flow in these reservoirs deviates from the classical Darcy law and conforms to the one of nonlinear seepage instead. Horizontal well is one of the key technologies to exploit the low permeability reservoir. According to this characteristic, this paper presents a mathematical model for the analysis of pressure behaviour of a horizontal well in low permeability reservoirs, which takes into consideration of threshold pressure gradient (TPG). The solution of the non-homogeneous Bessel equation is obtained through Laplace and Fourier cosine transform and the Green Function methods. Integrating along the horizontal well and using Stehfest numerical inversion method, the dimensionless pressure response and type curves of horizontal well with the consideration of TPG are obtained, and the sensitivities of related parameters are discussed. The results of this paper have significance in the well test analysis, production performance prediction and production data analysis of horizontal well in low permeability reservoirs. [Received: June 17, 2013; Accepted: July 21, 2013]

The use of permeability tensors is required when modelling fluid flow in anisotropic and heterogeneous reservoirs presenting multiple zones of directional permeability, or those categorized as naturally fractured reservoirs. A general procedure for characterizing complex reservoirs utilizing their permeability tensor is being developed by integrating data and methods from different disciplines. Permeability tensors for geologically defined fracture patterns are derived, and finally these small-scale descriptors are incorporated into a reservoir simulation program capable of handling full tensor permeabilities. The application and convenience of the method presented in this paper is illustrated with a field example from a naturally fractured reservoir. Introduction For many years, substantial research has been conducted in the areas of geosciences and engineering in order to characterize naturally fractured reservoirs. Numerous approaches have been presented to properly overcome this difficult task. Geoscientists have focused their research towards understanding the process of fracturing (rock mechanics), and the subsequent description of fracture characteristics such as density and orientation. Engineers, on the other hand, have focused their attention to the description of the fluid flow in the fracture systems, and in the development of accurate models (reservoir simulators) to reproduce the history, and predict the hydrocarbon production, for these complex systems. One of the first matrix-fracture models was presented by Warren and Root(1), who presented an idealized sugar cube model with two classes of porosity, a primary porosity that is intergranular, and a secondary porosity that is induced by fractures. Even though the sugar cube model has been widely accepted as the forerunner of the modern interpretation of dual-porosity systems, its limitations in describing the behaviour of some complex fractured reservoir systems have been observed. It is now well known that almost all fracture systems are much more complicated than the suggested model of three orthogonal sets of uniform fractures. One of the most important factors that has been identified as a necessary addition to improve the overall description of such complex reservoirs is the definition of a nine-component permeability tensor for the fracture system. This tensor is used to model fluid flow in complex reservoirs with multiple zones of directional permeability, where the orientation and magnitude of the principal permeabilities may vary between different zones in the reservoir. Snow(2, 3) studied the convenience of using mathematical equivalents of parallel plate openings to simulate fractures dispersed in orientation, distributed in aperture, and of arbitrary spacing. Models for fractured media which contained any number of planar conductors of any orientation and any fine aperture were presented. A key assumption was that all of the conduits have smooth parallel plane walls of indefinite extent (infinite fractures), and an rbitrary aperture. As a result, a permeability tensor could be obtained by superposition of contributions due to the fractures, and due to the permeable matrix. Long et al.(4, 5) addressed the more realistic scenario of finite or discrete fracture systems, where properties such as shape, orientation and location of the fractures in an impermeable matrix were considered to be random variables.

