CN117010260A - Automatic history fit model prediction method, system and equipment for fractured reservoir - Google Patents
Automatic history fit model prediction method, system and equipment for fractured reservoir Download PDFInfo
- Publication number
- CN117010260A CN117010260A CN202210454076.6A CN202210454076A CN117010260A CN 117010260 A CN117010260 A CN 117010260A CN 202210454076 A CN202210454076 A CN 202210454076A CN 117010260 A CN117010260 A CN 117010260A
- Authority
- CN
- China
- Prior art keywords
- scale
- crack
- fractured reservoir
- fracture
- small
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 72
- 238000005457 optimization Methods 0.000 claims abstract description 51
- 238000009826 distribution Methods 0.000 claims abstract description 39
- 238000004088 simulation Methods 0.000 claims abstract description 34
- 230000006870 function Effects 0.000 claims description 57
- 238000004590 computer program Methods 0.000 claims description 17
- 238000003860 storage Methods 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 238000004519 manufacturing process Methods 0.000 abstract description 5
- 239000003795 chemical substances by application Substances 0.000 description 15
- 238000010586 diagram Methods 0.000 description 8
- 238000011161 development Methods 0.000 description 7
- 239000002245 particle Substances 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- BVKZGUZCCUSVTD-UHFFFAOYSA-L Carbonate Chemical compound [O-]C([O-])=O BVKZGUZCCUSVTD-UHFFFAOYSA-L 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000009977 dual effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000012512 characterization method Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000035699 permeability Effects 0.000 description 2
- 230000002195 synergetic effect Effects 0.000 description 2
- 239000004215 Carbon black (E152) Substances 0.000 description 1
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- -1 carbonate hydrocarbon Chemical class 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 238000013209 evaluation strategy Methods 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 229930195733 hydrocarbon Natural products 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Geometry (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method, a system and equipment for predicting an automatic history fit model of a fractured reservoir, which are used for establishing a numerical simulation model of the fractured reservoir and dividing a fracture network into large-scale fractures and small-scale fractures; parameterizing the large-scale crack and the small-scale crack respectively; establishing a crack network structure by combining large-scale crack parameters and small-scale crack parameters; an improved agent-assisted collaborative population optimization algorithm is established through a fracture network structure, and a radial basis function of an interpolation form is changed into a radial basis function; and carrying out inversion prediction on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm and a data-driven-based evolution algorithm, wherein the obtained optimal solution is an actual fracture distribution parameter. The distribution form of cracks which are difficult to identify in common earthquakes can be predicted according to the production history of the well, and then the distribution state of the residual oil which approximates the actual situation is obtained.
Description
Technical Field
The application belongs to the field of oil and gas reservoir development, and relates to a method, a system and equipment for predicting an automatic history fit model of a fractured reservoir.
Background
Of the oil and gas reserves which have been ascertained worldwide, the reserve occupation scale of carbonate reservoirs is more than half, and the carbonate reservoirs are important reservoir types for oil and gas exploration and development. The carbonate hydrocarbon reservoirs are widely distributed in China, and have great potential for exploration and development. However, the carbonate reservoir has strong heterogeneity and complex geological conditions, and besides, the crack scale is often multi-scale, so that the development effect of the oil reservoir is greatly affected. The development condition of a carbonate reservoir fracture-cavity system cannot be described precisely by the existing geophysical prospecting technology, and the distribution form of cracks is extremely complex and has extremely large uncertainty, so that the prediction technology of earthquake cracks is required to be researched, the knowledge of the seepage rule of the reservoir can be improved, and further accurate residual oil distribution is obtained.
In the oil reservoir fracture prediction method, the history fitting method can be used for realizing inversion of a geological model, and the history fitting process is a method for continuously adjusting model parameters according to history production data so as to further determine actual model parameters. However, the manual history fitting is greatly influenced by human factors, is long in time consumption and has more limitations. The automatic history fitting is a method for solving the inverse problem by combining a computer technology, and is quicker and convenient to operate compared with manual history fitting. By automatic history fitting, uncertainty in fracture network distribution can be reduced.
