Chapter1-M.pdf

Chapter 1

1.1

Intr Introd oduc ucti tion on

Overview Numerical simulators have come to play an increasingly important role in the oil industry. Knowledge of their use, strengths and limitations as tools for the evaluation and prediction of reservoir performance is valuable for understanding and m 1anaging petroleum reservoirs. Prior to simulation, hand calculation methods were the basis for reservoir management. Essentially, this required that elements of the overall problem were de-coupled. Zero dimensional material balances, one and twodimensional flow analyses in single and two-phase flow, well performance and lift were studied as separate independent problems. Success in applying the highly idealized models depended on the engineer's skill in recognizing the degree and the effects of interdependence and integrated concept of the whole operation. The advent of computer simulation, enabling full coupling of all elements of the system, with increasingly detailed characterization of the reservoir, has changed the situation dramatically. Now many difficult development scenarios can be applied to several different geological interpretations so that key uncertainties can be identified and data acquisition programs defined. The situation has changed from one where much of the available data was used only intuitively (through a highly personal 'expert system') to one where the modern system can accommodate explicitly more data than is generally available. This has the additional advantage that feed back is now possible, the outcome or implications of particular interpretations can be fed back to individual specialists for revision or confirmation. The continued developments in mathematical methods of manipulating equations (Finite

1.

Applied Reservoir Simulation by Dr. Tayyar Sezgin DALTABAN

1-1

Introduction

Difference, Finite Element, Boundary Integral, etc.) has been paralleled by developments in characterization with local grid refinement, corner point geometry, non-neighbor connections, etc. Effective simulation still depends on the individual skills of the contributing specialists. Without these skills model construction and history matching can become arbitrary number generating exercises. The real objective, of understanding how a reservoir will behave under an improved production scenario, can be lost. The degree of realism in the simulation predictions depends heavily upon the quality of the reservoir description used. Reservoir description is an art of combining data from different sources and scales (See Figure 1-1) and to find a realization which is acceptable at a given time for the subsequent performance studies. Reservoir characterization is a very complex task, details of which will be presented in the related section of this manual. Some information will become available only when a reservoir is in production, and inferences about communication become possible, emphasizing the importance of the early “what if" scenarios. It is therefore a continuing exercise, identifying infill drilling locations, potential pilot scheme areas, etc. Reservoir simulation plays the central role in a reservoir management process. It helps to assess the feasibility of any stage of the reservoir operations, including prospect analysis, exploration and appraisal, development planning, production operations and abandonment. Figures 1-2 and 1-3 demonstrate the role of simulation at various stages of reservoir management activities. From these Figures, it is clear that even at discovery stage, simulation plays a crucial role in assessing the feasibility of the project and deciding to go ahead with the appraisal stage. The reservoir description at this stage usually relies on data obtained by one well, 2-D seismic calibrated by this well, and analog information. Due to huge uncertainties, “what if” scenarios play a major role in the decision mechanism. Many alternative scenarios provide the management a broad view about the project robustness, the points of strength and weaknesses. At the exploration stage, after every well is drilled, a review of the project feasibility with the help of simulation has become a routine practice. The role of simulation in development planning is of foremost importance especially in offshore operations because of very high front end loading prior to a single droplet of oil recovery (for a typical 1 billion bbls of STOIIP, the cost of the installation is about the same in dollars). At later stages of field development and exploitation, simulation allows continuous monitoring and assessment of the feasibility of the project.

1-2

Applied Reservoir Simulation Simul ation by Dr. Tayyar Sezgin Sezg in DAL DALTABAN TABAN

Introduction

Figure 1-1: S e is m i c

O u tc r o p

P o r e S c a le

C o r e o

o

o o

o

o o o o o o o

o o

o o

o

o

o o

o o oo

o

o

o o o

o o o

o

W e ll te s t

W e ll L o g s

D a t a U p s c a li n g /D o w n s c a li n g I n t e g r a t io n

F i n i t e E le m e n t M e s h

P (r

e

, t) P (r

w

, t)

i

i+ 1

F in i t e D i f f e r e n c e

M esh

f o r M a c r o - F lo w


1-3

Introduction

Figure 1-2:

The Role of Simulation at Different Stages of Field Development Discovery Phase

2D Seismic

Standard Well Log

Core Pressure PVT Measurements Measurements Analysis

Well Tests

Upscaling

Reservoir

O.K?

Simulation No! Production Planning

Yes!

Iterate? Yes!