Practical workflow to improve numerical performance in time-consuming reservoir simulation models using submodels and shorter period of time

Summary Accounting for poro-mechanical effects in full-field reservoir simulation studies and uncertainty quantification workflows is still limited, mainly because of their high computational cost. We introduce a new approach that couples hydrodynamics and poro-mechanics with dual-porosity flow diagnostics to analyze how poro-mechanics could affect reservoir dynamics in naturally fractured reservoirs without significantly increasing computational overhead. Our new poro-mechanically informed dual-porosity flow diagnostics account for steady-state and single-phase flow conditions in the fractured medium while the fracture-matrix fluid exchange is approximated using a physics-based transfer rate coefficient, which models two-phase flow using an analytical solution for spontaneous imbibition or gravity drainage. The deformation of the system is described by the dual-porosity poro-elastic theory, which is based on mixture theory and micromechanics to compute the effective stresses and strains of the rock matrix and fractures. The solutions to the fluid flow and rock deformation equations are coupled sequentially. The governing equations for fluid flow are discretized using a finite-volume method with two-point flux-approximation while the governing equations for poro-mechanics are discretized using the virtual element method. The solution of the coupled system considers stress-dependent permeabilities for fractures and matrix. Our framework is implemented in the open-source MATLAB Reservoir Simulation Toolbox (MRST). We present a case study using a fractured carbonate reservoir analog to illustrate the integration of poro-mechanics within the dual-porosity flow diagnostics framework. The extended flow diagnostics calculations enable us to quickly screen how the dynamics in fractured reservoirs (e.g., reservoir connectivity, sweep efficiency, and fracture-matrix transfer rates) are affected by the complex interactions between poro-mechanics and fluid flow where changes in pore pressure and effective stress modify petrophysical properties and hence affect reservoir dynamics. Because of the steady-state nature of the calculations and the effective coupling strategy, these calculations do not incur significant computational overheads. They provide an efficient complement to traditional reservoir simulation and uncertainty quantification workflows because they enable us to assess a broader range of reservoir uncertainties (e.g., geological, petrophysical, and hydromechanical uncertainties). The capability of studying a much broader range of uncertainties allows the comparison and ranking from a large ensemble of reservoir models and select individual candidates for more detailed full-physics reservoir simulation studies without compromising on assessing the range of uncertainties inherent to fractured reservoirs.

The computation of flowing-well bottomhole pressure from the pressure of the block containing the well or of well flow rate when the flowing bottomhole pressure is specified are important considerations in reservoir simulation. While this problem has been addressed by several authors, some important aspects of the problem are not treated adequately in the literature.We present an analytical method for computing the well-block factors (constants of the PI) for a well located anywhere in a square or rectangular block (aspect ratio between ½ and 2). Equations for well geometric factors and well fraction constants are given for gridblocks of various types, containing a single well, encountered in reservoir simulation studies. The equations given in this paper can be used for both block-centered and point-distributed grids in five- and nine-point two-dimensional (2D), finite-difference formulations. The radial flow assumption used in deriving the equations in this paper is not always strictly valid; however, for most practical situations it provides an adequate approximation for near-well flow.