The history fit fractured reservoir is first to build a reservoir numerical simulation model that simulates the flow of fluid in the fracture network. Conventional methods for modeling fracture network models include a dual media model, an equivalent continuous media model, a discrete fracture network model (DFM), and an embedded discrete fracture network model (EDFM). In the dual media model, the fracture and matrix are achieved by defining two types of permeability and porosity. The equivalent continuous medium model, matrix and fracture are characterized by the continuous medium alone, defining equivalent permeability porosity. The dual medium model and the equivalent continuous medium model do not need to separate grids, so the numerical simulation process is simple and stable, but the two methods have difficulty in clearly describing the influence of cracks on seepage. Discrete fracture network models (DFMs) typically use unstructured structuring to explicitly describe the location of the fracture, resulting in computational expense. While the embedded discrete fracture network model (EDFM) uses structured grids, the computational cost is reduced, however, the accuracy of simulating the fracture is poorer than that of the discrete fracture network model.
The ensemble-based method is widely used in reservoir history fitting, such as the Markov chain Monte Carlo Method (MCMC) to estimate model parameter distribution characteristics by sampling the posterior probability function of a history-fit objective function. In addition, genetic Algorithm (GA), particle Swarm Optimization (PSO) and other algorithms are adopted, and the most probable crack distribution situation is found out by solving an objective function. Unfortunately, fractured reservoirs often have a large number of parameters, and the resulting high dimensional problems have not been effectively resolved.
In summary, in the process of solving the history fit of the fractured reservoir, the optimization algorithm capable of adapting to the high-dimensional problem is relatively lacking.
Disclosure of Invention
The application aims to overcome the defects of the prior art and provide a method, a system and equipment for predicting an automatic history fit model of a fractured reservoir, which can predict the distribution form of cracks which are difficult to identify by common earthquakes according to the production history of a well, so as to obtain the distribution state of residual oil similar to the actual situation.
In order to achieve the purpose, the application is realized by adopting the following technical scheme:
an automatic history fit model prediction method for a fractured reservoir comprises the following steps:
step one, a numerical simulation model of a fractured reservoir is established, and a fracture network in the numerical simulation model of the fractured reservoir is divided into large-scale fractures and small-scale fractures;
step two, parameterizing the large-scale cracks and the small-scale cracks respectively;
step three, a crack network structure is established by combining parameters of the large-scale cracks and the small-scale cracks;
establishing an improved proxy auxiliary cooperative group optimization algorithm through a crack network structure, wherein in the improved proxy auxiliary cooperative group optimization algorithm, a radial basis function of a radial basis neural network is changed into a radial basis function of an interpolation form;
and fifthly, performing inversion prediction on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm and a data-driven-based evolution algorithm, wherein the obtained optimal solution is an actual fracture distribution parameter.
Preferably, the characteristic parameters of the fracture in the numerical simulation model of the fractured reservoir are fracture length, azimuth and midpoint coordinates.
Preferably, the large-scale fracture parameterization process is as follows: and randomly generating large-scale crack parameters through the ranges of the crack length, the azimuth and the midpoint coordinates, calculating the opening degree of the crack based on the cube law, and then combining the large-scale crack parameters.
Preferably, the small-scale fracture parameterization process is as follows: setting the maximum and minimum lengths of the cracks and the fractal dimension, calculating to obtain the number of all the cracks, generating a small-scale crack network through the relation between the lengths and the number of the cracks, dividing the small-scale cracks into data sets, and combining small-scale crack parameters.
Preferably, in the third step, the parameters of the large-scale cracks and the small-scale cracks are combined to form a crack network structure, and numerical simulation is performed by using an embedded discrete crack network model according to the parameters of the crack network structure.
Preferably, the radial basis function in interpolated form is expressed as:
wherein f (x) is a radial basis function in an interpolation form; n is the number of hidden nodes; omega k Is a weight coefficient; phi is a basis function; x is x k Is the center point.
