Changes

  | isbn    = 0891816607

  | isbn    = 0891816607

}}

}}

Fluid flow in a reservoir is controlled by bed continuity, the presence of baffles to flow, and the [[permeability]] distribution (see “Fundamentals of Fluid Row”). Reservoir heterogeneities influencing fluid flow range from large scale faults and discontinuities down to thin shale intercalations, sedimentary structures, and even pore scale features ([[:file:reservoir-modeling-for-simulation-purposes_fig1.png|Figure 1]]).

Fluid flow in a reservoir is controlled by bed continuity, the presence of baffles to flow, and the [[permeability]] distribution (see [[Fundamentals of fluid flow]]). Reservoir heterogeneities influencing fluid flow range from large scale faults and discontinuities down to thin shale intercalations, sedimentary structures, and even pore scale features ([[:file:reservoir-modeling-for-simulation-purposes_fig1.png|Figure 1]]).

[[file:reservoir-modeling-for-simulation-purposes_fig1.png|thumb|{{figure number|1}}Classification of reservoir heterogeneity types.]]

[[file:reservoir-modeling-for-simulation-purposes_fig1.png|thumb|{{figure number|1}}Classification of reservoir heterogeneity types.]]

Simulation studies (see “Conducting a Reservoir Simulation Study: An Overview”) performed at early stages of field development are done to estimate parameters such as optimal well spacing, while at later stages the objectives may be infill drilling or secondary recovery. The role of the geologist is to provide reservoir models that give a sufficient description of those parameters that control the fluid flow relevant to the planned simulation study<ref name=pt10r14>Harris, D. G., 1975, The role of geology in reservoir simulation studies: Journal of Petroleum Technology, May, p. 625–632.</ref>. The key to effective and economic field development planning lies in early recognition of those reservoir characteristics that control drive mechanisms, sweep efficiency, and consequently, well spacing requirements and possible need for pressure support. The significance of the various heterogeneity types for oil recovery is summarized in Table 1, where • denotes a strong effect, and × a moderate effect.

Simulation studies (see [[Conducting a reservoir simulation study: an overview]]) performed at early stages of field development are done to estimate parameters such as optimal well spacing, while at later stages the objectives may be infill drilling or secondary recovery. The role of the geologist is to provide reservoir models that give a sufficient description of those parameters that control the fluid flow relevant to the planned simulation study.<ref name=pt10r14>Harris, D. G., 1975, The role of geology in reservoir simulation studies: Journal of Petroleum Technology, May, p. 625–632.</ref> The key to effective and economic field development planning lies in early recognition of those reservoir characteristics that control drive mechanisms, sweep efficiency, and consequently, well spacing requirements and possible need for pressure support. The significance of the various heterogeneity types for oil recovery is summarized in Table 1, where • denotes a strong effect, and × a moderate effect.

{| class = "wikitable"

{| class = "wikitable"

==Data gathering==

==Data gathering==

The realism and reliability of the resulting model is a function of the reservoir heterogeneity but also of the planning that has gone into the data gathering. Defining suitable data gathering schemes for a specific field requires multidisciplinary cooperation and a sound understanding of the significance of the data<ref name=pt10r15>Hurst, A., Archer, J. S., 1986, Sandstone reservoir description—an overview of the role of geology and mineralogy: Clay Minerals, v. 21, p. 791–809., 10., 1180/claymin</ref>. Analog cases and sensitivity studies on prototype models can guide this process which is crucial to the reliability of the results. In Table 2, an overview is given of the value of the different types of data for the identification and quantification of reservoir heterogeneities. Only a part of the data are of geological nature. Reliable information on reservoir connectivity is often derived from production and pressure tests (see “[[Production testing]]” and “[[Pressure transient testing]]”). In particular, it is very difficult to determine large scale vertical permeability from core and log data.

The realism and reliability of the resulting model is a function of the reservoir heterogeneity but also of the planning that has gone into the data gathering. Defining suitable data gathering schemes for a specific field requires multidisciplinary cooperation and a sound understanding of the significance of the data.<ref name=pt10r15>Hurst, A., Archer, J. S., 1986, Sandstone reservoir description—an overview of the role of geology and mineralogy: Clay Minerals, v. 21, p. 791–809., 10., 1180/claymin</ref> Analog cases and sensitivity studies on prototype models can guide this process which is crucial to the reliability of the results. In Table 2, an overview is given of the value of the different types of data for the identification and quantification of reservoir heterogeneities. Only a part of the data are of geological nature. Reliable information on reservoir connectivity is often derived from production and pressure tests (see [[Production testing]] and [[Pressure transient testing]]). In particular, it is very difficult to determine large scale vertical permeability from core and log data.