This article, written by Technology Editor Dennis Denney, contains highlights of paper IPTC 10281, "Muse: A Multimodal Reservoir Simulation Environment," by J.A. Pita, SPE, N.M. Al-Zamel, and A.H. Dogru, SPE, Saudi Aramco, prepared for the 2005 International Petroleum Technology Conference, Doha, Qatar, 21-23 November. A new environment integrating a voice-advisory system with multimodal visualization was built for Saudi Aramco's in-house reservoir simulator. Engineers can simultaneously visualize oil saturation, hear changes in reservoir pressure, and feel rock permeability interactively as they move a 3D mouse throughout the reservoir. Engineers also can provide operational and surveillance rules so that the computer automatically checks conditions on any variables. The system can take a number of actions, such as using sound synthesis to speak directly to the engineer, when these conditions occur. The system runs on a Microsoft Windows-XP-based personal computer (PC) with dual processors. Introduction New-generation reservoir simulators, capable of handling millions of cells, imply that geological detail can be handled without intensive scaling up, which can lead to direct seismic integration with reservoir simulation. Unfortunately, the ability to analyze information has not fully kept pace with these advances. Recently, devices that function as a "3D mouse" with haptic (i.e., touch) capabilities have appeared. These devices have the ability to navigate and probe inside Earth models. Saudi Aramco's Multimodal Simulation Environment (Muse) is an environment that integrates a voice-advisory system with multimodal visualization. This environment provides stereo-capable 3D visualization, force feedback by use of haptic devices, and data sonification. Multimodal Data Analysis In contrast with conventional environments that provide only 3D visualization as the means for data interpretation and analysis, this system uses two additional "channels" for sensorial interaction: touch (haptics) and sound (sonification). Reservoir-simulation results that are loaded into the system can be attached to one or more of these three channels. The number of data volumes (i.e., simulation variables) that can be loaded is limited only by the amount of addressable system memory, but only three volumes can be attached to channels. The other volumes resident in the system can be toggled at any time into the channels. All loaded volumes, either attached to channels or not, can be monitored by the system by use of rules prescribed by the engineer. For example, the engineer can visualize 3D pressure distribution, feel permeability changes with the haptic device, and hear oil or gas saturation through the system speakers. The values of these properties are assigned to a range of 0 to 255, similar to a typical color map. Also, simulation results such as 3D distribution of the mole fraction of hydrogen sulfide (H2S) or acoustic impedance from a petroelastic model can be tracked as events, even if they are not directly attached to any channels. Visualization Channel So-called "rainbow" color maps are typical (e.g., blue for low values, red for high). The color map panel is fully editable, and engineers can choose the upper and lower limits to concentrate on specific value ranges, if so desired. Figs. 1 through 3 show typical 3D displays of different simulation-output variables (pressure, gas saturation, and H2S mole fraction, respectively). Fig. 1 also shows the channel selection menu. A separate opacity map enables parts of the volume to be made translucent to reveal only specific ranges of the data. The data shown correspond to a nine-component compositional simulation of 500,000 cells for a gas/condensate reservoir.

The objective of this project is to demonstrate that detailed reservoir characterization of slope and basin clastic reservoirs in sandstones of the Delaware Mountain Group in the Delaware Basin of West Texas and New mexico is a cost-effective way to recover a higher percentage of the original oil in place through strategic placement of infill wells and geologically based field development. Project objectives are divided into two major phases. The objectives of the reservoir characterization phase of the project are to provide a detailed understanding of the architecture and heterogeneity of two fields, the Ford Geraldine Unit and Ford West field, which produce from the Bell Canyon and Cherry Canyon Formations, respectively, of the Delaware Mountain Group and to compare Bell Canyon and Cherry Canyon reservoirs. Reservoir characterization will utilize 3-D seismic data, high-resolution sequence stratigraphy, subsurface field studies, outcrop characterization, and other techniques. Once the reservoir characterization study of both fields is completed, a pilot area of approximately 1 mi{sup 2} in one of the fields will be chosen for reservoir simulation. The objectives of the implementation phase of the project are to (1) apply the knowledge gained from reservoir characterization and simulation studies to increase recovery from the pilot area, (2) demonstrate that economically significant unrecovered oil remains in geologically resolvable untapped compartments, and (3) test the accuracy of reservoir characterization and flow simulation as predictive tools in resource preservation of mature fields. A geologically designed, enhanced recovery program (CO{sub 2} flood, waterflood, or polymer flood) and well-completion program will be developed, and one to three infill wells will be drilled and cored.

The objective of this project is to demonstrate that detailed reservoir characterization of slope and basin clastic reservoirs in sandstones of the Delaware Mountain Group in the Delaware Basin of West Texas and New Mexico is a cost effective way to recover a higher percentage of the original oil in place through strategic placement of infill wells and geologically based field development. Project objectives are divided into two major phases. The objectives of the reservoir characterization phase of the project are to provide a detailed understanding of the architecture and heterogeneity of two fields, the Ford Geraldine unit and Ford West field, which produce from the Bell Canyon and Cherry Canyon Formations, respectively, of the Delaware Mountain Group and to compare Bell Canyon and Cherry Canyon reservoirs. Reservoir characterization will utilize 3-D seismic data, high-resolution sequence stratigraphy, subsurface field studies, outcrop characterization, and other techniques. Once the reservoir- characterization study of both fields is completed, a pilot area of approximately 1 mi{sup 2} in one of the fields will be chosen for reservoir simulation. The objectives of the implementation phase of the project are to (1) apply the knowledge gained from reservoir characterization and simulation studies to increase recovery from the pilot area, (2) demonstrate that economically significant unrecovered oil remains in geologically resolvable untapped compartments, and (3) test the accuracy of reservoir characterization and flow simulation as predictive tools in resource preservation of mature fields. A geologically designed, enhanced-recovery program (CO{sub 2} flood, waterflood, or polymer flood) and well-completion program will be developed, and one to three infill wells will be drilled and cored. Progress to date is summarized for reservoir characterization.