Preferably, the specific process of the fifth step is as follows: performing history fitting on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm, setting a history-fitted objective function, and solving the objective function by using the improved agent-assisted collaborative population optimization algorithm to obtain an optimal solution which is an actual fracture distribution parameter.
An automated history-fit model prediction system for a fractured reservoir, comprising:
the simulation model building module is used for building a numerical simulation model of the fractured reservoir and dividing a fracture network in the numerical simulation model of the fractured reservoir into large-scale fractures and small-scale fractures;
the parameterization module is used for parameterizing the large-scale crack and the small-scale crack respectively;
the crack network structure building module is used for building a crack network structure by combining large-scale crack parameters and small-scale crack parameters;
the optimization algorithm building module is used for building an improved proxy auxiliary cooperation group optimization algorithm through a crack network structure, wherein in the improved proxy auxiliary cooperation group optimization algorithm, a radial basis function of a radial basis neural network is changed into a radial basis function of an interpolation form;
the actual fracture distribution parameter acquisition module is used for carrying out inversion prediction on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm and a data-driven-based evolution algorithm, and the obtained optimal solution is the actual fracture distribution parameter.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the automated history-fit model prediction method for a fractured reservoir as claimed in any one of the preceding claims when the computer program is executed.
A computer readable storage medium storing a computer program which when executed by a processor performs the steps of the automated history-fit model prediction method for a fractured reservoir of any one of the above.
Compared with the prior art, the application has the following beneficial effects:
according to the application, the crack network is divided into large-scale cracks and small-scale cracks, the parameter scale used for representing the crack distribution is greatly reduced, the fitting difficulty is reduced, an improved agent-assisted collaborative population optimization algorithm is adopted, the radial basis function of the radial basis neural network is changed into the radial basis function in an interpolation form, the iteration speed of the evolutionary algorithm can be accelerated, and the evolutionary speed of the evolutionary algorithm in the optimization process is improved, so that the calculation time cost required by automatic history fitting is greatly reduced, the accuracy of predicting the distribution form of the crack can be improved by the whole method, and the residual oil distribution state similar to the real situation is obtained.
Drawings
FIG. 1 is a diagram of an original fracture distribution reference model of the present application;
FIG. 2 is an initialized fracture distribution profile of the present application;
FIG. 3 is a schematic diagram of an optimized fracture network model of the present application;
FIG. 4 is a schematic representation of the optimized model residual oil saturation of the present application;
FIG. 5 is a schematic representation of residual oil saturation for a reference model of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the application; all other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It should be noted that the words "front", "rear", "left", "right", "upper" and "lower" used in the following description refer to directions in the drawings, and the words "inner" and "outer" refer to directions toward or away from, respectively, the geometric center of a particular component.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the description of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
The automatic history fit model prediction method for the fractured reservoir comprises the following steps:
step one, a numerical simulation model of the fractured reservoir is established, and a fracture network in the numerical simulation model of the fractured reservoir is divided into large-scale fractures and small-scale fractures.
The characterization parameters of the fracture in the numerical simulation model of the fractured reservoir are fracture length, azimuth and midpoint coordinates.
And step two, respectively parameterizing the large-scale crack and the small-scale crack.
The large-scale crack parameterization process comprises the following steps: and randomly generating large-scale crack parameters through the ranges of the crack length, the azimuth and the midpoint coordinates, calculating the opening degree of the crack based on the cube law, and then combining the large-scale crack parameters.
The small-scale crack parameterization process comprises the following steps: setting the maximum and minimum lengths of the cracks and the fractal dimension, calculating to obtain the number of all the cracks, generating a small-scale crack network through the relation between the lengths and the number of the cracks, dividing the small-scale cracks into data sets, and combining small-scale crack parameters.
And thirdly, establishing a crack network structure by combining the parameters of the large-scale cracks and the small-scale cracks.
And combining large-scale crack parameters and small-scale crack parameters to form a crack network structure, and performing numerical simulation by using an embedded discrete crack network model according to the parameters of the crack network structure.
