CN114402233A

CN114402233A - Automatic calibration of forward deposition model

Info

Publication number: CN114402233A
Application number: CN202080024546.4A
Authority: CN
Inventors: 李宇鹏; 穆赫莱丝·穆斯塔法·梅兹加尼
Original assignee: Saudi Arabian Oil Co
Current assignee: Saudi Arabian Oil Co
Priority date: 2019-03-26
Filing date: 2020-03-25
Publication date: 2022-04-26
Also published as: EP3948365A1; CA3134777A1; WO2020198284A1; US20200308934A1

Abstract

The subject matter of this specification can be embodied in, among other things, a method for geologic modeling that includes: the method includes receiving a forward depositional model, determining a Latin Hypercube Sampling (LHS) stratigraphic model based on the projected forward depositional model, performing forward depositional modeling, transforming the forward depositional model from a time domain to a stratigraphic depth domain, determining one or more pseudo-wells based on the transformed model, determining a mismatch value based on the transformed forward depositional model and a set of simulated physical values, and determining a kriging proxy model based on the LHS stratigraphic model and the mismatch value.

Description

Automatic calibration of forward deposition model

PRIORITY CLAIM

This application claims priority from U.S. patent application No.16/365,217, filed on 3/26/2019, the entire contents of which are incorporated herein by reference.

Technical Field

The present application relates to techniques for predicting subsurface geological structures.

Background

The forward deposition modeling process typically includes several coupled or sequential sub-processes to simulate different deposition processes. This modeling process can numerically model the fluid flow, settling laws that affect erosion, transport, and deposition. Some sedimentation processes such as compaction and porosity reduction, fold deformation, diagenesis, and fluid maturation can also be numerically simulated during forward modeling of sedimentation.

The output of the numerical forward depositional simulation may be the stratigraphic space architecture (e.g., thickness of each group (formation)), and lithology or geologic facies within each group. Petrophysical properties of the simulated region, such as porosity and permeability, may also be derived in the final model.

Disclosure of Invention

In general, this document describes techniques for predicting subsurface geological structures.

In a first aspect, a method for geologic modeling comprises: receiving a forward modeling deposition model; determining a Latin Hypercube Sampling (LHS) formation model based on the projected forward depositional model; performing forward modeling deposition modeling; transforming the forward sedimentary model from a time domain to a formation depth domain; determining one or more pseudo-wells based on the transformed model; determining a mismatch value based on the transformed forward depositional model and the set of simulated physical values; and determining a Kriging (Kriging) surrogate model based on the LHS stratigraphic model and the mismatch value.

Various implementations may include some, all, or none of the following features. The forward depositional model may be based on well log data describing a drilling path through a predetermined geographic area. The method may further comprise: receiving well log data describing a drilling path through a predetermined geographic area; and projecting the drilling path to forward depositional model coordinates in the forward depositional model based on the log data. The set of simulated physical values may include a set of at least one of hydraulic values, geological values, and sedimentology values. The method may further comprise: determining a borehole path based on the identified set of predetermined input parameters; and drilling a well based on the determined borehole path. Determining the mismatch value based on the transformed forward depositional model and the set of simulated physical values may comprise: identifying a first set of geologic parameter values representing geologic properties measured at predetermined points along a drilling path; determining a second set of geologic parameter values representing simulated geologic properties of the LHS formation model at predetermined points along the drilling path; determining a set of difference values comprising differences between the selected geological parameter values in the first set and corresponding geological parameter values in the second set; and providing the set of differences as a mismatch value. Determining the kriging proxy model may include: determining a set of simulation models based on the set of LHS stratigraphic model and forward depositional model parameters; determining a set of approximation models that emulate a set of simulation models; ordering the set of approximation models based on a comparison of each approximation model to the forward depositional model; and identifying, based on the ranking, an approximation model of the set of approximation models that emulates the proxy model. Determining the set of simulation models based on the set of LHS formation model and forward depositional model parameters includes: a kriging prediction for locations not included in the forward depositional model is determined.

In a second aspect, a system for geographic modeling includes: a control system having one or more processors, and a non-transitory computer-readable medium storing instructions executable by the one or more processors to perform operations comprising: receiving a forward modeling deposition model; determining a Latin Hypercube Sampling (LHS) formation model based on the projected forward depositional model; performing forward modeling deposition modeling; transforming the forward sedimentary model from a time domain to a formation depth domain; determining one or more pseudo-wells based on the transformed model; determining a mismatch value based on the transformed forward depositional model and the set of simulated physical values; and determining a kriging proxy model based on the LHS stratigraphic model and the mismatch value.

Various embodiments may include some, all, or none of the following features. The system may further comprise: determining a borehole path based on the identified set of predetermined input parameters; and drilling a well based on the determined borehole path. Determining the mismatch value based on the transformed forward depositional model and the set of simulated physical values may comprise: identifying a first set of geologic parameter values representing geologic properties measured at predetermined points along a drilling path; determining a second set of geologic parameter values representing simulated geologic properties of the LHS formation model at predetermined points along the drilling path; determining a set of difference values comprising differences between the selected geological parameter values in the first set and corresponding geological parameter values in the second set; and providing the set of differences as a mismatch value. Determining the kriging proxy model may include: determining a set of simulation models based on the set of LHS stratigraphic model and forward depositional model parameters; determining a set of approximation models that emulate a set of simulation models; ordering the set of approximation models based on a comparison of each approximation model to the forward depositional model; and identifying, based on the ranking, an approximation model of the set of approximation models that emulates the proxy model. Determining the set of simulation models based on the set of LHS formation model and forward depositional model parameters includes: a kriging prediction for locations not included in the forward depositional model is determined.

In a third aspect, a non-transitory computer-readable medium stores instructions executable by a processing device to perform operations comprising: receiving a forward modeling deposition model; determining a Latin Hypercube Sampling (LHS) formation model based on the projected forward depositional model; performing forward modeling deposition modeling; transforming the forward sedimentary model from a time domain to a formation depth domain; determining one or more pseudo-wells based on the transformed model; determining a mismatch value based on the transformed forward depositional model and the set of simulated physical values; a kriging proxy model is determined based on the LHS stratigraphic model and the mismatch value.

Various embodiments may include some, all, or none of the following features. The forward depositional model may be based on well log data describing a drilling path through a predetermined geographic area. The operations may further include: receiving well log data describing a drilling path through a predetermined geographic area; and projecting the drilling path to forward depositional model coordinates in the forward depositional model based on the log data. The operations may further include: determining a borehole path based on the identified set of predetermined input parameters; and drilling a well based on the determined borehole path. Determining the mismatch value based on the transformed forward depositional model and the set of simulated physical values may further comprise: identifying a first set of geologic parameter values representing geologic properties measured at predetermined points along a drilling path; determining a second set of geologic parameter values representing simulated geologic properties of the LHS formation model at predetermined points along the drilling path; determining a set of difference values comprising differences between the selected geological parameter values in the first set and corresponding geological parameter values in the second set; and providing the set of differences as a mismatch value. Determining the kriging proxy model may include: determining a set of simulation models based on the set of LHS stratigraphic model and forward depositional model parameters; determining a set of approximation models that emulate a set of simulation models; ordering the set of approximation models based on a comparison of each approximation model to the forward depositional model; and identifying, based on the ranking, an approximation model of the set of approximation models that emulates the proxy model. Determining the set of simulation models based on the set of LHS formation model and forward depositional model parameters may include: a kriging prediction for locations not included in the forward depositional model is determined.