The objective of this project is to demonstrate that detailed reservoir characterization of slope and basin clastic reservoirs in sandstones of the Delaware Mountain Group in the Delaware Basin of West Texas and New Mexico is a cost effective way to recover a higher percentage of the original oil in place through strategic placement of infill wells and geologically based field development. Project objectives are divided into two major phases. The objectives of the reservoir characterization phase of the project are to provide a detailed understanding of the architecture and heterogeneity of two fields, the Ford Geraldine unit and Ford West field, which produce from the Bell Canyon and Cherry Canyon Formations, respectively, of the Delaware Mountain Group and to compare Bell Canyon and Cherry Canyon reservoirs. Reservoir characterization will utilize 3-D seismic data, high-resolution sequence stratigraphy, subsurface field studies, outcrop characterization, and other techniques. Once the reservoir- characterization study of both fields is completed, a pilot area of approximately 1 mi{sup 2} in one of the fields will be chosen for reservoir simulation. The objectives of the implementation phase of the project are to (1) apply the knowledge gained from reservoir characterization and simulation studies to increase recovery from the pilot area, (2) demonstrate that economically significant unrecovered oil remains in geologically resolvable untapped compartments, and (3) test the accuracy of reservoir characterization and flow simulation as predictive tools in resource preservation of mature fields. A geologically designed, enhanced-recovery program (CO{sup 2} flood, waterflood, or polymer flood) and well-completion program will be developed, and one to three infill wells will be drilled and cored. Accomplishments for this past quarter are discussed.

Documented here in part are the reservoir engineering techniques used during the Bay Marchand fire. Before any plans for the kill operations were made, a brief reservoir simulation study was conducted to gain insight into the mechanics of fluid flow. Some of the predictions made with the model are compared with actual field results. Introduction A fire broke out on Shells Oil Co.'s Platform "B" at Bay Marchand Offshore Louisiana on Dec. 1, 1970. Of the 22 completed wells (43 zone completions) and two drilling wells, 11 caught fire; and Shell immediately set up an emergency task force to handle the problem of extinguishing the fire and capping the wells. This paper is one of a series that documents the activities of that task force during the 136 days required to cap all of the wells on the platform. Questions To Be Answered The reservoir engineering section of the task force was rapidly deluged with questions, all pertinent and important, for which other engineering sections and management needed answers before they could adequately plan and conduct the kill operations. These plan and conduct the kill operations. These inquiries can be summarized into four basic questions that had to be answered for each of the 11 wells to be capped. (1) What is the possibility that circulation will belost if a relief well were to penetrate a blowoutzone within a 25-ft radius of the producing well? (2) What surface injection pressure can be anticipatedduring a high-injection-rate kill operation? (3) How much water must be pumped into the reliefwell before water breaks through in theproducing well. (4) How much water must be pumped into the reliefwell to achieve 100 percent water production inthe producing well? Techniques Employed Question 1 was easily answered; however, we were not able to come up immediately with simple answers to the remaining questions nor rapid methods of dealing with the problems. Intuition and experience with conventional fluid injection projects strongly suggested that the predominant flow pattern from the injector would be radial, but there was some uncertainty as to the effect of the producing well on this pattern, both before and after water breakthrough. Accordingly, we decided to conduct a relatively brief reservoir simulation study, having a very specific and limited objective: to gain insight into the mechanism involved in controlling a blowout by water injection from an adjacent well. The Reservoir Simulation Study All simulations of a blowout kill operation were made in a two-dimensional areal mode utilizing COMSIMa three-phase, three-dimensional, compressible simulator developed by Shell Development Co. The investigation and its results were all subject to the following assumptions and restrictions: (1) The reservoir was assumed to be horizontal, homogeneous, and isotropic. (2) Interference effects from other wells orreservoir boundaries were ignored. (3) Gravity and capillary effects were neglected. (4) Only the injection of water was simulated. JPT P. 234