And step four, an improved proxy auxiliary cooperative group optimization algorithm is established through a crack network structure, and in the improved proxy auxiliary cooperative group optimization algorithm, a radial basis function of the radial basis neural network is changed into a radial basis function of an interpolation form.
The radial basis function in interpolated form is expressed as:
wherein f (x) is a radial basis function in an interpolation form; n is the number of hidden nodes; omega k Is a weight coefficient; phi is a basis function; x is x k Is the center point.
And fifthly, performing inversion prediction on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm and a data-driven-based evolution algorithm, wherein the obtained optimal solution is an actual fracture distribution parameter.
Performing history fitting on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm, setting a history-fitted objective function, and solving the objective function by using the improved agent-assisted collaborative population optimization algorithm to obtain an optimal solution which is an actual fracture distribution parameter.
The specific process of the scheme is as follows:
step 1: and establishing a numerical simulation model of the fractured reservoir.
The establishment of the numerical simulation model of the fractured reservoir is based on an embedded discrete fracture network (EDFM) and is realized by using an MRST kit in MATLAB software. The characterization parameters of the crack are the length, azimuth and midpoint coordinates of the crack. The fracture network can be divided into two parts, namely a large-scale fracture and a small-scale fracture, and the small-scale fracture can be changed into a medium-scale fracture and a large-scale fracture according to different setting parameters.
Step 2: parameterizing large-scale cracks
For the large-scale crack parameterization method, the large-scale crack parameters can be generated randomly through the ranges of the crack length, the azimuth and the midpoint coordinates, and the opening degree of the crack is calculated based on the cube law, wherein the calculation formula is as follows:
e＝βl
where e is the opening of the crack, β is a constant, and l is the length of the crack.
Combining the large-scale fracture parameters, which can be expressed as:
wherein m is l Is a large-scale crack parameter set, l i Length of ith crack, x i ，y i For the ith crack midpoint coordinate, θ i For the direction of the ith crack, N ma Is the number of large cracks.
Step 3: and parameterizing the small-scale crack based on a fractal theory.
Firstly, setting the maximum and minimum lengths of the cracks and the fractal dimension, and calculating the number of all the cracks. The parameterization method of the small-scale cracks is based on a fractal theory, a small-scale crack network is generated through the relation between the lengths and the numbers of the cracks, and the relation between the lengths and the numbers of the cracks in the fractal theory is as follows:
wherein N is to For the total number of cracks, l max Length of longest crack, l min For the length of the shortest crack, D l Is the fractal dimension.
Then, a random number R is set, and the length of each crack can be obtained. For a specific length of each crack, the generation method is as follows:
wherein R is a random number between 0 and 1.
Further, the small-scale fracture is divided into data sets. Different data sets have different crack generation parameters and different properties, so that the multi-scale crack network is characterized. The intensity D of the cracks in the data set can be calculated according to the number of the cracks in different data sets s The calculation formula is as follows:
wherein the method comprises the steps ofN is the number of cracks in the crack set i mi N for the number of all small cracks se To divide the number of data sets.
Finally, the angle for each small crack is generated by a Gaussian distribution function with an average value ofVariance is sigma i The small-scale fracture parameters can be combined according to actual custom settings.
Step 4: and combining the large-scale cracks and the small-scale cracks to establish a crack network structure.
Combining large-scale crack and small-scale crack parameters to form a crack network structure, and expressing the parameters of the crack network structure as:
m＝{m 1 ，m 2 }
numerical simulations were performed using an embedded discrete fracture network model (EDFM) according to parameters of the fracture network.
Step 5: an improved proxy assisted collaborative population optimization algorithm is established.
The agent assisted collaborative population optimization algorithm (SACOSO) is an evolution algorithm based on a data driver, the algorithm based on the data driver can accelerate convergence, the calculation cost is reduced, and the SACOSO algorithm is suitable for the optimization problem of 50-100 dimensions. The traditional SACOSO algorithm solves the objective function through the synergistic effect of two particle swarm optimization algorithms, one is a particle swarm optimization algorithm based on an adaptive evaluation strategy, and the other is a particle swarm optimization algorithm based on a radial basis function. The synergistic effect of the two algorithms can balance the search effect of the evolutionary algorithm.