The systems and techniques described here may provide one or more of the following advantages. First, the system can provide a model of the subsurface geological structure that accurately reflects the observed conditions. Second, the system can perform modeling using relatively fewer computational resources than the prior art. Third, the system can perform modeling faster than the prior art. Fourth, the system may provide information that can improve the efficiency of drilling operations.

The details of one or more implementations are set forth in the accompanying drawings and the description to be presented. Other features and advantages will be apparent from the description and drawings, and from the claims.

Drawings

FIG. 1 is an example cross-sectional view of an example deposition environment.

FIG. 2 shows a table of several example designs of input variables.

Fig. 3 shows three examples of simulation models.

FIG. 4A is a flow diagram of an example inverse depositional modeling workflow.

FIG. 4B illustrates an example conceptual model of an example stratigraphic implementation.

FIG. 4C illustrates an example conceptual mathematical representation of a proxy model.

FIG. 4D illustrates an example proxy model represented as a mathematical surface.

FIG. 5 is a flow diagram illustrating a Latin hypercube sampling and physical transformation process.

Fig. 6A is a flow diagram of an example spatial depth transformation process.

FIG. 6B illustrates example two-dimensional and three-dimensional projections of an example well trajectory.

Fig. 7A is a flow diagram of an example mismatch determination process.

FIG. 7B shows a graphical example of the temporal-to-spatial transformation of a model.

FIG. 7C illustrates example grid values in a spatial transformation of a model.

Fig. 8 and 9 are flowcharts of an example mismatch determination process.

Fig. 10 is a flow diagram of an example parameter set generation process.

FIG. 11 is a flow chart of an example process for calibrating a forward deposition model.

Detailed Description

This document describes systems and techniques for Inverse Depositional Modeling (IDM) using an optimization method based on kriging (kriging) proxy models. The IDM method calibrates Forward Deposition Modeling (FDM) to previously observed data, expressed as well log data, and used as constraints on FDM. The following discussion describes techniques for fitting an FDM to an observed value through rapid and automatic adjustment of FDM input parameters.

One of the challenges of some forward depositional modeling approaches is to optimize various input parameters so that the simulated output will maximally match the available prior observed data (e.g., from data records obtained from physical drilling operations). This document describes techniques for finding appropriate sets of input parameters, including initial conditions and boundary conditions that can provide consistency between simulated deposits and previously observed data.

The proposed method described in this document can be applied to various specific forward depositional modeling methods that can be considered as "black boxes" in the inversion process. Implementing a kriging-agent-based optimization method may enable faster computations than previous techniques.

The forward deposition modeling process typically includes several coupled or sequential sub-processes to simulate different deposition processes. Such modeling processes can numerically model things such as fluid flow, settling laws that affect erosion, transport, and deposition. In forward modeling of sedimentation, numerical simulations can also be performed for some sedimentation processes such as compaction, porosity reduction, fold deformation, diagenesis, and fluid maturation.

The output of the quantified forward depositional simulation may be the stratigraphic space architecture (e.g., thickness of each group), lithology, or geologic facies within the group. In some embodiments, petrophysical properties of the simulated region, such as porosity and permeability, may also be derived.

Fig. 1 is an example cross-sectional view of an example deposition environment 100. The experimental deposition environment was a carbonate deposition environment. In some examples, a useful parameter for forward depositional simulations is the growth rate of each rock type. In the present example, four rock types (facies) are identified, namely lagoon 110, bank 120, algae plateau 130, and deeper open sea 140.

For forward depositional simulation processes, the input parameters are typically a set of initial terrain parameters and other input parameters describing hydraulic or depositional dispersion characteristics of the selected geological region. In some examples, a history of deformation and movement (e.g., sag) may also be used as parameters for forward depositional simulations to obtain true geological results. For some forward numerical models, the spatial and temporal distribution of physical properties and their boundary conditions may also be used as model parameters.

FIG. 2 shows a table 200 of several example designs of input variables. Each rock type (e.g., those shown in the example deposition environment 100 of fig. 1) may be associated with a range of growth rates for that rock type. For example, lagoon 110 may be associated with a growth rate of 10-80 meters per million years. The bank ridge 120 may be associated with a growth rate of 40-110 meters per million years. The algae plateaus 130 can be associated with growth rates of 0-70 meters per million years. The deep open sea 140 may be associated with a growth rate of 0-70 meters per million years. The first step is to perform a set of initial Latin Hypercube Sampling (LHS) designs. Table 200 presents an example of such an initial LHS design.

Fig. 3 shows three examples of simulation models and a legend 301 for explaining the models.

With the initial LHS design available, a set of parameters for the forward deposition model can be obtained. Forward deposition modeling may be obtained by a batch process using these parameters as inputs. In some embodiments, the process may be automated, thereby mitigating a significant amount of labor. For example, 10, 20, 50, 100 or more models may be simulated and used for such experiments.

In fig. 3, model 310 represents the phase stacking pattern of wells (the vertical section actually observed from the wells). The model 310 is based on observed data, such as measurements (logs) obtained while drilling. The model shows the various layers of dike ridges 302, lagoon 303, algae terrace 304, and deep sea area 305 layers.

The model 320 represents the simulated stack pattern (vertical section extracted from the simulation model using the proposed method) generated from manual parametric inference. The model 330 represents a stacked pattern of simulation results obtained from a process that will be described in more detail in the subsequent paragraphs (vertical sections extracted from the simulation model, but inversion parameters are manually inferred). Comparison of the three models 310-330 shows that the model 330 more closely approximates the model 310 than the model 320.

The relationship (contrast) between the initial set of LHS designs and the mismatch values calculated from the model and well comparisons is then ranked and identified using Kriging agent modeling techniques. In some examples, the efficient global search and prospective refinement principles implemented in the technology may increase the likelihood of: the next proposed design will reduce the uncertainty of the proxy model built between the LHS design and the mismatch values. Based on the recommended parameters, the process runs forward until the iterations meet various predetermined criteria, such as by determining that the mismatch value is less than a predetermined amount, or by determining that the process has iterated more than a predetermined maximum iteration count. In some embodiments, during the proxy modeling and iterative process, if the maximum expected improvement is small enough, a new mismatch value will be calculated from the proxy model rather than the forward modeling simulation. The modeling techniques will now be described in more detail.

Using the techniques described in this document, any appropriate, identified input parameters (e.g., initial water depth measurements, input sediment composition, river flow rate, transport efficiency) for forward depositional modeling may be automatically inferred with the goal of reproducing the previous observations.

The techniques described in this document provide an automated method of performing calibration of a forward depositional model based on available prior observation data (sometimes expressed as well logs). The process includes the following general stages:

(1) conditioning data (e.g., previously collected well log data) preprocessing is performed. For example, data previously collected during drilling of the well path may be projected to forward depositional model grid coordinates. The process transforms the well path into a grid cell index in the simulation model for extracting attributes from the model and for mismatch value computation.

(2) The initial experimental design was constructed using the Latin hypercube sampling design (LHS). The LHS design is transformed into physical values such as hydraulic, geological and sedimentary values.