Appraisal No? Abandon Field Appraisal Stage

3D Seismic

Production

Logging

Interference Testing

Outcrop Surveys

Tracer Testing

Upscaling Update O.K?

Reservoir Simulation

No! Revised Production Strategy

Iterate?

Yes!

Yes! Appraisal No? Abandon Field

1-4


Introduction

Figure 1-3: Additional Data Development Phase

4D Seismic?

Well Logging(Sor)

Tracer Testing(Sor)

Predrilled Production Prediction?

Predrilled Production Prediction

Upscaling

Update O.K?

Reservoir Simulation

No! Revised Production Strategy

Iterate? Yes! No? What is the next step?


Yes!

Continue with Installing Platform & Implement Production Strategy

1-5

Introduction

1.2

Aim of the Book The book consists of a series of lectures, supported by carefully designed tutorials. These will give participants the opportunity to vary key parameters in the models and establish the impact on field performance. This course intends to: 

•

•

•

•

•

•

•

1-6

develop, a reasonable understanding of the mechanics of reservoir simulation, which is one of the preconditions for the effective use of the simulators by the engineers and geoscientists. Lack of this knowledge and understanding means inputting information and getting results without any appreciation of the relationship between them, explain the limitations and the structural aspects of the models. If these are not clearly understood, the users of the models will not be able to prepare appropriate input data. discuss with and develop amongst participants a sufficient background on engineering and geologic data acquisition techniques, data structures, and data processing techniques, review data scales and their interrelationships, soft data generation, and in that respect, explain state of the art reservoir characterization and reservoir model generation techniques. Normally, reservoir models are fine grid realizations of the reservoirs, and may comprise tens of millions of grid blocks with currently available resources, develop a sufficient background on generating reservoir models for simulation. Due to restrictions in computational resources, the simulation grid blocks are usually orders of magnitude greater than those used in constructing fine grid reservoir models. This, therefore, requires upscaling from a fine grid to a reservoir simulation grid. In depth discussion on the techniques used in single and multiphase case will be carried out both under static and dynamic conditions. The accuracy of the simulation model, therefore, depends strongly on the reservoir characterization. As a consequence, incorrectly compiled data in building a reservoir model, and improper use of upscaling procedures will yield unacceptable results. develop skills in conducting a simulation study, and framework for checking the results, their quality and integrity. discuss the current modeling practices, the models available with their unique features, and the degree of consistency among them. form the necessary background on history matching using simulators.


Introduction

1.3

Definitions and Descriptions The dictionary meaning of the word ' simulate' is ‘to give an appearance of ',' to make so as to resemble the real or genuine thing' or in other words ' to imitate or create the conditions of' ’. The word Simulation refers to ' utilization of a model to obtain some insight into the behavior of a real system'. To simulate any physical process is, therefore, a means to investigate behavior through a system necessary which is called a model. Models can be of two types, namely: 1.

Physical Model: It is essentially a scaled down reproduction of the original.

2.

Mathematical Model: It is a system of equations describing the physical behavior of the process of concern.

The core of the reservoir simulator is the mathematical model and in this context a reservoir simulator can be defined as: 'The process of inferring the behavior of a hydrocarbon reservoir from the performance of a mathematical model of that reservoir' The mathematical model of a reservoir simulator is a set of partial differential equations with appropriate boundary conditions sufficient to represent the physical processes that may occur in a reservoir. These partial differential equations are always non-linear meaning that the primary unknowns of the system such as saturation, pressure and concentration exhibit non-linear A typical flow equation is given below:

∂ ∂Φ ∂ ∂Φ ∂ ∂Φ ∂ Φ S -----  λ -------  + -----  λ ------- + -----  λ ------- = ----  ------- ∂ x  ∂ x  ∂ y  ∂ x  ∂ z  ∂ z  ∂ t  β 

(1)

where β is formation volume factor, t is time, S is saturation, x, y and z are cartesian co-ordinate directions, φ is porosity, Φ flow potential. If there are three phases in a given medium (oil + water + gas), then, there are three similar partial differential equations. In the above equation, permeabilities, porosities, flow potentials, formation volume factors, viscosities are to be calculated simultaneously. Except in a few simple cases like Buckley-Leverett type displacement problems, an analytical solution to the partial differential equations cannot be found due to these non-linearities. This has tempted mathematicians to