The specific process is as follows:
1) Combining a particle swarm optimization algorithm with fitness value evaluation, wherein the fitness value evaluation needs to combine the numerical simulation in the step 4, update iteration and find an optimal solution;
2) The data in the first step of algorithm searching process is formed into a sample set, and a proxy model of a radial basis function is trained;
3) Combining a particle swarm optimization algorithm with a radial basis function proxy model evaluation, updating iteration to find an optimal solution, and guiding the searching process in the first step by the optimal solution;
4) Repeating the above 1-3 processes until the algorithm converges.
The improved agent assisted collaborative population optimization algorithm (moSACOSO) is based on the SACOSO algorithm, the radial basis function (RBF-IN) IN an interpolation form is changed from a radial basis function (RBF-NN), the radial basis function (RBF-IN) IN the interpolation form is faster IN modeling process, and meanwhile, more accurate effects can be obtained. The radial basis function (RBF-IN) IN interpolated form can be expressed as:
wherein f (x) is a radial basis function in an interpolation form; n is the number of hidden nodes; omega k Is a weight coefficient; phi is a basis function; x is x k Is the center point. The basis function phi may be selected as:
φ(r)＝r 3
weight vector ω= (ω) 1 ,...,ω N ) T Can be expressed as:
ω＝(Φ T Φ) -1 Φ T y
wherein y= (y) 1 ,...,y N ) T 。
Step 6: inversion prediction is performed on the fractured reservoir by using a data-driven evolution algorithm.
An automated history fit is performed on the fractured reservoir model using a modified agent assisted collaborative population optimization algorithm (moSACOSO).
First, a history-fit objective function is set. The relation between the model parameters and the historical production data is as follows:
d＝g(m)+ε
where d is the observed data, m is the model parameters, g (m) is the numerical simulation data, and ε is the error.
According to bayesian theory, the posterior probability can be represented by a priori probability and likelihood function, and in the reservoir automatic history fitting problem, the posterior probability density function p (d|m) can be represented as:
wherein C is d For observed data error covariance, C is a constant.
Thus, the objective function of the history fit is set to:
then, the objective function is solved by using an improved agent-assisted collaborative population optimization algorithm (moSACOSO), and the obtained optimal solution is the actual fracture distribution parameter.
The original crack distribution is shown in fig. 1, which contains 6 large cracks and a plurality of small cracks.
The improved method results are shown in fig. 2 and 3, fig. 2 is an initialized crack distribution characteristic, the characteristic parameters of the crack are optimized through an evolutionary algorithm, and the finally obtained optimal crack inversion result is shown in fig. 3, wherein the distribution characteristic and the position of the large crack can be relatively matched with the actual crack position in fig. 1, and the accuracy is high.
On the basis of the fracture model, the residual oil saturation is also predicted, as shown in fig. 4 and 5, fig. 4 is a prediction result, and fig. 5 is a saturation distribution result of the actual model. From comparison of the two graphs, the residual oil distribution state of the predicted result is basically consistent with the state of an actual oil reservoir.
Application range and application prospect.
The application range is as follows: the method is suitable for the fractured reservoir, and can approximate fit the underground fracture distribution condition. The method is particularly suitable for well group models with smaller ranges.
Application prospect: carbonate reservoir fractures are the primary oil transportation channels, so subsurface fracture prediction is also an important direction for research in current development. The method can approximate the distribution state of underground cracks, can improve the knowledge of technicians on the underground seepage characteristics of oil reservoirs, and can know the distribution characteristics of residual oil, and has important significance for subsequent well position deployment and development technical policy adjustment. Meanwhile, the method can be suitable for various fractured reservoirs and has a larger application space.