(3) These initial experimental design runs were performed in a batch mode that automatically performed the proposed inverse deposition process for each initial experimental design run.

(4) The original depositional model is transformed from the time domain to the formation depth domain. In the case where the outputs of the simulation models are not in the expected order (e.g., where the outputs of the simulation models are saved according to time from oldest to newest, rather than from newest to oldest), at this stage the models may be transformed into the spatial domain.

(5) A mismatch value calculation is constructed based on the prior observed data and the simulation model. Depending on the previous observations (which may be thickness, phase type, porosity or permeability), different mismatch value calculation methods may be implemented.

(6) Given a predetermined set of specified input parameters, a kriging proxy model is created based on the inputs and outputs of the forward depositional modeling program.

(7) After using the searched input parameters for forward depositional modeling, a global optimization process is used to find a set of input parameters that can maximally reproduce the observed hard data.

FIG. 4A is a flow chart 400 of an example inverse depositional modeling workflow. At 410, an initial LHS design is determined, for example, based on observed well log data.

At 412, an initial forward simulation is determined. The forward deposition modeling process may be considered a numerical experiment in the experimental design notation. In some embodiments, ignoring knowledge of the internal functions modeling the evolutionary deposition may allow the process to be viewed as a black box model, where only the input-output variable relationships are considered.

In many modeling programs, depositional modeling is computationally intensive, and therefore the maximum amount of information should be obtained during a limited number of forward simulation runs. To optimize the gain of information from a certain set of runs, these runs were performed according to a suitable experimental design method.

Different experimental design approaches are typically involved in building a proxy model as an approximation of the efficient computation of numerical experiments at high computational cost. In this example process, Latin Hypercube Sampling (LHS) is used in order to make the representation of the entire range of all input parameters more uniform in each design. However, in some examples, other experimental design methods may also be used.

This step is performed to transform the experimental design from homogeneous space to physical space and write a set of forward deposition modeling parameter files for later proxy modeling. For each forward modeling, several identified parameters are used in the inversion process. For example, the parameters may be expressed as:

x＝{x1，x2，…，x_k}。

hereinafter, the design space or the design domain will be referred to as D^k. Different samples from the domain may constitute a sampling scheme.

For each parameter x in x_iThere will be maximum and minimum values to derive the depositional values to run the forward modeling process. The maximum and minimum values may be expressed as

And

for example, given a parameter x_iLHS of l_iThen x_iCan be calculated as:

in some examples, several groups (e.g., an initial set of samples) may be run. For example, N groups may be run. Given a

And

the entire parametric design matrix can be obtained:

X＝{x¹，x²，…，x^N}。

based on the forward deposition modeling parameter templates, a complete set of parameter files may be generated based on the processes described herein. The forward modeling procedure may be performed N times automatically using batch scripts and templates. In subsequent iterations, a parameter file is constructed using the parameter sets identified from the optimization process, and a forward stratigraphic modeling process is run.

After the forward deposition model is constructed, it is compared to previous observations. In some examples, the forward depositional model and the previous observations may not be in the same data or coordinate format. In such an example, some pre-processing or decoding work may be done on the simulation run before the compare or mismatch operation is performed. For example, different forward engines and different pre-processing modules may be used to process different model output formats.

In some examples of forward sedimentary procedures, the extraction from the simulation model may not be able to follow the true well trajectory with sufficient accuracy. As a solution, stratigraphic comparisons are performed prior to comparing well trajectories, which provides a basis for property extraction of real trajectories along the well path. A temporal stratigraphic comparison is performed based on the group signature of the well log data and the average thickness of the model. The comparison process projects the well path to the same temporal formation comparison as the simulated depositional model. Given a well trajectory and a grid definition of a depositional model in the spatial depth domain, the index for each cell may be searched and indexed. The along-track properties are then obtained from the simulated deposition model for later mismatch value calculation.

At 414, an initial mismatch result is determined. For example, the initial forward simulation may be compared to the observed well log data to determine a measure of how accurate (or inaccurate) the initial forward simulation emulates the observed data. The mismatch value guides proxy model building and optimization. In some examples, the challenge of log-based calibration relates to the type and nature of the data to be calibrated. As will be discussed in more detail in later paragraphs, in these examples, three types of mismatch value calculation methods are described. They are specific to interval data (e.g., group thickness), continuous variables (e.g., permeability and porosity), and categorical variables (e.g., facies type or rock type). These are three common and common types of observed data obtained from well log data.

At 416, a stop criterion is received. For example, an iteration limit value may be obtained that represents a maximum number of iterations that the modeling process is allowed to perform, after which the best available model may be identified and provided as the final output of the process 400. In another example, a time limit may be obtained that represents a maximum amount of time that the modeling process is allowed to run, after which the best available model may be identified and provided as the final output of the process 400. In yet another example, a model mismatch threshold may be received that represents a mismatch value for a simulation model that would be considered an adequate simulation of the observed data.

At 420, a proxy model is determined. The proxy model is an approximation that emulates the relationship between the evolution simulation and the observed differences in the data. The proxy model is generally simpler and therefore less computationally intensive to run than the more fully functional forward simulation. In the reverse, only one proxy model will be built. The proxy model is updated after an initial set of (x, y). FIG. 4B illustrates an example conceptual model of an example stratigraphic implementation. FIG. 4C illustrates an example conceptual mathematical representation of a proxy model.

After running the forward model, the simulated forward model output is compared to the previously observed hard data. This comparison provides a suggestion how to compare these models to previous observations used as constraints. In some embodiments, the comparison data may be referred to as the experimental response y, which is a quantitative measure of the difference between the model and the measured data.

Given that logs of porosity are available at certain locations along the vertical direction in the investigation region, the log data is denoted S^obs. A simulated log of porosity, denoted S, may be obtained from the same location^simThe difference between the logs may then be defined as the response variable of the formation model:

y＝diff(S^obs-S^sim)。

proxy solutions implement a "proxy" model for low evaluation computational cost

The model emulates the high computation of the forward earth model process f (x)Response to cost. Here, f (x) is defined by a k-vector of design variables x ∈ D. Hereinafter, D denotes a design space or a design domain. Different samples from the domain will constitute a sampling scheme X ═ X⁽¹⁾，x⁽²⁾，…，x⁽ⁿ⁾}。

Based on the availability of the forward stratigraphic modeling program, the { x } may be passed through a number of observations or samples⁽ⁱ⁾→y⁽ⁱ⁾＝f(x⁽ⁱ⁾) I is 1,2, …, n to obtain f (x). These are computationally expensive to obtain and are therefore rarely used. The task of proxy modeling is to construct an approximation using the sparse set of samples

It can be used for performance prediction with lower computational cost for design x ∈ D.

Using the identified design variables k, a learning data set { (x) can be determined⁽¹⁾,y⁽¹⁾),(x⁽²⁾,y⁽²⁾),…,(x⁽ⁿ⁾,y⁽ⁿ⁾)}. Using a common structure

The shape of the model is determined by a set of parameters w. The early step in the process is to select the vector w so that the model will fit to the hard data. At 422, a minimum mismatch in the proxy model is identified. Assume that a proxy model (expressed as a surface in fig. 4D) has been established. From the fitted surface/proxy model, a minimum search may be performed. If the searched minimum is greater than a predetermined stopping minimum criterion, the proxy model (surface) may be identified as not good enough and another LHS design and corresponding y-variables should be added. The minimum mismatch value will be compared to a predefined stopping criterion to decide on a subsequent stopping action, such as shown at 430.