1-7

Introduction

explore approximate solution techniques to the full mathematical model with complete reservoir description. These solutions are numerical in their nature. Most frequently, there are two classes of numerical solutions used; namely, finite element and finite difference methods. It follows from this that a mathematical model should be translated into an approximate numerical model. The reservoir simulator is then an implementation of numerical algorithms on a computer in the form of software to find an approximate solution to the mathematical model. The approximate solutions to the flow equations are based on gridded realizations of the petroleum reservoirs. This is a final outcome of the efforts summarized by Figure 1-1. For each node in the grid system, a value for the following parameters is required: •

Permeability

•

Porosity

•

Thickness

•

Elevation

•

Grid dimensions

•

Initial saturation for each phase

•

Initial pressure

•

Rock compressibility

Fluid characteristics are assigned by the following relationships: •

Oil formation volume factor versus pressure

•

Water formation volume factor versus pressure

•

Gas formation volume factor versus pressure

•

Oil viscosity versus pressure

•

Water viscosity versus pressure

•

Gas viscosity versus pressure

•

Solution gas-oil ratio versus pressure

•

Solution gas-water ratio versus pressure

•

Liquid to gas ratio versus pressure

•

Oil density

•

Gas density

•

Water density

The interactions of forces between rock and fluids are given by the following saturation dependent functions:

1-8


Introduction

•

Relative permeability for each phase

•

Capillary pressure between oil and water

•

Capillary pressure between gas and oil

Additional data may come from wells and include:

1.4

•

Producing interval.

•

Oil production rate versus time

•

Water production rate versus time

•

Gas production rate versus time

•

Observed pressure versus time

Major Steps in a Simulation Study These are: 1.

Reservoir Description: Starting with a meaningful subdivision of a given reservoir into hydraulic units; further discretization of hydraulic units into grid blocks, and arriving at estimates of the parameters in the governing flow equations describes the process as a function of spatial position. The parameter estimates have the meaning of average or pseudo values at the scale of grid block in the discretized version of a continuous reservoir. The deliverable will be a reservoir model. The resolution of the reservoir model may go up to hundreds of millions of mesh points.

2.

Recovery Mechanism Identification: The decision has to be made concerning the method of recovery like water injection, gravity drainage, natural depletion, gas injection, etc.

3.

Mathematical Model: Selection of mathematical model is required here which may be black oil formulation, compositional formulation, single phase model, multi-phase model, single dimension, multi-dimension, etc.

4.

Engineering Model: This is also called simulation model. The objective is to generate an acceptable representation of the reservoir based on the reservoir model generated. Due to computational resource limitations, the grid to be used for the engineering model can be much coarser than the reservoir model (on the order of tens of thousands or hundreds of thousands). To this end, some upscaling efforts are carried out.

5.

Numerical Model: There are several different numerical models available including: Finite Element Model, Finite Difference Model, Boundary Integral Model, and Streamlines Model. Among these, the Finite Difference


1-9

Introduction

approach is the most commonly used approach, almost to the exclusion of others. 6.

Computer Model: (ECLIPSE, etc.)

7.

Validate Model: Prior to a History Match, the model needs to be validated. This may be done by comparing STOIIP estimate using, for example, the Monte-Carlo Approach or Deterministic Approach. Fine-tuning may be necessary around the control points (say wells) by matching well test pressures, production rates and the expected disturbances in the vicinity of the wellbore during the tests. The gridded reservoir realization must also conform with the reservoir description.

8.

History Match: The model is compared with the available history and finetuning of the model is made when necessary. The quality of the history match depends heavily upon the quality and the degree of realism in the reservoir description efforts.

9.

Performance Predictions: Once a satisfactory history match is obtained, performance predictions can be carried out.

10. Update: The whole process may be updated starting from the reservoir de-

scription. Probably the best known application of numerical reservoir modeling is that of matching historical performance and then predicting future reservoir behavior. In matching, one uses the best data available to estimate all of the parameters listed in the previous paragraph. Then the wells are allowed to produce at the observed rate for one of the phases. Next, pressure behavior for all wells and the production rate of the remaining phases are calculated. Calculated pressures and rates are then compared with observed pressures and rates. The comparison between these two sets of values will indicate how accurate an initial estimate was made for the input data. Next, it may be necessary to modify some of the input data until all observed and calculated data compare favorably. No hard and fast rules exist to indicate when a match is obtained. The number of runs made prior to obtaining a satisfactory match depends on the complexity of the reservoir and the length of history. When a match is achieved in this manner, a rather sophisticated reservoir representation has been obtained, and it can be used to predict the future behavior of the field. During prediction, one may set the desired production rates for all wells or for the entire field. Another option is to set a production rate to be maintained until the reservoir pressure falls to a certain point. The various criteria imposed on prediction runs are usually meant to reflect alternative development/ depletion scenarios. Each prediction run will represent a specific physical development

1 - 10

Applied Reservoir Simulation by Dr. Tayyar Sezgi n DALTABAN

Introduction

plan or reservoir management scheme to be evaluated. Then, with these hypothetical rate schedules, the economic performance of the field is studied. In this manner, various exploitation schemes may be evaluated, economics may be applied to the results, and the "optimum" exploitation scheme may be selected. A word of caution appears warranted. The reservoir representation was calibrated with a certain kind of history; for example, depletion at moderate rates. One should not expect any degree of accuracy if suddenly the mechanism is changed to that of water injection without any history to predict the behavior of that water injection scheme. Similarly, if suddenly the field is produced at very high rates causing pressure to drop below the bubble point, the predicted results should be viewed with caution. A general rule of thumb may be that one should not predict more than twice the period used for matching under similar modes of operation. Particular care is necessary when using a model to predict behavior when no history is available for calibrating the model. Although reservoir modeling was originally used to study overall field performance and to predict that performance following matching, it has many other applications. Considerable use can be made of these models to study sections of fields, which are considered to be typical. Then, by assigning the best known values to that section, one may study the field's producing mechanisms. One may then change some of the parameters to see what effect they have on the overall mechanism. In turn, one may find what parameter(s) need additional study in the laboratory or the field in order to better understand the performance observed in the field. Numerical reservoir modeling is, however, a tool and is best used in conjunction with the other types of tools. Ideally, the results of the different types of analysis will tend to reinforce the conclusions of each, leading to a higher degree of confidence in the answer. Over the past two decades, reservoir simulators have gained wide acceptance and have become a standard tool for reservoir engineers. The tremendous increase in computing speed and capacity (by a factor of approximately 1000) has helped to drive this acceptance. This reduction in unit cost; development of new and efficient algorithms, and the ability of the simulators to handle most general features of hydrocarbon reservoirs makes simulation a useful engineering tool.


1 - 11

Introduction

1.5

Classification of Simulators 1.5.1

Classification Based on Linearization

Due to strong non-linearity of Equation (1), it is necessary to linearize it to obtain unknown parameters. Hence, the first classification of the simulators is based on the linearization technique used. There are two generic techniques, namely: IMPES: Implicit Pressure and Explicit Saturation. In this application, flow equations of water, oil and gas are coupled and solved first for pressure and then saturation. This technology assumes that the change in capillary pressure over a time step should be negligible. In addition, the pressures are calculated by the old time level saturation and pressure dependent parameters. They are updated after pressures and saturations are updated. Fully Implicit: In this application, all of the unknowns are solved simultaneously. The flow equations are linearized by using Newtonian approach. 1.5.2

Classification Based on Fluid Characteristics

Simulators are classified based on fluid characterization as Black Oil Simulators or Compositional Simulators.

1 - 12

1.

Black Oil Simulators: This type of simulator treats hydrocarbons as two components; gas and oil. They are applicable to dissolved gas, medium gravity oil-bearing reservoirs under moderate reservoir pressures and temperatures. They can be applied to almost all conventional water flooding simulation studies. If the oil formation volume factor is less than two, they can safely be applied to solution-gas drive, gas cap expansion or gas injection studies. Black oil simulators can also be used for some cases where the formation volume factor is greater than 2. That is possible if oil and gas formation volume factors, gas in solution, and oil and gas viscosities, are plotted as a function of pressure and can be determined accurately by calculation or experiment.

2.

Compositional Simulators: Compositional simulators are those which use cubic Equations of State forms like Peng Robinson, Soave-Redlich-Kwong, Redlich-Kwong, Schmidt Wenzel and Patel-Teja. Instead of tracking the phases, as in Black Oil Simulators, track constituent components of hydrocarbons like Methane, Ethane, Butane, Propane, Nitrogen, Carbon Dioxide, etc. Because


Introduction

of this multicomponent treatment of reservoir fluids, the simulation is capable of handling: •

•

•

•

Enhance Oil Recovery by carbon-Dioxide or enriched gas injection, Multiple Contact Miscibility studies Natural depletion and Injection of Gases such as Nitrogen or residue gas into gas condensate reservoirs. Natural Depletion or Gas Injection into Volatile Oil Reservoirs. Re-evaporation of residual oil by injecting residual gas

Despite the excellent capability of simulating the compositional phenomena in gas condensate reservoirs, compositional simulators can be as much as 100 times more expensive than black oil simulators. The main source of this additional computation cost is the Equation of State (EOS) calculations (which may be up to 80% of the total cost). 1.5.3

Classification Based on Temperature Dependence •

•

1.5.4

Isothermal: Simulators that consider the temperature of the reservoir constant Thermal: In this case the following equations are involved: 1.

Energy equation.

2.

Oxygen for in-situ conditions

3.

Gas for in-situ conditions

4.

Hydrocarbon components: light, medium and heavy components may be necessary

5.

Phase: gas, liquid and solid phases (4 phase: oil + water + gas + solid)

Classification Based on Grid Dimensions and Types (See Figures 1-4 to 1-8): 1-Dimensional: These models cannot be used for fieldwide simulation applications because they cannot handle either areal sweep or the gravity effects. They can be used for sensitivity towards some selected reservoir parameters prior to full-scale simulation. The effect of viscous forces and mobility ratios on the recovery can be tested. They are especially used for testing the accuracy of the simulators against known analytic solutions like that of BuckleyLeverett. Also, most of the finite difference techniques can be tested


1 - 13

Introduction

first with one-dimensional case and then their use can be extended to multi-dimensions. They are also useful in assessing the heterogeneity in the direction of flow. 1-Dimensional Cartesian - Horizontal. 1-Dimensional Cartesian -Vertical: These are usually used for vertical equilibrium. 1-Dimensional Radial: In addition to general objectives as stated, they are also used for assessing the productivity impairment in gas condensate reservoirs as well as volatile oil reservoirs. They can also simulate well testing (i.e. radial flow).

Figure 1-4: Grid Dimensions X

(a) 1-D Linear

r (b) 1-D Radial

Z

(c) 1-D VE

1 - 14


Introduction

2-Dimensional: These include areal cartesian, areal radial and cross-sectional cartesian, and cross-sectional radial models. 2-Dimensional Cartesian Areal: Although petroleum reservoirs are three dimensional, in some cases, especially for thin reservoirs where gravitational effects are negligible, Z-direction may not be important. Under the circumstances, these type of models are used when areal flow patterns dominate the reservoir performance; and if the areal heterogeneities are important. Most areal models use pseudo functions to account for flow in vertical direction. However, they are not necessary if the reservoir is thin and stratification is not important. Under the circumstances, normal reservoir engineering studies can be carried out with these models including well position optimization; distribution of injection and withdrawal rates; timing for installation of artificial lift and modification of surface facilities. 2-Dimensional Radial Areal: This model investigates the effect of well performance as in the case of 1-Dimensional radial models, with the addition of areal heterogeneities. 2-Dimensional Cross-sectional Cartesian: In this case, instead of neglecting flow in the vertical direction, one of the horizontal directions can be discounted. This type of model can be used in cases where vertical flow is dominant. A highly stratified reservoir is a typical case study for this type of model. The effect of segregation and the effect of stratification are the main focus areas in this type of modeling. They are also used in developing well functions, pseudo functions and coning functions; simulating peripheral gas injection. Crestal gas injection, or other processes in which frontal velocities toward producers are largely uniform help to justify simplification in modeling of entire fields or field segments. Studying miscible processes to assess the gravity and heterogeneities on displacement efficiency and the sweep efficiency are also candidates. 2-Dimensional Cross-Sectional Radial: This model is also applied to reservoirs where gravity forces and stratification dominate the flow, and areal property distribution of the reservoir is relatively homogeneous. This approach is used particularly to investigate the coning problem. They can also be used to develop coning functions and well functions.