The following are device embodiments of the present application that may be used to perform method embodiments of the present application. For details of the device embodiment that are not careless, please refer to the method embodiment of the present application.
In still another embodiment of the present application, an automatic history fit model prediction for a fractured reservoir is provided, and the automatic history fit model prediction for a fractured reservoir may be used to implement the automatic history fit model prediction for a fractured reservoir, and the method specifically includes a simulation model establishment module, a parameterization module, a fracture network structure establishment module, an optimization algorithm establishment module, and an actual fracture distribution parameter acquisition module.
The simulation model building module is used for building a numerical simulation model of the fractured reservoir and dividing a fracture network in the numerical simulation model of the fractured reservoir into large-scale fractures and small-scale fractures.
And the parameterization module is used for parameterizing the large-scale crack and the small-scale crack respectively.
And the crack network structure building module is used for building a crack network structure by combining large-scale crack parameters and small-scale crack parameters.
The optimization algorithm building module is used for building an improved proxy auxiliary cooperation group optimization algorithm through the fracture network structure, and in the improved proxy auxiliary cooperation group optimization algorithm, the radial basis function of the radial basis neural network is changed into the radial basis function of the interpolation form.
The actual fracture distribution parameter acquisition module is used for carrying out inversion prediction on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm and a data-driven-based evolution algorithm, and the obtained optimal solution is the actual fracture distribution parameter.
In yet another embodiment of the present application, a terminal device is provided, the terminal device including a processor and a memory, the memory for storing a computer program, the computer program including program instructions, the processor for executing the program instructions stored by the computer storage medium. The processor may be a central processing unit (Central Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf Programmable gate arrays (FPGAs) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., which are the computational core and control core of the terminal adapted to implement one or more instructions, in particular adapted to load and execute one or more instructions to implement a corresponding method flow or a corresponding function; the processor of the embodiment of the application can be used for automatic history fit model prediction of the fractured reservoir, and the method comprises the following operations:
step one, a numerical simulation model of the fractured reservoir is established, and a fracture network in the numerical simulation model of the fractured reservoir is divided into large-scale fractures and small-scale fractures.
And step two, respectively parameterizing the large-scale crack and the small-scale crack.
And thirdly, establishing a crack network structure by combining the parameters of the large-scale cracks and the small-scale cracks.
And step four, an improved proxy auxiliary cooperative group optimization algorithm is established through a crack network structure, and in the improved proxy auxiliary cooperative group optimization algorithm, a radial basis function of the radial basis neural network is changed into a radial basis function of an interpolation form.
And fifthly, performing inversion prediction on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm and a data-driven-based evolution algorithm, wherein the obtained optimal solution is an actual fracture distribution parameter.
In still another embodiment, the present application also provides a computer-readable storage medium (Memory) that is a Memory device in a terminal device for storing programs and data. It will be appreciated that the computer readable storage medium herein may include both a built-in storage medium in the terminal device and an extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space storing an operating system of the terminal. Also stored in the memory space are one or more instructions, which may be one or more computer programs (including program code), adapted to be loaded and executed by the processor. The computer readable storage medium herein may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory.
One or more instructions stored in a computer-readable storage medium may be loaded and executed by a processor to implement the respective steps of the method in relation to automated history-fit model prediction of a fractured reservoir in the above-described embodiments; one or more instructions in a computer-readable storage medium are loaded by a processor and perform the steps of:
step one, a numerical simulation model of the fractured reservoir is established, and a fracture network in the numerical simulation model of the fractured reservoir is divided into large-scale fractures and small-scale fractures.
And step two, respectively parameterizing the large-scale crack and the small-scale crack.
And thirdly, establishing a crack network structure by combining the parameters of the large-scale cracks and the small-scale cracks.
And step four, an improved proxy auxiliary cooperative group optimization algorithm is established through a crack network structure, and in the improved proxy auxiliary cooperative group optimization algorithm, a radial basis function of the radial basis neural network is changed into a radial basis function of an interpolation form.