For example, after a forward modeling depositional model is run through the simulation, the obtained formation model may be compared to previously observed data. The comparison may provide an indication of the different degrees to which these models are compared to the previous observations used as constraints. This will be the response of the experiment, denoted as y_iIt is a simulation model anda quantitative measure of the difference between the observed data.

Based on a previous observation of well log data at a certain position along the vertical direction in the investigation region, the well log data is denoted S_obs. The same location based simulated log is denoted S_simAnd a total of N simulated deposition models. Then, all mismatch values would be calculated as:

yⁱ＝diff(S_obs，S_sim)，i＝1，2，…，N。

wherein y isⁱIs the difference between each simulation model and the current previous observed data set (this would be some thickness or facies logs or some porosity or permeability sequence). The operator diff is a difference operator. However, in some embodiments, operators may be defined differently depending on the nature of the observed data.

At 430, a determination is made. If the mismatch does not satisfy the stopping criteria, another determination is made at 440. If the amount of time or number of iterations exceeds a threshold of the maximum number of iterations or maximum amount of time for performing iterations, the process 400 continues at 442. For example, the process 400 may be configured to stop after 1000 iterations, or after 10 minutes of computation.

At 442, a new proxy model design is obtained. If the improvement of the new model over the previous model is small enough at 450, a new mismatch value is calculated at 452. If the improvement of the new model is not small enough at 450, forward modeling is performed again at 454. In either case, the LHS design and mismatch results are updated at 465, and a new batch of proxy models is built based on this information at 420.

However, if either determination at 430 or 440 is positive (e.g., time to stop), then the file is generated at 460. For example, the best proxy model found during the previous step may be saved to a storage medium, such as an electronic document that may be saved, archived, and transmitted.

At 462, forward modeling is performed based on the saved file. For example, the stored model may be transmitted through a flash drive to another computer for forward modeling, and the computer may read the flash drive as part of the forward modeling process.

Fig. 5 is a flow diagram of an example latin hypercube sampling and physical transformation process 500. In some implementations, the process 500 may be step 410 of the example process 400 shown in fig. 4A.

At 510, a total number of experimental variables is received. Here, the experimental variable total is the number of key parameters identified from the modeler. For example, for fig. 1, the total number is 4.

At 515, a LHS design total is obtained. Here, the total is the initial LHS design quantity. For example, in the example of fig. 2, the total number of LHS designs is 100.

At 525, an LHS design is performed. The results of LHS design are in the range of 0 to 1 (or [0,1 ]). The result is a matrix. The matrix size is determined from 510 and 515. For example, in the example of fig. 2, the size is 4 × 100.

At 530, ranges for each target inversion variable are obtained. For example, the growth rate of each sedimentation type can be obtained. At 535, given the range of each geological parameter and the LHS design in [0,1] space, the LHS design in geological space is obtained.

The initial forward depositional modeling results of step 530 are saved in terms of geologic time coordinates. For example, along the vertical direction, the time increments are equal, and all analog attributes (e.g., thickness, phase type, or other continuous attributes) are saved within the time range.

At 540, a forward model parameter file template is obtained. In some embodiments, here, the forward modeling program may be run according to a parameter file or template. In some embodiments, there may be many variables for forward modeling to run. Some of them may be assigned different values to obtain different models.

At 545, a set of keywords for the design is identified from the template. The identified inversion target forward modeling parameter variables are marked in the template. The tagged variables identify the variables in the parameter range space that are to be replaced by the LHS design, as shown in figure 5.

At 550, an initial forward modeling parameter file set is generated. In this set, each parameter file is obtained from the LHS design and is sufficient to perform normal forward runs.

Fig. 6A is a flow diagram of an example spatial depth transformation process 600. In some implementations, the process 600 may be performed based on the output of the example process 500 of fig. 5. In some examples, the temporal-to-spatial depth transformation may be performed prior to extracting the analog data values and comparing with the well log data (which is typically expressed in depth).

At 610, a stratigraphic model expressed in geologic time coordinates is received. For example, the output of the example process 500 of fig. 5 may be received.

At 620, information regarding the depth and thickness of the geological formation may be extracted from the received model. In some implementations, this information may be saved as part of the final model but not in an explicit format, according to forward modeling. In order to perform the spatial transformation, some decoding and extraction work is required to pick up from the saved forward models the information that changes from one forward modeling routine to another.

At 630, the stratigraphic model is transformed into spatial depth coordinates. For example, the model may be analyzed to determine that the model describes a lagoon layer 10 meters thick, a bank layer 100 meters thick, an algae terrace 150 meters thick, and a deep open sea layer 90 meters thick.

At 640, spatial contrast in spatial depth coordinates is provided. After transforming the model from time to space, simulated properties of some pseudo-wells are extracted at 650. However, the top depths of the wells may not all have the same depth. At 660, spatial stratigraphic comparisons in spatial depth coordinates are performed on the extracted pseudo-wells. The spatial contrast starts with the top of all wells at the same depth. The comparison also allows for proper mismatch calculations to be made based on each attribute of all wells. In the reservoir, the observed wells have their respective spatial locations and respective trajectories. These position and trajectory information may be projected to a numerical model, an example of which is shown in FIG. 6B. FIG. 6B illustrates example two-dimensional and three-dimensional projections of an example well trajectory.

Fig. 7A is a flow diagram of an example mismatch determination process. After transforming the model from the time domain to the spatial depth domain (e.g., in process 600 of FIG. 6A), the transformed stratigraphic contrast is applied to the simulation domain, which will ensure that the model is preserved in the simulation domain divided into small cells, and that the cells (3D cube cells) have the same size. FIG. 7B shows a graphical example of the temporal-to-spatial transformation of a model. FIG. 7C shows example grid values of a model in a spatial transformation.

Fig. 7B and 7C show a small example of a process that illustrates the temporal-to-spatial transformation in the model and in the well.

Sub-diagrams

750a and 760a show examples of simulated outputs from a conventional forward depositional model. The attributes are recorded in terms of analog time increments in the vertical direction. Sub-graphs 750b and 760b illustrate an example of the transformation of a simulation model from a time recording to spatial coordinates. The top or bottom may not occur at the same spatial depth.

Subgraphs

750c and 760c show examples of stratigraphic contrast transforms. Model attributes are stored in each cell.

In some embodiments, different output data retention logic may be employed depending on the different forward engines used, but the same general principles may be implemented. That is, to extract the simulated attributes from the previous observed positions, the coordinates should be the same. The proposed method is illustrated here with specific output model data. The groups are saved in chronological order from oldest to newest (e.g., by year) according to forward modeling software. Attributes such as the thickness of the group, the facies type, and other (including continuous) measurements (e.g., porosity, permeability, sand rate, and any other suitable combination of these or any other suitable attributes of the group) are stored in each time interval, which is modeled in terms of substantially equally-growing geologic time.

The model may be expressed in the spatial depth domain based on the vertical depth and thickness information for each geologic simulation time. After the model is transformed from the time domain to the spatial depth domain, the transformed stratigraphic contrast is applied to the simulation domain, which will ensure that the model is preserved in a "candy-chunk" -like conceptual arrangement of grid cells. In some embodiments, the intent of this arrangement may be to provide easy and accurate hard data extraction for later mismatch function calculations.

For stratigraphic contrast transformation, the new relative spatial locations are obtained using the following equations:

here, Z_relRepresenting relative spatial depth, Z, in the spatial domain_cbIndicating the bottom of the formation, Z_ctRepresenting the top of the formation, T represents Z_cbAnd Z_ctAverage thickness in between. Converting all depth measurements to Z_relAllows modeling of each reservoir time formation in conventional cartesian coordinates, thereby facilitating comparison between the model and the well.

Fig. 7A is a flow diagram of an example mismatch determination process 700. In some implementations, the process 700 may be performed on data provided by the example process 600 of fig. 6A.

At 710, a target group thickness is received. For example, process 600 may provide a set of information describing group thicknesses, and may receive this information for use at step 710.

At 720, a corresponding thickness is identified from the simulation model. For example, the well log data and the simulation data may each include a first layer, a second layer, and a third layer (and so on), each layer having its own thickness.

At 730, a mismatch between the target group thickness and the simulated group thickness is calculated. For example, the well log data may indicate that there are layers of 100 meters, 150 meters, 75 meters, and 120 meters thick, respectively, while the simulation model may describe layers of 110 meters, 100 meters, 80 meters, and 120 meters thick, respectively. In this example, the mathematical differences between the two data sets may be compared to determine the amount of mismatch between the layers, which in this example would be 10 meters, 50 meters, 5 meters and 0 meters, respectively.

Typically, prior to performing forward depositional modeling, a stratigraphic framework of the target reservoir is established. In this example, assume that the target formation has been identified by the modeler and has been labeled as horizon data.

Such grids (expressed as horizons) can also be identified from the simulation model along the vertical direction according to a clear geologic time definition. The process 700 identifies the top and bottom of the target well, here denoted as h_obs. Then, from each simulation model, at the same well location, a simulated thickness, here denoted as

The difference between the observation well and each simulation model is represented as:

here, the index i is an index of the simulation model.

Fig. 8 is a flow diagram of an example mismatch determination process 800. In some implementations, the process 800 may be performed on data provided by the example process 600 of fig. 6A.

At 810, a target facies stacking pattern is received. For example, the process 600 may provide a collection of information describing how the geological layer types are arranged on top of each other, and may receive this information for use at step 810.

At 820, a corresponding stacking pattern is identified from the simulation model. For example, the log data and the simulation data may each include bank layers stacked on lagoon layers, stacked on algae plateau layers, stacked on deep sea bed layers (etc.).

At 830, a mismatch between the target group thickness and the simulated group thickness is calculated. For example, the well log data may indicate the presence of a bank layer stacked on a lagoon layer, stacked on an algae plateau layer, stacked on a deep sea bed layer, and the simulation model may describe a bank layer stacked on an algae plateau layer, stacked on another bank layer, stacked on an algae plateau layer, stacked on a deep sea bed layer. In this example, the differences in ordering between the two data sets may be compared to determine the amount of mismatch between the layers. For example, the simulation model may be expressed as a 10%, 20%, 1%, or any other suitable descriptive value mismatched from the observed data.

For facies stacking patterns, the first step is to study the numerical encoding of the facies types of the fields. As an example, the phase in the research domain may be [ ultralong coarse face rock, sand, shale ]]([domits，sand，shale]) And these values can be numerically transformed to represent 1,2,3]. In some examples, the phase codes may be transformed into an integer set [1,2, …, K]The purpose is to make a numerical comparison of them. In the present example, it is assumed that the observation wells of the current group are known and there will be a total of n observations from top to bottom. The structure of the group may be denoted as S_obs＝{k₁，k₂，…，k_nWhere each observation k_iIs from the set [1,2, …, K ]]A phase type of (2).

From the simulation models, a simulated facies stacking sequence, represented as a simulated facies stacking sequence, may also be extracted from the same well location for each simulated formation model

In many examples, the number of facies observations from the simulation model is different from the number obtained from previously observed well data. Assuming that the number of observation phases is m, the structure of the simulated formation can be expressed as:

each simulated well is then resampled. For example, the maximum number of phase observations may be implemented each time the maximum number of phase observations is equal to m or n. The phase observation sequences may be resampled according to the maximum number m or n. After the sequences are resampled, they have the same observation length. Indicator transformations are performed on these layers:

here, index i would be the simulation model and index j would be along the modelThe type of phase found by the selected trajectory. Then, from the sequence pair

Response variable y of_iIs calculated as:

fig. 9 is a flow diagram of an example mismatch determination process 900. In some implementations, the process 900 can be performed on data provided by the example process 600 of fig. 6A.

At 910, a target (continuous) log is received. For example, the process 600 may provide a set of information describing sensor data detected at various points along the length of the wellbore (which may be serpentine rather than perfectly linear or vertical), and this information may be received for use at step 910.

At 920, a corresponding simulated log is identified from the simulation model. For example, the path of the actual well may be reconstructed in a simulation model, and a simulation record may be obtained that simulates the wellbore.

At 930, a mismatch between the target log and the simulated log is calculated. For example, the log data may indicate that there was a bank ridge in the first 100 meters, a lagoon next 150 meters, an algae plateau next 50 meters, another lagoon next 50 meters, and a simulation model may describe that there was a bank ridge in the first 105 meters, a lagoon next 200 meters, and a deep sea area next 50 meters. In this example, the mathematical difference between the two data sets may be compared to determine an amount of mismatch between the logs, which may be expressed as a fractional value, a percentage value, or any other suitable expression of mismatch between the data sets.

Continuous measurements along a well trajectory generally refer to measurements of properties such as porosity, permeability, or sand. Assume that the observed sequential attribute data is:

S_obs＝f_jwhere j is 1,2, …, n.

From one of the simulation models, a simulation measurement can be determined for the same well trajectory and expressed as:

wherein j is 1,2, …, n; i is 1,2, …, N.

In some examples, the simulated wells at the observation locations may be different. In such an example, the simulated pseudo-wells extracted from each simulated well may be resampled. The entire sequence can be resampled according to the number n and the resulting mismatch values from successive measurements

Will be calculated as:

the index i represents the simulation model and the index j represents the measured property along the selected trajectory in each simulation model.

In some examples of the inversion process, a sequence calibration process may be employed. For example, a certain formation group thickness may be fitted first. Thereafter, classification variables (e.g., facies or rock types) can be fitted, and then continuous measurements (e.g., porosity and permeability) can be fitted. In another example, a user may define a global objective function for one calibration. In this case, the global mismatch value may be given as:

wherein λ_thickness、λ_catAnd λ_contIs a weight for each component of the mismatch calculated from the thickness, the classification variable and the continuous variable.

Fig. 10 is a flow diagram of an example parameter set generation process 1000. At 1010, a current proxy model is received. For example, the process 1000 may be used for a general iterative process in which a plurality of proxy models are generated and refined, and at 1010, one of the proxy models generated in the current iteration may be received. In some implementations, the proxy model received at 1010 may be an output of any of the example process 700 of fig. 7A, the example process 800 of fig. 8, and the example process 900 of fig. 9.

At 1020, it is determined whether the process 1000 is at its final iteration. If so, at 1030, a mismatch function is performed to correlate the mismatch scores from the proxy model, which is also a surface fitted based on the inverted target variables. From the minimum mismatch value, the LHS design can be solved, which will be the best value for the inversion process and will be used for the process in 1050.

At 1032, an LHS design having a relatively minimum degree of mismatch is identified based on the mismatch score. This is the core technology of optimization. The proxy model concatenates the input parameters with mismatch values calculated from a comparison of previously observed well logs to the simulation model. It will therefore replace the function of a complex forward depositional simulation engine and save computation when searching for the best input parameters for the best calibration of previous hard data. In some examples, prior data for the depositional model may be collected through seismic surveys, well logs, and other processes. In the examples described in this document, emphasis has been placed on calibration of well log data. However, in other examples, the general principles of inversion are substantially the same, with the difference in calibration of different data being the mismatch value calculation used.

If at 1020 it is determined that process 1000 is not at its final iteration, process 1000 continues at 1040. At 1040, designs that provide a relatively maximum amount of improvement to the proxy model are identified. Since in this example the minimum mismatch value from the proxy surface is still greater than the stopping criterion, further proxy work is performed. The maximum improvement algorithm will ensure that the next LHS design will bring the maximum improvement to the proxy model.

At 1042, the best candidate for the next LHS design is obtained from 1040. The LHS design is in [0,1] space. It will be transformed to geological space and will be used to construct a set of parameters for process 1050.

At 1050, a parameter set based on the output of 1032 or 1042 is determined. For example, a set of parameters may be generated for use in forward stratigraphic modeling processes. Agent-based optimization techniques utilize low-computational-cost "agent" models

The "proxy" model emulates the computationally expensive response of the forward modeling process f (x). Here, f (x) is defined by the k-vector of design variables: x ═ x₁，x₂，…，x_k}. Different designs x from the design domainⁱExperimental sampling scheme X ═ X for compositional values¹，x²，…x^NWhere N is the total number of experimental samples.

Each design sample x from the sample design domainⁱWill be a set of input forward depositional model parameters and will obtain a simulated stratigraphic model after being fed to the specified forward engine f (x)

That is, based on the availability of forward depositional modeling programs, some simulation models may be obtained and they may be represented as:

the computational cost of acquiring these models is relatively high and therefore may only be marginally available. The goal of the proxy modeling technique is to construct an approximation using the sparse set of design input/output observation samples

It can be used for any new design x^*Low cost performance prediction is performed.

Based on the prior observation data S_obsUsing equation yⁱ＝diff(S_obs，S_sim)，i＝1，2，…, N to calculate the mismatch value.

Then, the pairs of input factors and their associated responses will form a set of data { (x)¹，y¹)，(x²，y²)，…，(x^N，y^N) Is used for establishing a proxy model

Selecting a generic structure

And the shape of the proxy model is based on a set of parameters W. Thus, the first step in this sub-process is to select the vector W so that the model will best fit the hard data.

Kriging prediction is also performed at non-sampled locations. For example, a new design position in the experimental sample domain may be represented as x^*And the prediction to be made is expressed as:

x^*of

The prediction will be generally consistent with the observed data (the initial sampling plan and observations calculated from the simulation model) and thus will be generally consistent with the calculated contrast parameters. Thus, prediction (given the contrast parameters and predictions) increases the likelihood that sample data will be selected for the next proxy modeling construct iteration.

An Efficient Global Optimization (EGO) process is designed for global optimization of numerical models at high evaluation computational cost. The EGO algorithm is employed to find a representative set of input parameters in its super-dimensional space. The initial design is determined (as described previously). The algorithm will then sequentially access the current global maximum of expected improvement to the current proxy model and update the kriging proxy model at each iteration.

In some embodiments, a workflow may be executed by: evaluating y at an initial set using LHS designⁱEstimating the covariance using the initial design samplesVariance function, determining expected improvement of each candidate location in space, and searching design space D^kThe greatest expected improvement in. The position of greatest improvement is the next sample point x⁺It will bring the best improvement to the proxy model. The kriging method is then performed on the picked location and a prediction is added to the measured data. The stopping criteria are then identified based on the maximum expected improvement. When the stopping criterion is met, stopping and using the new set of sample points to evaluate the covariance function and iterating.

Evaluating y at an initial set for using LHS designsⁱAssuming that the original forward model has been run N times with different sets of input parameters. Classical space filling methods such as latin hypercube sampling may be used. For the step of estimating the covariance function, a kriging-based proxy modeling is employed and the training data set obtained in the previous step will be used, as described previously.

Fig. 11 is a flow diagram of an example process 1100 for calibrating a forward deposition model (such as those described in the previous paragraphs).

At 1110, a forward depositional modeling routine is received. In some embodiments, forward deposition modeling may be accomplished by a software program.

At 1112, a Latin Hypercube Sampling (LHS) design is determined based on the projected forward depositional model input variables.

At 1114, the LHS design is transformed into a set of simulated physical values. In some implementations, the set of simulated physical values can be a set of at least one of hydraulic values, geological values, and sedimentology values.

At 1116, a forward modeling process is performed to obtain all forward models.

At 1118, the forward depositional model is transformed from the time domain to the formation depth domain.

At 1120, one or more pseudo-wells are determined (extracted) based on the simulation model. In general, the location of the extracted pseudo-wells should be the same as the location of previously observed or drilled wells in the study area.

At 1122, a mismatch value is determined based on the transformed forward depositional model and the set of simulated physical values. In some implementations, determining the mismatch value based on the transformed forward depositional model and the set of simulated physical values may include: identifying a first set of geologic parameter values representing geologic properties measured at predetermined points along a drilling path; determining a second set of geologic parameter values representing simulated geologic properties of the first stratigraphic model at predetermined points along the drilling path; determining a set of difference values representing differences between the selected values of the geological parameter in the first set and corresponding values of the geological parameter in the second set; and providing the set of differences as a mismatch value.

At 1124, a kriging proxy model is determined based on the set of LHS designs (which are x variables), mismatch values (calculated as y values from comparison of all the simulation forward models and the well bore). The proxy model will use a function to connect x and y together.

At 1126, the proxy model and iteration criteria are examined and a determination is made at 1130. If it is determined that the proxy model is insufficient, process 1100 continues at 1116. However, if the proxy model is determined to be good, then a minimum mismatch search is performed based on the current proxy model at 1140.

At 1142, a new LHS design is determined based on the current minimum mismatch search results. For example, if the proxy model surface is not good enough, a new LHS design can be obtained to improve the proxy model. The EGO algorithm can ensure that the next proposed LHS design improves the proxy model over the previously determined model.

At 1144, a set of parameters is generated for forward stratigraphic modeling. For example, when a new LHS design is proposed, the design may provide x for the proxy model function. y may be calculated by running a forward depositional model and comparing the simulated output to the observed value.

Implementations of the subject matter and the functional operations described in this specification can be implemented in digital electronic circuitry, in tangibly embodied computer software or firmware, in computer hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Software implementations of the described subject matter can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible, non-transitory computer-readable computer storage medium for execution by, or to control the operation of, data processing apparatus. Alternatively or additionally, the program instructions may be encoded on an artificially generated propagated signal (e.g., a machine-generated electrical, optical, or electromagnetic signal) that is generated to encode information for transmission to suitable receiver apparatus for execution by data processing apparatus. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or a combination of computer storage media.

The terms "data processing apparatus," "computer," or "electronic computer device" (or equivalents thereof as understood by those of ordinary skill in the art) refer to data processing hardware and include various apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. An apparatus may also be, or may also include, special purpose logic circuitry, e.g., a Central Processing Unit (CPU), a Field Programmable Gate Array (FPGA), or an Application Specific Integrated Circuit (ASIC). In some embodiments, the data processing apparatus or dedicated logic circuit (or a combination of the data processing apparatus or dedicated logic circuit) may be hardware-based or software-based (or a combination of hardware-based and software-based). Alternatively, the apparatus may comprise code that creates an execution environment for the computer program, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of execution environments. The present disclosure contemplates the use of a data processing device with or without a conventional operating system (e.g., LINUX, UNIX, WINDOWS, MAC OS, ANDROID, or IOS or any other suitable conventional operating system).

A computer program (which can also be referred to or described as a program, software application, module, software module, script, or code) can be written in any form of programming language, including compiled or interpreted languages, or declarative or procedural languages. A computer program can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. The computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network. While the portions of the programs shown in the various figures are illustrated as individual modules implementing various features and functions through various objects, methods, or other processes, the programs may instead include multiple sub-modules, third party services, components, libraries, and the like, as appropriate. Rather, the features and functionality of the various components may be combined into a single component as appropriate. The threshold value for making the calculation determination may be determined statistically, dynamically, or both.

Implementations of the described subject matter may include one or more features, either alone or in combination.

For example, in a first aspect, a method for geologic modeling comprises: receiving a forward modeling deposition model; determining a Latin Hypercube Sampling (LHS) formation model based on the projected forward depositional model; performing forward modeling deposition modeling; transforming the forward sedimentary model from a time domain to a formation depth domain; determining one or more pseudo-wells based on the transformed model; determining a mismatch value based on the transformed forward depositional model and the set of simulated physical values; and determining a kriging proxy model based on the LHS stratigraphic model and the mismatch value.

Implementations of the subject matter and the functional operations described in this specification can be implemented in digital electronic circuitry, in tangibly embodied computer software or firmware, in computer hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Software implementations of the described subject matter can be implemented as one or more computer programs. Each computer program may include one or more modules of computer program instructions encoded on a tangible, non-transitory computer-readable computer storage medium for execution by, or to control the operation of, data processing apparatus. Alternatively or additionally, the program instructions may be encoded on an artificially generated propagated signal. By way of example, the signals may be machine-generated electrical, optical, or electromagnetic signals that are generated to encode information for transmission to suitable receiver apparatus for execution by the data processing apparatus. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or a combination of computer storage media.

The terms "data processing apparatus," "computer," and "electronic computing device" (or equivalents thereof as understood by those of ordinary skill in the art) refer to data processing hardware. For example, a data processing apparatus may encompass all types of apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. An apparatus may also comprise special purpose logic circuitry, e.g., a Central Processing Unit (CPU), Field Programmable Gate Array (FPGA), or Application Specific Integrated Circuit (ASIC). In some embodiments, the data processing apparatus or dedicated logic circuit (or a combination of the data processing apparatus or dedicated logic circuit) may be hardware-based or software-based (or a combination of hardware-based and software-based). Alternatively, the apparatus may comprise code that creates an execution environment for the computer program, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of execution environments. The present disclosure contemplates the use of data processing devices with or without conventional operating systems (e.g., LINUX, UNIX, WINDOWS, MAC OS, ANDROID, or IOS).

A computer program (which may also be referred to or described as a program, software application, module, software module, script, or code) can be written in any form of programming language. The programming language includes, for example, a compiled, interpreted, declarative, or procedural language. A computer program can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or unit suitable for use in a computing environment. The computer program may, but need not, correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files that hold one or more modules, sub programs, or portions of code. A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network. While the portions of the program shown in the figures may be illustrated as individual modules implementing various features and functions through various objects, methods, or processes, the program may alternatively include multiple sub-modules, third party services, components, and libraries. Rather, the features and functionality of the various components may be combined into a single component, as appropriate. The threshold value for making the calculation determination may be determined statistically, dynamically, or both.

The methods, processes, or logic flows described in this specification can be performed by one or more programmable computers executing one or more computer programs to perform functions by operating on input data and generating output. The methods, processes, or logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., a CPU, FPGA, or ASIC.

A computer suitable for executing a computer program may be based on one or more general purpose and special purpose microprocessors, as well as other types of CPUs. The elements of a computer are a CPU for executing instructions and one or more memory devices for storing instructions and data. Generally, a CPU can receive instructions and data from (and write data to) a memory. A computer will also include, or be operatively coupled to store data, one or more mass storage devices for storing data. In some implementations, a computer can receive data from and transfer data to a mass storage device, which can include, for example, a magnetic, magneto-optical disk, or optical disk. Further, the computer may be embedded in another device, e.g., a mobile telephone, a Personal Digital Assistant (PDA), a mobile audio or video player, a game player, a Global Positioning System (GPS) receiver, or a portable storage device such as a Universal Serial Bus (USB) flash drive.

Computer-readable media (transitory or non-transitory, as applicable) suitable for storing computer program instructions and data can include all forms of persistent/non-persistent and volatile/non-volatile memory, media and memory devices. The computer-readable medium may include, for example, semiconductor memory devices such as Random Access Memory (RAM), Read Only Memory (ROM), phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Erasable Programmable Read Only Memory (EPROM), Electrically Erasable Programmable Read Only Memory (EEPROM), and flash memory devices. The computer readable medium may also include, for example, magnetic devices such as magnetic tapes, cassettes, cartridges, and internal/removable disks. The computer-readable medium may also include magneto-optical disks and optical storage devices and technologies, including, for example, Digital Video Disks (DVDs), CD ROMs, DVD +/-R, DVD-RAMs, DVD-ROMs, HD-DVDs, and Bluetooth. The memory may store various objects or data, including: caches, classes (classes), frames, applications, modules, backup data, jobs, web pages, web page templates, data structures, database tables, knowledge bases, and dynamic information. The objects and data types stored in memory may include parameters, variables, algorithms, instructions, rules, constraints, and references. Further, the memory may include logs, policies, security or access data, and reporting files. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

Embodiments of the subject matter described in this disclosure can be implemented on a computer having a display device for providing interaction with a user, including displaying information to the user (and receiving input from the user). Types of display devices may include, for example, Cathode Ray Tubes (CRTs), Liquid Crystal Displays (LCDs), Light Emitting Diodes (LEDs), and plasma monitors. The display device may include a keyboard and a pointing device (including, for example, a mouse, trackball, or touch pad). User input may also be provided to the computer through the use of a touch screen, such as a tablet computer surface with pressure sensitivity, or a multi-touch screen using capacitive or electrical sensing. Other types of devices may be used to provide interaction with the user, including receiving user feedback, including, for example, sensory feedback (including visual feedback, auditory feedback, or tactile feedback). Input from the user may be received in the form of sound, speech or tactile input. Further, the computer may interact with the user by sending documents to and receiving documents from the device used by the user. For example, by sending a web page to a web browser on a user client device in response to a request received from the web browser.

The terms "graphical user interface" or "GUI" may be used in the singular or plural to describe one or more graphical user interfaces and each display of a particular graphical user interface. Thus, the GUI may represent any graphical user interface, including but not limited to a web browser, touch screen, or Command Line Interface (CLI) that processes information and effectively presents the results of the information to the user. In general, a GUI may include a plurality of User Interface (UI) elements, some or all of which are associated with a web browser, such as interactive fields, drop-down lists, and buttons. These and other UI elements may be related to or represent functionality of a web browser.

Embodiments of the subject matter described in this specification can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server). Moreover, the computing system can include a front-end component, e.g., a client computer having one or both of a graphical user interface and a web browser through which a user can interact with the computer. The components of the system can be interconnected by any form or medium of wired or wireless digital data communication (or combination of data communication). Examples of communication networks include a Local Area Network (LAN), a Radio Access Network (RAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), Worldwide Interoperability for Microwave Access (WIMAX), a Wireless Local Area Network (WLAN) (using, for example, 802.11a/b/g/n or a combination of 802.20 or protocols), all or a portion of the internet, or any other communication system (or combination of communication networks) at one or more locations. The network may transport, for example, Internet Protocol (IP) packets, frame relay frames, Asynchronous Transfer Mode (ATM) cells, voice, video, data, or a combination of communication types between network addresses.

The computer system may include clients and servers. A client and server may be generally remote from each other and may typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship.

The clustered file system may be any file system type accessible from multiple servers for reading and updating. Since the locking of the swap file system may be done at the application layer, locking or consistency tracking may not be necessary.

While this specification contains many specific implementation details, these should not be construed as limitations on the scope of what may be claimed, but rather as descriptions of features that may be specific to particular implementations. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination. Furthermore, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination.

Specific embodiments of the present subject matter have been described. Other implementations, modifications, and substitutions of the described implementations are apparent to those of skill in the art and are within the scope of the following claims. Although operations are depicted in the drawings or claims in a particular order, this should not be understood as: it may be desirable to perform the operations in the particular order shown, or in sequential order, or to perform all of the operations shown (some of which may be considered optional) in order to achieve desirable results. In some cases, multitasking or parallel processing (or a combination of multitasking and parallel processing) may be advantageous and performed as appropriate.

Moreover, the separation or integration of various system modules and components in the foregoing implementations should not be understood as requiring such separation or integration in all implementations, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

Accordingly, the foregoing example embodiments do not define or limit the present disclosure. Other changes, substitutions, and alterations are also possible without departing from the spirit and scope of this disclosure.

Moreover, any claimed embodiments are considered applicable to at least one computer-implemented method; a non-transitory computer-readable medium storing computer-readable instructions for performing a computer-implemented method; and a computer system comprising a computer memory interoperably coupled with a hardware processor configured to perform a computer-implemented method or instructions stored on a non-transitory computer-readable medium.

Claims

1. A method for geologic modeling, comprising:

receiving a forward modeling deposition model;

determining a Latin hypercube sampling LHS stratum model based on the projected forward modeling deposition model;

performing forward modeling deposition modeling;

transforming the forward depositional model from a time domain to a formation depth domain;

determining one or more pseudo-wells based on the transformed model;

determining a mismatch value based on the transformed forward depositional model and the set of simulated physical values; and

determining a kriging proxy model based on the LHS stratigraphic model and the mismatch value.

2. The method of claim 1, wherein the forward depositional model is based on well log data describing a drilling path through a predetermined geographic area.

3. The method of claim 2, further comprising:

receiving the well log data describing a drilling path through a predetermined geographic area; and

projecting the drilling path to forward depositional model coordinates in the forward depositional model based on the well log data.

4. The method of claim 1, wherein the set of simulated physical values comprises a set of at least one of hydraulic values, geological values, and sedimentology values.

5. The method of claim 1, further comprising:

determining a borehole path based on the identified set of predetermined input parameters; and

drilling a well based on the determined borehole path.

6. The method of claim 1, wherein determining a mismatch value based on the transformed forward depositional model and the set of simulated physical values comprises:

identifying a first set of geologic parameter values representing geologic properties measured at predetermined points along a drilling path;

determining a second set of geologic parameter values representing simulated geologic properties of the LHS formation model at predetermined points along the drilling path;

determining a set of difference values comprising differences between the selected geological parameter values of the first set and corresponding geological parameter values of the second set; and

providing the set of difference values as the mismatch value.

7. The method of claim 1, wherein determining a kriging proxy model comprises:

determining a plurality of simulation models based on the set of LHS formation models and forward depositional model parameters;

determining a plurality of approximation models that emulate the plurality of simulation models;

ranking the plurality of approximation models based on a comparison of each approximation model to the forward depositional model; and

based on the ranking, identifying ones of the plurality of approximation models that emulate the proxy model.

8. The method of claim 7, wherein determining a plurality of simulation models based on the set of LHS formation model and forward depositional model parameters comprises: determining a kriging prediction for locations not included in the forward depositional model.

9. A system for geographic modeling, comprising:

a control system comprising one or more processors; and

a non-transitory computer-readable medium storing instructions executable by the one or more processors to perform operations comprising:

receiving a forward modeling deposition model;

performing forward modeling deposition modeling;

determining one or more pseudo-wells based on the transformed model;

10. The system of claim 9, further comprising:

drilling a well based on the determined borehole path.

11. The system of claim 9, wherein determining a mismatch value based on the transformed forward depositional model and the set of simulated physical values comprises:

providing the set of difference values as the mismatch value.

12. The system of claim 9, wherein determining a kriging proxy model comprises:

13. The system of claim 12, wherein determining a plurality of simulation models based on the set of LHS stratigraphic model and forward depositional model parameters comprises: determining a kriging prediction for locations not included in the forward depositional model.

14. A non-transitory computer-readable medium storing instructions executable by a processing device to perform operations comprising:

receiving a forward modeling deposition model;

performing forward modeling deposition modeling;

determining one or more pseudo-wells based on the transformed model;

determining a mismatch value based on the transformed forward depositional model and the set of simulated physical values;

15. The non-transitory computer-readable medium of claim 14, wherein the forward depositional model is based on well log data describing a drilling path through a predetermined geographic area.

16. The non-transitory computer-readable medium of claim 15, the operations further comprising:

receiving well log data describing a drilling path through the predetermined geographic area; and

17. The non-transitory computer-readable medium of claim 14, the operations further comprising:

drilling a well based on the determined borehole path.

18. The non-transitory computer-readable medium of claim 14, wherein determining a mismatch value based on the transformed forward depositional model and the set of simulated physical values comprises:

providing the set of difference values as the mismatch value.

19. The non-transitory computer-readable medium of claim 14, wherein determining the kriging proxy model comprises:

20. The non-transitory computer-readable medium of claim 19, wherein determining a plurality of simulation models based on the set of LHS stratigraphic model and forward depositional model parameters comprises: determining a kriging prediction for locations not included in the forward depositional model.