1 - 15

Introduction

Figure 1-5: Grid Dimensions AREAL-CARTESIAN

CROSS-SECTIONAL CARTESIAN

AREAL-RADIAL

2-DIMENSIONAL DOMAN REALISATIONS CROSS SECTIONAL RADIAL

1 - 16


Introduction

Figure 1-6: Grid Dimensions

lock Centered Geometry

5800

3000

4000

5000

6000

6200

6600

7000

7400

7800

Block Centered Geometry XZ plane

orner Point Geometry 5800

3000

4000

5000

6000

6200

6600

7000

7400

7800

Corner point Geometry XZ plane


1 - 17

Introduction

3-Dimensional Models: They can be Cartesian, Radial-Cylindrical or Spherical. 3-Dimensional Cartesian: Reservoir geometry can sometimes be too complex to be represented by two-dimension. For example, reservoirs having shales or other flow barriers that are continuous over large areas, but with permeable windows where crossflow occurs, are difficult, if not impossible, with two dimensions. Reservoir mechanics may be so complex that two-dimensional realizations are difficult to analyze. Reservoirs which are at a more advanced stage of depletion fall into this category and require careful and precise modeling to distinguish between performances resulting from alternative depletion plans. The displacement to be studied may be dominated by vertical flow as, for example, near wells where both cusping and coning may occur. Both areal and vertical details needed can be obtained only in a 3D segment model. Occasionally 2-D studies are more troublesome and expensive than 3-D modeling. Reservoirs with a complex facies structure may require an excessive number of pseudoisations to be represented with two dimensions. 3-Dimensional Cylindrical: This model has the same objectives as 1-D radial, 2-D Radial areal and 2-D radial cross-sectional models. In addition, it is able to capture the impact of vertical and areal heterogeneities on the flow. 3-Dimensional Spherical: This approach is used to investigate the partial penetration effects on well testing results and to model minipermeametry flow.

1 - 18


Introduction


3-DIMENSIONAL CARTESIAN 3-DIMENSIONAL CYLINDRICAL



1 - 19

Introduction

1.6

Benefits of Reservoir Simulators The reason for using reservoir simulators is to estimate reservoir performance under a variety of production schemes. We have only a single opportunity to produce from an actual reservoir with a considerable expense whereas with simulation we can test several alternatives and assess them prior to deciding the actual field operation. The cost of simulation is considerably cheaper and the time necessary is usually negligible to the actual operation. Some of the applications and the benefits of the simulation can be summed up as follows: 1.

The performance of a hydrocarbon reservoir under natural depletion, water injection or cycling can be examined.

2.

Type of water flooding can be judged. For example, it is possible to see the relative merits of flank water injection and pattern waterflooding.

3.

The effect of well location and spacing can be critically evaluated.

4.

The effect of the production rate on the hydrocarbon recovery can be estimated.

5.

For a given number of wells at certain specified locations, it is possible to predict total field gas deliverability.

6.

In heterogeneous hydrocarbon reservoirs, it is possible to estimate the leaseline drainage.

7.

To maximize hydrocarbon recovery, best methods of field development and production schemes can be found.

8.

Best Enhanced Oil Recovery (EOR) scheme and its implementation can be determined.

9.

The reasons why the reservoir behavior deviates from the earlier predictions can be explained.

10. The ultimate economic hydrocarbon recovery can be predicted. 11. Laboratory and field data requirements and their subsequent effect on the

performance predictions can be assessed. 12. The best completion schemes for the wells can be established.

1 - 20


Introduction

13. The section of reservoir from which the hydrocarbon is produced can be

identified. 14. Critical parameters to be measured from the field in an application of a re-

covery scheme can be identified. 15. It is possible to decide whether it is necessary to do physical model studies

of the reservoir and if so how can the findings can be scaled up for field applications.

1.7

ECLIPSE – State of the Art Reservoir Simulator Immense advances in the areas of data acquisition and integration, especially in the last 10 years, dictate ever increasing complexities in reservoir description in order to model reservoir behavior. The detail required by reservoir management and the detail introduced by the integrated multidisciplinary reservoir characterization efforts, require robust and fast simulation technologies which are parallelisable and cost effective. These technologies must be able to cope with all of the complexities of the field production and operations and must be flexible to absorb new concepts. They must be user friendly as the ever-growing demand for the simulators encourages non-specialists to make use of them in their day to day field management efforts. In this respect, the front-end processors of such applications must be able to detect the physical inconsistencies in the input data to a measurable degree. The increasing demand for the simulators also dictates that simulation software be available for PC's as well as workstations. The pre and post processing facilities must be extremely powerful so that engineers and geoscientists can monitor their simulation studies efficiently. Integration of various different data processing and simulation modules and data transferability between them is another important aspect required within the current reservoir management environment. In fact, the advances in data exchange and integration between different disciplines in the recent years have begun forming the firm and necessary basis for multi-disciplinary co-operation and working. ECLIPSE has been a leading state of the art software in the oil industry dominating currently 70% of the market. Some of the main reasons for its dominant role in the oil industry are its stability, robustness, and high degree of material conservation, mathematical accuracy and flexibility. The available functions of the ECLIPSE are summarized in the form of need and solution:


1 - 21

Introduction

1 - 22

NEED

SOLUTION

Handling multi-phase flow problems

Eclipse 100, Eclipse 200 and Eclipse 300

Handling Complex Geometries

Block-center geometry, Corner Point Geometry, Unstructured Gridding, Pebby/Voronoi Gridding, Non-Neighbor Connections and Local Grid Refinement

Unconditional stability and robustness

Fully Implicit

Compromise solution between accuracy and computational speed

IMPES and Adaptive Implicit

Modeling Structural Geological Features like faults

Non-Neighbor Connections

Flexibility in Model Size and Representation

Run Time Dimensioning

Modeling Segregated flow in vertical direction

Vertical Equilibrium

Fractured Reservoi r Modeling

Dual Porosity and Dual Porosi ty/permeability options

Relative Permeability and Capillary Pressure Treatment

Directional Relative Permeabilities, Relative Pe rmeability and Capillary Pressure hysteresis, Saturation Table Scaling

Gas Field Operations, Gas Lift Optimization

ECLIPSE 200

PVT Data

ECLIPSE 100 and 200 for Black Oil data ECLIPSE 300 for Compositional data, API Tracking

Modeling Flow of Methane in Coalbed

ECLIPSE 200

Collapse of pore channels due to change in the pore pressure

Rock Compaction Option

Designing and modeling single well and interwell tracer testing

Tracer Tracking

Cold water injection into a re servoir cooling effects need to be handled, energy conservation must be maintained

Temperature Model

Modeling First Contact and Multiple Contract Miscibilities

ECLIPSE 300 compositional formulation ECLIPSE 100/ 200 three component model to handle First Contact Miscibility

Managing Field and Well schedules

Individual Well Controls Group and Field Production Controls Multi-Level Grouping Hierarchy Group Injection Controls Sales Gas Production Control Crossflow and Comingling in Wells Highly Deviated/Slanted and Horizontal Wells Special Facilities for Gas Wells Surface Networks

Modeling Polymer, Surfactant and Foam

ECLIPSE 200

Accurate Distribution of Initial Fluid in Place

Fine Grid Equilibration

Aquifer Modeling

Non-neighbor Connections Analytic Aquifers (Fetkovitch and Carter Tracy) Numerical Aquifers

Interpolation of sparse data

FILL Program

Flow in the Wellbore

VFP Module comprising the following options:-Aziz, Govier and Fogarasi Orkiszewski, Hagedorn and Brown, Beggs and Brill Mukherjee and Brill, Gray

Pre and Post Processing Facility

GRID, GRAF, and RT View


Introduction

NEED

SOLUTION

Well Test Interpretation and Analysis

Well Test 200

Accurate and Efficient Preparation, Evaluation and Quality Checking of the Well Production and Completion Data

SCHEDULE

Relative Permeability and Capillary Pressure Pseudoisations

PSEUDO

Coupling Multiple reservoirs and Parallelization Compatibility

ECLIPSE 200

Between Different Modules

YES

Automatic History Matching

Sim Opt


1 - 23

Introduction

References Ames, W.F. (1969): Numerical Methods for Partial Differential Equations, Thomas Nelson and Sons, Don Hills. Aziz, K. and Settari, A. (1979): A Petroleum Reservoir Simulation, Applied Science Publishers, LTD., Wilmette, IL. Bech, N. (1984): “Classification of Reservoir Simulators,” Energiminsteriets Energiforskningsprogram, Olie-og gasreservoirmodeller , Report No. 6, RisoM-2421, Riso Natl. Laboratory, Rosklide, Denmark. Burewell, R.B. and R.E. Hadlow (1976): “Reservoir Management of the Blackjack Creek Field,” paper SPE 6195 presented at the 1976 SPE Annual Technical Conference and Exhibition, New Orleans, Oct. 3-6. Coats, K.H. (1983): “Author’s Reply to Discussion of Reservoir Simulation: State of the Art,” JPT (Nov.), p. 1176 Coats, K.H. (1982): “Reservoir Simulation: State of the Art,” JPT (Aug.), pp. 1633-1642. Coats, K.H. (1969): “Use and Misuse of Reservoir Simulation Models,” JPT , (Nov.), pp. 1391-1398. Crichlow, H.B. (1977): Modern Reservoir Engineering Aspects - A Simulation Approach , Prentice Hall, Englewood Cliffs, NJ. Dake, L.P. (1978): Fundamentals of Reservoir Simulation, Elsevier. Daltaban, T.S. (1989): “Petroleum Engineering Studies Educational Model,” SPE 19145, published in the Proceedings of SPE Petroleum Computer Conference, San Antonio, TX., June 26-28. Daltaban, T.S. and C.G. Wall: “Petroleum Reservoir Management - Past, Present and Future,” Journal of Mining Tech., (Nov.), v. 78, No. 903, pp. 297305. Ferguson D.S. and H.D. Attra (1961): “The Uses and Limitations of Computers in Petroleum Engineering Work,” JPT (July), pp. 625-628.

1 - 24


Introduction

Kazemi, H. (Oct. 1996): “Future of Reservoir Simulation”, Society of Petroleum Engineers Computer Applications, pp. 120-121. Khatib, A.K. (1983): “Discussion of Reservoir Simulation: State of the Art, JPT , (June), p. 1176 Mattax, C.C. and R.L. Dalton (1990): Reservoir Simulation, SPE Monograph #13, Richardson, TX: Society of Petroleum Engineers. Odeh, A.S. (1969): “Reservoir Simulation... What is it?,” JPT , (Nov.), pp. 1383-1388. Peaceman, D.W. (1977): Fundamentals of Numerical Reservoir Simulation, Amsterdam: Elsevier. Richardson, J.G. and R.J. Blackwell (1971): “Use of Simple Mathematical Models for Predicting Reservoir Behavior,” JPT, (Sept.), pp. 1145-1154; Trans. AIME , 251. Schlumberger GeoQuest Reservoir Technologies, (1998): RMT Manual, Houston, TX. Schlumberger Technology Corporation, (1998): ECLIPSE 100 Reference Manual, Houston, TX. Schlumberger Technology Corporation, (1998): ECLIPSE 100 Technical Manual, Houston, TX. Schlumberger Technology Corporation, (1998): ECLIPSE 200 Reference Manual, Houston, TX. Schlumberger Technology Corporation, (1998): ECLIPSE 300 Reference Manual, Houston, TX. Schlumberger Technology Corporation, (1998): GRID Reference Manual, Houston, TX. Serra, J.W. and R.C. Wilson (1976): “Reservoir Simulation Studies Reveal Key to Maximizing Waterflood Oil Recovery, Virginia Hills Beaverhill Lake ‘A’ Pool,” CIM 7617, presented at the 1976 Annual Technical Meeting of the Petroleum Soc. of CIM, Calgary, June 7-11.


1 - 25

Introduction

Sheldon, J.W., C.D. Harris and D. Bavly (1960): “A Method for General Reservoir Behavior Simulation on Digital Computers,” SPE 1521-G, presented at the 1960 SPE Annual Meeting, Denver, CO., Oct. 2-5. Smith, G.D. (1964): Numerical Solution of Partial Differential Equations, Oxford University Press. Staggs, H.M. and E.F. Herbeck (1971): “Reservoir Simulation Models - An Engineering Overview,” JPT , (Dec.), pp. 1428-1436. Thachuk A.R., and R.A. Wattenbarger, (1970): “The What, Why, When, and How of Reservoir Simulation,” Canad. Pet. (April), pp. 86-92. Thakur, G.C. (1996): “What is Reservoir Management?” JPT , pp. 520-525 Thomas, G.W. (1982): Principles of Hydrocarbon Reservoir Simulation, Boston: International Human Resources Development Corporation. Toronyi, R.M. and N.G. Saleri (1988): “Engineering Control in Reservoir Simulation,” SPE 17937, Proceedings of 1988 Society of Petroleum Engineers Fall Conference, Oct. 2-5. Van Poollen, H.K. (1971): “The Wise Use of Reservoir Models,” APEA J. v. 11, pp. 131-134. Wiggins, M.L. and R.A. Startzman (1990): “An Approach to Reservoir Management,” Paper SPE 20747, Proceedings of 65th Annual Society of Petroleum Engineers Journal, pp. 323-338; and “Part II - Implementation,” Society of Petroleum Engineers Journal, pp. 339-344. Stags, H.M. and HErbeck, E.F.(1971): “Reservoir Simulation ModelsMythology or Methodology?” SPE 3304, American Institute of Mining, Metallurgical, and Petroleum Engineers, Dallas, TX., May.

1 - 26


Chapter1-M.pdf

Recommend Documents