And fifthly, performing inversion prediction on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm and a data-driven-based evolution algorithm, wherein the obtained optimal solution is an actual fracture distribution parameter.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
It is to be understood that the above description is intended to be illustrative, and not restrictive. Many embodiments and many applications other than the examples provided will be apparent to those of skill in the art upon reading the above description. The scope of the present teachings should, therefore, be determined not with reference to the above description, but instead should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled. The disclosures of all articles and references, including patent applications and publications, are incorporated herein by reference for the purpose of completeness. The omission of any aspect of the subject matter disclosed herein in the preceding claims is not intended to forego such subject matter, nor should the applicant be deemed to have such subject matter not considered to be part of the disclosed subject matter.
Claims (10)
1. The automatic history fit model prediction method for the fractured reservoir is characterized by comprising the following steps of:
step one, a numerical simulation model of a fractured reservoir is established, and a fracture network in the numerical simulation model of the fractured reservoir is divided into large-scale fractures and small-scale fractures;
step two, parameterizing the large-scale cracks and the small-scale cracks respectively;
step three, a crack network structure is established by combining parameters of the large-scale cracks and the small-scale cracks;
establishing an improved proxy auxiliary cooperative group optimization algorithm through a crack network structure, wherein in the improved proxy auxiliary cooperative group optimization algorithm, a radial basis function of a radial basis neural network is changed into a radial basis function of an interpolation form;
and fifthly, performing inversion prediction on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm and a data-driven-based evolution algorithm, wherein the obtained optimal solution is an actual fracture distribution parameter.
2. The method for predicting an automatic history fit model for a fractured reservoir according to claim 1, wherein the characteristic parameters of the fracture in the numerical simulation model for the fractured reservoir are fracture length, azimuth and midpoint coordinates.
3. The method for predicting an automatic history fit model for a fractured reservoir according to claim 1, wherein the process of parameterizing the large-scale fracture is: and randomly generating large-scale crack parameters through the ranges of the crack length, the azimuth and the midpoint coordinates, calculating the opening degree of the crack based on the cube law, and then combining the large-scale crack parameters.
4. The method for predicting an automatic history fit model for a fractured reservoir according to claim 1, wherein the process of small-scale fracture parameterization is as follows: setting the maximum and minimum lengths of the cracks and the fractal dimension, calculating to obtain the number of all the cracks, generating a small-scale crack network through the relation between the lengths and the number of the cracks, dividing the small-scale cracks into data sets, and combining small-scale crack parameters.
5. The method for predicting an automatic history fit model for a fractured reservoir according to claim 1, wherein in the third step, a fracture network structure is formed by combining parameters of large-scale fractures and small-scale fractures, and numerical simulation is performed by using an embedded discrete fracture network model according to the parameters of the fracture network structure.
6. The method for predicting an automatic history-fit model for a fractured reservoir according to claim 1, wherein the radial basis function in the form of interpolation is expressed as:
wherein f (x) is a radial basis function in an interpolation form; n is the number of hidden nodes; omega k Is a weight coefficient; phi is a basis function; x is x k Is the center point.
7. The method for predicting an automatic history fit model for a fractured reservoir according to claim 1, wherein the specific process of the fifth step is as follows: performing history fitting on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm, setting a history-fitted objective function, and solving the objective function by using the improved agent-assisted collaborative population optimization algorithm to obtain an optimal solution which is an actual fracture distribution parameter.