Compositional reservoir simulation is a time bound activity demanding complex physics. We review the advantages of machine learning in complex compositional reservoir simulations to determine fluid properties, such as critical temperature and saturation pressure. A machine learning approach to predict critical temperatures during simulation based on the Heidemann-Khalil method is implemented, resulting in more accurate results with lower computational cost, outperforming the standard method and improving performance on a giant field model with compositional gradient and miscible gas injection. The fluid column grades from black oil to gas condensate; accurate phase behavior and miscibility modelling involves a significant number of components. This makes simulation performance one of the biggest challenges. Critical temperature of the mixture is commonly used to determine phase state (phase labeling), a crucial process of reservoir simulation. Mislabeling can result in incorrect physics and convergence issues, particularly in cells with gas-oil displacement. Simulators generally use the Li correlation to calculate the pseudo-critical temperature from the weighted average of component critical temperatures, leading to inaccuracies. The Heidemann-Khalil method is computationally costly, proportional to the cube of the number of components, prohibiting its use for complex compositional simulation models. We use machine learning to efficiently incorporate it into simulation. By using the machine learning approach, neural networks are trained based on a combination of feeds to reproduce the Heidemann-Khalil method with great precision. The accuracy of critical temperature and other fluid property determinations is thus improved, with machine learning ensuring a very low computational cost. Convergence is also improved. With the traditional Li method, especially at the beginning of gas injection, we faced numerical difficulties, and the runtime was slowed down. By implementing the machine learning based method, the convergence is smooth through the entire gas injection cycle, leading to a reduction of total iteration counts. We experience an overall four times speed-up of the simulation model, which greatly enhances the usage of this model in simulation studies. The use of machine learning methods to replace physics in the simulator is an evolving area. By showing a field example of a successful application that improves both accuracy and performance, we contribute to fostering research into new possibilities where physics-informed models will enhance simulation studies.

The Ubit Field, located southeast of the Niger Delta, is a Nigerian National Petroleum Corporation (NNPC)/Mobil Producing Nigeria (MPN) Joint Venture (JV) asset that has been on production for more than 40 years (September 1970). Ubit has approximately 2.4GBO OOIP, with production peaking at ~137,000 BOPD in 1997, and is currently at ~85,000 BOPD. With much of the reserves yet to be produced, it is important that an optimized field development plan be implemented to maximize recovery. Guided by reservoir simulation models, a revised Ubit Field re-development plan was approved in December 2009. Infill development has proceeded as designed, and the first infill drilling program is underway. Integration of results from reprocessed seismic, geologic interpretation, core, log and reservoir simulation studies, coupled with state of the art technology in well completions, were critical in developing an optimized depletion plan. Reservoir simulation results consistently indicate higher reserves capture with increased well density from infill drilling that targets unswept locations around and down-dip of abandoned completions. Results from the integrated studies suggest varying the optimum well spacing requirements for the different sections of the reservoir. The recovery factor in the lower quality reservoir section (Disturbed Biafra) could be vastly improved by increasing well density. The multidisciplinary study for field development indicates that approximately 100 additional wells should be drilled to tap unswept resource in the field. This aggressive approach will maximize ultimate recovery, but requires closer well spacing than minimum regulatory specification. The technical basis for this strategy, primarily reservoir simulation models anchored in measured hard-data, is the core focus of this paper.