{| class = "wikitable"

{| class = "wikitable"

|}

|}

The detail to which a reservoir can be modeled is a function of both the degree of heterogeneity and the data density. At an early stage of development, one may be restricted to seismic data, sparse well data, and production tests only. In these circumstances, we reap maximum benefit from knowledge of the typical flow characteristics associated with the prevailing facies and diagenetic overprint and of a data base to predict reservoir body continuity and architecture. Regional experience in analog fields can often be used to guide the modeling process<ref name=pt10r29>Slatt, R. M., Hopkins, G. L., 1990, Scaling geologic reservoir description to engineering needs: Journal of Petroleum Technology, Feb., p. 202–210.</ref>.

The detail to which a reservoir can be modeled is a function of both the degree of heterogeneity and the data density. At an early stage of development, one may be restricted to seismic data, sparse well data, and production tests only. In these circumstances, we reap maximum benefit from knowledge of the typical flow characteristics associated with the prevailing facies and diagenetic overprint and of a data base to predict reservoir body continuity and architecture. Regional experience in analog fields can often be used to guide the modeling process.<ref name=pt10r29>Slatt, R. M., Hopkins, G. L., 1990, Scaling geologic reservoir description to engineering needs: Journal of Petroleum Technology, Feb., p. 202–210.</ref>

The data that form the basis for reservoir modeling ideally comprise regional data, seismic data, cores, logs, pressure measurements, wireline formation tests, pulse tests, and well-planned production tests (Table 2). Modern three-dimensional seismic data can be used for a range of modeling purposes from structural analysis to reservoir properties such as thickness, lithology, porosity, and pore fill<ref name=pt10r27>Ruijtenberg, P. A., Buchanan, R., Marke, P. A. B., 1990, Three-dimensional data improve reservoir mapping: Journal of Petroleum Technology, Jan., p. 22–61.</ref> (also see Part 7 on Geophysical Methods).

The data that form the basis for reservoir modeling ideally comprise regional data, seismic data, cores, logs, pressure measurements, wireline formation tests, pulse tests, and well-planned production tests (Table 2). Modern three-dimensional seismic data can be used for a range of modeling purposes from structural analysis to reservoir properties such as thickness, lithology, porosity, and pore fill<ref name=pt10r27>Ruijtenberg, P. A., Buchanan, R., Marke, P. A. B., 1990, Three-dimensional data improve reservoir mapping: Journal of Petroleum Technology, Jan., p. 22–61.</ref> (also see [[Geophysical methods]]).

The effect of large scale features such as faults can be estimated on the basis of geological experience and modeling<ref name=pt10r33>Weber, K. J., Mandl, G., Pilaar, W. F., Lehner, F., Precious, R. G., 1978, The role of faults in hydrocarbon migration and trapping in Nigerian growth fault structures: Offshore Technical Conference, Houston, OTC 3356.</ref> or it must be evaluated by pressure measurements or fluid level differences.

The effect of large scale features such as faults can be estimated on the basis of geological experience and modeling<ref name=pt10r33>Weber, K. J., Mandl, G., Pilaar, W. F., Lehner, F., Precious, R. G., 1978, The role of faults in hydrocarbon migration and trapping in Nigerian growth fault structures: Offshore Technical Conference, Houston, OTC 3356.</ref> or it must be evaluated by pressure measurements or fluid level differences.

[[file:reservoir-modeling-for-simulation-purposes_fig3.png|left|thumb|{{figure number|3}}Log-facies calibration and determination of facies-related rock characteristics.]]

[[file:reservoir-modeling-for-simulation-purposes_fig3.png|left|thumb|{{figure number|3}}Log-facies calibration and determination of facies-related rock characteristics.]]

The first priority in describing the reservoir rock is the determination of the environment of deposition and the range of lithofacies that occur within the reservoir (see “Lithofacies and Environmental Analysis of Clastic Depositional Systems” and “Carbonate Reservoir Models: Facies, Diagenesis, and Flow Characterization”). Regional stratigraphic information, cores, and sidewall samples are used for this purpose. Of particular interest is the rock typing through a study of porosity, permeability, petrography, and capillary properties ([[:file:reservoir-modeling-for-simulation-purposes_fig2.png|Figure 2]]) (see Part 5 on Laboratory Methods). For simulation purposes, permeability is a major parameter, and estimating the permeability profile in noncored wells is of prime importance<ref name=pt10r38>Wolf, M., Pelissier-Combescure, J., 1982, Faciolog-automatic electrofacies determination: SPWLA Third Annual Logging Symposium Transactions, July.</ref>. The basis for these techniques is multivariate analysis of the combined logging data. Discriminant analysis of log response using a core calibrated system usually leads to the best results. In general, one has to combine several rock types into an electrofacies class mainly because of the poor vertical resolution of the nuclear logs if run in standard fashion. If the porosity and permeability relationships of combined rock classes differ little, this is an acceptable simplification ([[:file:reservoir-modeling-for-simulation-purposes_fig3.png|Figure 3]]).

The first priority in describing the reservoir rock is the determination of the environment of deposition and the range of lithofacies that occur within the reservoir (see [[Lithofacies and environmental analysis of clastic depositional systems]] and [[Carbonate reservoir models: facies, diagenesis, and flow characterization]]). Regional stratigraphic information, cores, and sidewall samples are used for this purpose. Of particular interest is the rock typing through a study of porosity, permeability, petrography, and capillary properties ([[:file:reservoir-modeling-for-simulation-purposes_fig2.png|Figure 2]]) (see [[Laboratory methods]]). For simulation purposes, permeability is a major parameter, and estimating the permeability profile in noncored wells is of prime importance.<ref name=pt10r38>Wolf, M., Pelissier-Combescure, J., 1982, Faciolog-automatic electrofacies determination: SPWLA Third Annual Logging Symposium Transactions, July.</ref> The basis for these techniques is multivariate analysis of the combined logging data. Discriminant analysis of log response using a core calibrated system usually leads to the best results. In general, one has to combine several rock types into an electrofacies class mainly because of the poor vertical resolution of the nuclear logs if run in standard fashion. If the porosity and permeability relationships of combined rock classes differ little, this is an acceptable simplification ([[:file:reservoir-modeling-for-simulation-purposes_fig3.png|Figure 3]]).

[[file:reservoir-modeling-for-simulation-purposes_fig4.png|thumb|{{figure number|4}}Correlation of reservoir units and subdivision of reservoir In flow units.]]

[[file:reservoir-modeling-for-simulation-purposes_fig4.png|thumb|{{figure number|4}}Correlation of reservoir units and subdivision of reservoir In flow units.]]

==Well correlation==

==Well correlation==

The next step in reservoir modeling is correlating the reservoir intervals from well to well ([[:file:reservoir-modeling-for-simulation-purposes_fig4.png|Figure 4]]). This procedure is strongly dependent on facies and well spacing. A requirement is a sound database of genetic unit geometry, for example, width to thickness ratio of a specific sand body type. When no deterministic correlation can be made of reservoir units, it may be necessary to use probabilistic modeling techniques, but in such cases only prototype simulations are usually carried out<ref name=pt10r13>Haldorsen, H. H., Brand, P. J., Macdonald, C. J., 1987, Review of the stochastic nature of reservoirs: Proceedings of the Seminar on Mathematics of Oil Production, Cambridge, U., K.</ref>.

The next step in reservoir modeling is correlating the reservoir intervals from well to well ([[:file:reservoir-modeling-for-simulation-purposes_fig4.png|Figure 4]]). This procedure is strongly dependent on facies and well spacing. A requirement is a sound database of genetic unit geometry, for example, width to thickness ratio of a specific sand body type. When no deterministic correlation can be made of reservoir units, it may be necessary to use probabilistic modeling techniques, but in such cases only prototype simulations are usually carried out.<ref name=pt10r13>Haldorsen, H. H., Brand, P. J., Macdonald, C. J., 1987, Review of the stochastic nature of reservoirs: Proceedings of the Seminar on Mathematics of Oil Production, Cambridge, U., K.</ref>

The framework for constructing simulation models is controlled by facies distribution, major permeability contrasts, and impermeable layers<ref name=pt10r35>Weber, K. J., Van Geuns, L. C., 1990, Framework for constructing clastic reservoir simulation models: Journal of Petroleum Technology, Oct., p. 1248–1297.</ref>. Again, maximum use should be made of reservoir engineering data with emphasis on wireline formation pressures. Geological predictions of sealing surfaces are often unreliable because of the presence of small scale faults or local erosion.

The framework for constructing simulation models is controlled by facies distribution, major permeability contrasts, and impermeable layers.<ref name=pt10r35>Weber, K. J., Van Geuns, L. C., 1990, Framework for constructing clastic reservoir simulation models: Journal of Petroleum Technology, Oct., p. 1248–1297.</ref> Again, maximum use should be made of reservoir engineering data with emphasis on wireline formation pressures. Geological predictions of sealing surfaces are often unreliable because of the presence of small scale faults or local erosion.

==Grid block patterns==

==Grid block patterns==

The reservoir has to be subdivided in such a way that all major baffles are represented and that the individual layers and bodies can be described using average petrophysical properties that lead to realistic flow simulations. For this purpose, the term ''flow unit'' has been introduced, which is defined as a volume of rock characterized by a specific combination of geological and petrophysical properties that influence fluid flow (see “Flow Units for Reservoir Characterization”). Flow units can be composed of a combination of stratigraphic elements if they do not essentially differ in fluid flow criteria<ref name=pt10r29 />).

The reservoir has to be subdivided in such a way that all major baffles are represented and that the individual layers and bodies can be described using average petrophysical properties that lead to realistic flow simulations. For this purpose, the term ''flow unit'' has been introduced, which is defined as a volume of rock characterized by a specific combination of geological and petrophysical properties that influence fluid flow (see [[Flow units for reservoir characterization]]). Flow units can be composed of a combination of stratigraphic elements if they do not essentially differ in fluid flow criteria.<ref name=pt10r29 />

The subdivision of the reservoir into flow units is typically a multidisdplinary activity because geological and reservoir engineering information has to be used together. Economic considerations with respect to numbers of grid blocks to be used also play a role.

The subdivision of the reservoir into flow units is typically a multidisdplinary activity because geological and reservoir engineering information has to be used together. Economic considerations with respect to numbers of grid blocks to be used also play a role.

Reservoir simulation operates on the principle of simultaneously solving the flow equations between adjacent blocks of rock in response to offtake from wells. The larger the number of grid blocks used, the closer the model resembles the geological prototype. However, even with the fastest computers, the simulation models are restricted to a maximum of some 60,000 grid blocks. For economic reasons, most simulations make use of less than 10,000 grid blocks. Consequently, grid block sizes normally range from 100 to [[length::300 m]] in diameter and from 5 to [[length::15 m]] in thickness (see “Conducting a Reservoir Simulation Study: An Overview”).

Reservoir simulation operates on the principle of simultaneously solving the flow equations between adjacent blocks of rock in response to offtake from wells. The larger the number of grid blocks used, the closer the model resembles the geological prototype. However, even with the fastest computers, the simulation models are restricted to a maximum of some 60,000 grid blocks. For economic reasons, most simulations make use of less than 10,000 grid blocks. Consequently, grid block sizes normally range from 100 to [[length::300 m]] in diameter and from 5 to [[length::15 m]] in thickness (see [[Conducting a reservoir simulation study: an overview]]).

Practical grid block sizes usually imply the amalgamation of a few geological features. Thus, an averaging procedure is required to obtain realistic overall flow characteristics for the blocks as a whole. The influence of discontinuous shale breaks on vertical permeability, for example, can be estimated using a statistic approach<ref name=pt10r1>Begg, S. H., Chang, D. M., Haldorsen, H. H., 1985, A simple statistical method for calculating the effective vertical permeability of a reservoir containing discontinuous shales: Society of Petroleum Engineers Symposium on Reservoir Simulation, Dallas, TX, Feb. 10–13, SPE 14271.</ref>. Averaging horizontal and vertical permeability over grid block size units is a difficult task. In practice, the effective horizontal permeability usually ranges from the arithmetic to the geometric average of the permeability profile of the block. The more continuous the sublayers of the flow unit, the closer the average Ues to the arithmetic average. The more random the permeability, the closer it gets to the geometric average. Geostatistical methods have become popular to tackle these problems<ref name=pt10r17>Journel, A. G., Alabert, F. G., 1990, New method for reservoir mapping: Journal of Petroleum Technology, Feb., p. 212–218.</ref>. Vertical permeabilities are difficult to measure, and the values used are often based either on experience for a given facies or on vertical pulse tests or other pressure data evaluation.

Practical grid block sizes usually imply the amalgamation of a few geological features. Thus, an averaging procedure is required to obtain realistic overall flow characteristics for the blocks as a whole. The influence of discontinuous shale breaks on vertical permeability, for example, can be estimated using a statistic approach.<ref name=pt10r1>Begg, S. H., Chang, D. M., Haldorsen, H. H., 1985, A simple statistical method for calculating the effective vertical permeability of a reservoir containing discontinuous shales: Society of Petroleum Engineers Symposium on Reservoir Simulation, Dallas, TX, Feb. 10–13, SPE 14271.</ref> Averaging horizontal and vertical permeability over grid block size units is a difficult task. In practice, the effective horizontal permeability usually ranges from the arithmetic to the geometric average of the permeability profile of the block. The more continuous the sublayers of the flow unit, the closer the average is to the arithmetic average. The more random the permeability, the closer it gets to the geometric average. Geostatistical methods have become popular to tackle these problems.<ref name=pt10r17>Journel, A. G., Alabert, F. G., 1990, New method for reservoir mapping: Journal of Petroleum Technology, Feb., p. 212–218.</ref> Vertical permeabilities are difficult to measure, and the values used are often based either on experience for a given facies or on vertical pulse tests or other pressure data evaluation.

[[file:reservoir-modeling-for-simulation-purposes_fig5.png|left|thumb|{{figure number|5}}Mapping of reservoir properties per grid block layer to provide input for the reservoir simulation.]]

[[file:reservoir-modeling-for-simulation-purposes_fig5.png|left|thumb|{{figure number|5}}Mapping of reservoir properties per grid block layer to provide input for the reservoir simulation.]]

Modeling of carbonate reservoirs is generally more difficult than modeling clastic reservoirs. The reason is that carbonate rocks usually undergo a much more complex pattern of diagenetic processes. As a result, the permeability distribution can be complex and poorly related to original facies distribution. A further complication can be the occurrence of open fractures.

Modeling of carbonate reservoirs is generally more difficult than modeling clastic reservoirs. The reason is that carbonate rocks usually undergo a much more complex pattern of diagenetic processes. As a result, the permeability distribution can be complex and poorly related to original facies distribution. A further complication can be the occurrence of open fractures.

Carbonate reservoirs without fractures can in principle be treated similarly to clastic reservoirs. Reservoir simulation of so-called dual porosity reservoirs is difficult because of the problem of quantifying the degree of capillary contact across fractures. Recently, methods have been proposed to tackle the problem of block-to-block interaction<ref name=pt10r26>Por, G. J., 1989, A fractured reservoir simulator capable of modelling block/block interaction: SPE Annual Technical Conference and Exhibition, San Antonio, TX, Oct. 8–11, SPE 19807.</ref>. In all cases, carbonate reservoirs require a considerable amount of core studies and frequent use of the scanning electron microscope and its auxiliary equipment.

Carbonate reservoirs without fractures can in principle be treated similarly to clastic reservoirs. Reservoir simulation of so-called dual porosity reservoirs is difficult because of the problem of quantifying the degree of capillary contact across fractures. Recently, methods have been proposed to tackle the problem of block-to-block interaction.<ref name=pt10r26>Por, G. J., 1989, A fractured reservoir simulator capable of modelling block/block interaction: SPE Annual Technical Conference and Exhibition, San Antonio, TX, Oct. 8–11, SPE 19807.</ref> In all cases, carbonate reservoirs require a considerable amount of core studies and frequent use of the scanning electron microscope and its auxiliary equipment.

==See also==

==See also==

* [[Waterflooding]]

* [[Waterflooding]]

* [[Fundamentals of fluid flow]]

* [[Fundamentals of fluid flow]]

* [[Conducting a reservoir simulation study: An overview]]

* [[Conducting a reservoir simulation study: an overview]]

* [[Introduction to reservoir engineering methods]]

* [[Introduction to reservoir engineering methods]]

* [[Petroleum reservoir fluid properties]]

* [[Petroleum reservoir fluid properties]]