8. An automated history-fit model prediction system for a fractured reservoir, comprising:
the simulation model building module is used for building a numerical simulation model of the fractured reservoir and dividing a fracture network in the numerical simulation model of the fractured reservoir into large-scale fractures and small-scale fractures;
the parameterization module is used for parameterizing the large-scale crack and the small-scale crack respectively;
the crack network structure building module is used for building a crack network structure by combining large-scale crack parameters and small-scale crack parameters;
the optimization algorithm building module is used for building an improved proxy auxiliary cooperation group optimization algorithm through a crack network structure, wherein in the improved proxy auxiliary cooperation group optimization algorithm, a radial basis function of a radial basis neural network is changed into a radial basis function of an interpolation form;
the actual fracture distribution parameter acquisition module is used for carrying out inversion prediction on the fractured reservoir by using an improved agent-assisted collaborative population optimization algorithm and a data-driven-based evolution algorithm, and the obtained optimal solution is the actual fracture distribution parameter.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, performs the steps of the method for automated history-fit model prediction for a fractured reservoir according to any one of claims 1 to 7.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of the automated history-fit model prediction method for a fractured reservoir of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210454076.6A CN117010260A (en) | 2022-04-27 | 2022-04-27 | Automatic history fit model prediction method, system and equipment for fractured reservoir |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210454076.6A CN117010260A (en) | 2022-04-27 | 2022-04-27 | Automatic history fit model prediction method, system and equipment for fractured reservoir |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117010260A true CN117010260A (en) | 2023-11-07 |
Family
ID=88571357
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210454076.6A Pending CN117010260A (en) | 2022-04-27 | 2022-04-27 | Automatic history fit model prediction method, system and equipment for fractured reservoir |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117010260A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117421939B (en) * | 2023-12-18 | 2024-03-19 | 中国石油大学(华东) | Shale oil fracture system simulation agent method based on track piecewise linearization |
-
2022
- 2022-04-27 CN CN202210454076.6A patent/CN117010260A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117421939B (en) * | 2023-12-18 | 2024-03-19 | 中国石油大学(华东) | Shale oil fracture system simulation agent method based on track piecewise linearization |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103440361B (en) | The modeling method of yield is etched in a kind of plasma etch process | |
| CN111465944A (en) | Graphical neural network system for generating a structured representation of an object | |
| Karssenberg et al. | Conditioning a process-based model of sedimentary architecture to well data | |
| WO2016074202A1 (en) | Method and device for simulating particle etching or depositional evolution, and computer readable medium | |
| CN111058840A (en) | Organic carbon content (TOC) evaluation method based on high-order neural network | |
| CN112927769A (en) | MOFs material defect structure prediction method based on pore size distribution curve | |
| CN117010260A (en) | Automatic history fit model prediction method, system and equipment for fractured reservoir | |
| CN104732067A (en) | Industrial process modeling forecasting method oriented at flow object | |
| US11513255B2 (en) | Method of characterising a subsurface volume | |
| CN116415482A (en) | Wing flow field analysis method based on graph neural network MeshGraphNet | |
| Ding et al. | Dynamic evolutionary model based on a multi-sampling inherited hapfnn for an aluminium electrolysis manufacturing system | |
| CN114372332A (en) | Multi-target robust design method based on subset simulation | |
| CN114861519A (en) | Initial ground stress field acceleration optimization inversion method under complex geological condition | |
| Couckuyt et al. | Towards efficient multiobjective optimization: multiobjective statistical criterions | |
| CN113468466A (en) | Multi-working-condition one-dimensional wave equation solving method based on neural network | |
| CN116993548A (en) | Incremental learning-based education training institution credit assessment method and system for LightGBM-SVM | |
| Pehlivanoglu | Direct and indirect design prediction in genetic algorithm for inverse design problems | |
| CN113240094A (en) | SVM-based LSTM hyper-parameter optimization method, system, medium and device | |
| CN112016956B (en) | Ore grade estimation method and device based on BP neural network | |
| CN112649869A (en) | Reservoir characteristic parameter prediction method and system based on GA-WNN | |
| CN118261022A (en) | Multi-target fracturing fracture parameter prediction method, system, equipment and medium | |
| CN113705878B (en) | Method and device for determining water yield of horizontal well, computer equipment and storage medium | |
| CN118036428A (en) | Modeling and predicting method of integrated proxy model based on Kriging interpolation | |
| CN117313905A (en) | Tracing algorithm based on improved arithmetic optimization algorithm | |
| Bogoclu et al. | Reliability analysis of non-linear and multimodal limit state functions using adaptive Kriging |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |