Shale oil reservoir development simulation method and device based on machine learning
By optimizing shale reservoir development simulation using the finite volume method and time step prediction model based on machine learning, the problems of multi-scale seepage and multi-phase and multi-component seepage were solved, achieving efficient and stable shale reservoir development simulation.
Patent Information
- Application Number
- CN202410312102.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-19
- Publication Date
- 2025-09-19
AI Technical Summary
Existing technologies in shale reservoir development have multi-scale seepage characteristics and multi-phase and multi-component seepage problems, which make numerical simulation complex and time-consuming, making it difficult to achieve efficient shale reservoir development.
A machine learning-based method is used to perform initial simulation using the finite volume method, train time step prediction models for the water, oil, and gas phases, use the maximum residual value as input, optimize the simulation time step, and combine subsequent simulations with machine learning algorithms to improve simulation speed and accuracy.
It accelerates the simulation process of shale oil reservoir development, improves computing efficiency, stability and simulation accuracy, reduces numerical simulation oscillation, and simplifies the operation process.
Smart Images
Figure CN120671485A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of shale oil reservoir development, and in particular to a shale oil reservoir development simulation method and device based on machine learning. Background Art
[0002] To achieve efficient exploitation of shale reservoirs, it is necessary to understand their development mechanisms. Currently, the primary method for shale reservoir development is to utilize quasi-natural energy after hydraulic fracturing. After hydraulic fracturing, reservoirs exhibit typical multiscale flow characteristics, meaning that permeability varies at different scales, posing challenges for reservoir development and simulation. Furthermore, shale reservoir development involves multiphase and multicomponent flow. Gas and liquid phases coexist within the reservoir, and these phases contain multiple components, such as different hydrocarbons. This multiphase and multicomponent flow makes numerical simulations complex and time-consuming. Numerical simulations are relatively slow due to the dual complexity of the fluid and matrix in the reservoir. The interplay of factors within the reservoir, such as the fracture network, pore structure, and rock mechanical properties, creates challenges in numerical simulation. Therefore, to better understand and optimize shale reservoir development, sophisticated numerical simulations are necessary to improve development efficiency and maximize the utilization of this abundant energy resource. Summary of the Invention
[0003] In order to obtain more efficient shale oil reservoir development simulation results, an embodiment of the present invention provides a shale oil reservoir development simulation method and device based on machine learning.
[0004] In a first aspect, an embodiment of the present invention provides a shale oil reservoir development simulation method based on machine learning, the method comprising:
[0005] Obtaining the initial geological parameters, initial simulation time step, and preset maximum residual value combination of the target shale oil reservoir;
[0006] Based on the initial geological parameters, the initial simulation time step, and the preset maximum residual value, a finite volume method is used to perform a shale reservoir development simulation until a preset first simulation stop condition is reached, thereby obtaining a step parameter data set and new geological parameters;
[0007] Based on the step parameter data set, a water phase time step prediction model, an oil phase time step prediction model, and a gas phase time step prediction model are trained;
[0008] Combining the preset maximum residual value combination with the new geological parameters, the results are input into the oil phase time step prediction model, the water phase time step prediction model, and the gas phase time step prediction model, respectively, and obtaining a new simulation time step based on the corresponding oil phase time step, water phase time step, and gas phase time step;
[0009] Performing shale oil reservoir development simulation based on the new geological parameters and the new simulation time step to obtain updated geological parameters;
[0010] Repeat the steps of obtaining an updated simulation time step according to the updated geological parameters, performing a shale reservoir development simulation according to the updated simulation time step, and obtaining new updated geological parameters until a preset second simulation stop condition is reached, thereby obtaining a shale reservoir development simulation result.
[0011] In one or some optional implementations of the embodiment of the present application, the preset maximum residual value combination includes: maximum oil phase residual, maximum water phase residual, maximum gas phase residual;
[0012] The combination of the preset maximum residual value and the new geological parameters is input into the water phase time step prediction model, the oil phase time step prediction model, and the gas phase time step prediction model, respectively, and a new simulation time step is obtained according to the corresponding oil phase time step, water phase time step, and gas phase time step, including:
[0013] The maximum oil phase residual, the maximum water phase residual, and the maximum gas phase residual are combined with the new geological parameters and input into the oil phase time step prediction model, the water phase time step prediction model, and the gas phase time step prediction model, respectively. Based on the following formula, the oil phase time step, the water phase time step, and the gas phase time step are obtained:
[0014]
[0015] Where Δt 1max , Δt 2max , Δt 3max represents the oil phase time step, water-gas phase time step and gas phase time step respectively, f1, f2 and f3 represent the oil phase time step prediction model, water phase time step prediction model and gas phase time step prediction model respectively, φ represents porosity, k is absolute permeability, S o 、S w 、S g Represent the saturation of oil phase, water phase and gas phase respectively, p o 、p w 、p g Represent the pressure values of oil phase, water phase and gas phase respectively, q O ,q W ,q G are the underground mass sources and sinks of oil, water and gas components respectively, R omax 、R wmax 、R gmax They represent the maximum oil phase residual, the maximum water phase residual, and the maximum gas phase residual, respectively;
[0016] The minimum value among the oil phase time step, the water phase time step and the gas phase time step is taken as the new simulation time step.
[0017] One or some optional implementations of the embodiments of the present application may further include:
[0018] In the process of obtaining an updated simulation time step according to the updated geological parameters and performing a shale reservoir development simulation according to the updated simulation time step to obtain new updated geological parameters, calculating a later cumulative time length of the shale reservoir development simulation after training the water phase time step prediction model, the oil phase time step prediction model, and the gas phase time step prediction model;
[0019] If the later accumulated time length reaches a multiple of the first preset time length, all simulation time steps and their corresponding geological parameters obtained in the shale oil reservoir development simulation are obtained to obtain an updated step parameter data set;
[0020] A new water phase time step prediction model, a new oil phase time step prediction model and a new gas phase time step prediction model are trained based on the updated step parameter data set.
[0021] In one or some optional implementations of the embodiment of the present application, the finite volume method is used to perform shale reservoir development simulation based on the initial geological parameters, the initial simulation time step, and the preset maximum residual value combination until a preset first simulation stop condition is reached, thereby obtaining a step parameter data set and new geological parameters, including:
[0022] Based on the initial geological parameters and the initial simulation time step, a residual equation of the three-component numerical simulation model is established; the residual equation of the three-component numerical simulation model includes a residual equation of the oil component mass conservation equation, a residual equation of the water component mass conservation equation, and a residual equation of the gas component mass conservation equation;
[0023] Based on the initial geological parameters and the initial simulation time step, the residual equation of the three-component numerical simulation model is solved using the finite volume method. When the residuals of the oil component mass conservation equation within the initial simulation time step are all less than the maximum oil phase residual, the residuals of the water component mass conservation equation are all less than the maximum water phase residual, and the residuals of the gas component mass conservation equation are all less than the maximum gas phase residual, the first geological parameters are obtained, and the first time step is obtained based on the following formula:
[0024]
[0025] Where, Δt new represents the first time step, Δt oldrepresents the initial simulation time step, fac1 and fac2 represent the time increase factor and time decrease factor respectively, dp ph is the maximum pressure change value within the initial simulation time step, dp phmax Increase the maximum allowable pressure change value for the initial simulation time step, dS ph is the maximum saturation change value within the initial simulation time step, dS phmax The maximum saturation change value allowed for increasing the initial simulation time step, ph represents the category, and the possible values are o, w, and g, where o, w, and g represent oil, water, and gas, respectively;
[0026] The above process of simulating shale reservoir development based on the first geological parameter and the first time step is repeatedly executed until a preset first simulation stop condition is reached, thereby obtaining the step parameter data set and the new geological parameters.
[0027] In one or some optional implementations of the embodiment of the present application, the preset first simulation stop condition is that the cumulative time length of the shale oil reservoir development simulation reaches a first preset time length;
[0028] Whether the progress of shale oil reservoir development simulation reaches the preset first simulation stop condition is determined by the following method:
[0029] determining a first cumulative time length for shale reservoir development simulation based on the initial simulation time step and all simulation time steps obtained in the iterative solution process, and determining whether the first cumulative time length reaches a first preset time length;
[0030] If yes, obtain all simulation time steps and their corresponding geological parameters obtained in the iterative solution process, construct the step parameter data set, and use the geological parameters obtained in the last iteration as the new geological parameters;
[0031] If not, the above steps of performing shale oil reservoir development simulation based on the first geological parameter and the first time step and iteratively solving the problem are repeated.
[0032] In one or some optional implementations of the embodiment of the present application, the preset second simulation stop condition is that the cumulative time length of the shale oil reservoir development simulation reaches a second preset time length; the second preset time length is greater than the first preset time length;
[0033] Whether the progress of shale oil reservoir development simulation reaches the preset second simulation stop condition is determined by the following method:
[0034] determining a second cumulative time length of the shale reservoir development simulation based on the initial simulation time step and all simulation time steps obtained during the shale reservoir development simulation, and determining whether the second cumulative time length reaches a second preset time length;
[0035] If yes, obtaining the shale oil reservoir development simulation result based on all simulation time steps and their corresponding geological parameters obtained during the shale oil reservoir development simulation process;
[0036] If not, the above steps of obtaining an updated simulation time step according to the updated geological parameters are re-executed, and shale oil reservoir development simulation is performed according to the updated simulation time step to obtain new updated geological parameters.
[0037] In one or some optional implementations of the embodiment of the present application, establishing the residual equation of the three-component numerical simulation model based on the initial geological parameters and the initial simulation time step includes:
[0038] Based on the initial geological parameters and the initial simulation time step, the residual equation of the oil component mass conservation equation is established based on the following formula:
[0039]
[0040] Where R o is the residual of the oil component mass conservation equation, φ represents the porosity, ρ Oo is the density of the oil component in the oil phase, S o is the oil phase saturation, t is the initial simulation time step, u o is the oil phase flow rate, q O represents the underground mass source and sink of oil components, o represents oil, is the divergence operator, is the partial derivative operator;
[0041] Based on the initial geological parameters and the initial simulation time step, the residual equation of the water component mass conservation equation is established based on the following formula:
[0042]
[0043] Where R w is the residual of the water component mass conservation equation, φ represents the porosity, ρ w is the density of the water phase, S w is the water phase saturation, t is the initial simulation time step, u w is the water phase flow rate, q W represents the underground mass source and sink of water components, w represents water, is the divergence operator, is the partial derivative operator;
[0044] Based on the initial geological parameters and the initial simulation time step, the residual equation of the gas component mass conservation equation is established based on the following formula:
[0045]
[0046] Where R g is the residual of the gas component mass conservation equation, φ represents the porosity, ρ Go is the density of the gas component in the oil phase, ρ G and ρ g is the gas phase density, S o is the oil phase saturation, S G is the gas phase saturation, t is the initial simulation time step, u o is the oil phase flow rate, u g is the gas phase flow rate, q G represents the underground mass source and sink of gas components, g represents gas, is the divergence operator, is the partial derivative operator.
[0047] In one or some optional implementations of the embodiments of the present application, the flow rates of the oil phase, water phase, and gas phase are calculated based on the following formulas:
[0048]
[0049] Where u ph is the flow rate of the pH phase, k is the absolute permeability, k rph is the relative permeability of the pH phase, μ ph is the viscosity of the pH phase, p ph is the pressure of the ph phase, ρ ph is the density of the ph phase, g is the acceleration of gravity; D is the depth; ph represents the category, and the possible values are o, w, and g, where o, w, and g represent oil, water, and gas, respectively.
[0050] In a second aspect, an embodiment of the present invention provides a shale oil reservoir development simulation device based on machine learning, the device comprising:
[0051] The first acquisition module is used to obtain the initial geological parameters, initial simulation time step and preset maximum residual value combination of the target shale oil reservoir;
[0052] a first simulation module, configured to perform a shale reservoir development simulation using a finite volume method based on the initial geological parameters, the initial simulation time step, and the preset maximum residual value combination, until a preset first simulation stop condition is reached, thereby obtaining a step parameter data set and new geological parameters;
[0053] A first training module is used to train a water phase time step prediction model, an oil phase time step prediction model, and a gas phase time step prediction model based on the step parameter data set;
[0054] a first prediction module, configured to combine the preset maximum residual value combination with the new geological parameters, and input the combined values into the oil phase time step prediction model, the water phase time step prediction model, and the gas phase time step prediction model, respectively, and obtain a new simulation time step based on the corresponding oil phase time step, water phase time step, and gas phase time step;
[0055] A second simulation module is used to perform shale oil reservoir development simulation based on the new geological parameters and the new simulation time step to obtain updated geological parameters;
[0056] The second acquisition module is used to repeatedly execute the steps of obtaining an updated simulation time step according to the updated geological parameters, and performing shale oil reservoir development simulation according to the updated simulation time step to obtain new updated geological parameters until a preset second simulation stop condition is reached, thereby obtaining a shale oil reservoir development simulation result.
[0057] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned machine learning-based shale oil reservoir development simulation method.
[0058] In a fourth aspect, an embodiment of the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the shale oil reservoir development simulation method based on machine learning as described above is implemented.
[0059] In a fifth aspect, an embodiment of the present invention provides a computer program product comprising instructions, which, when executed on a computer device, enables the computer device to execute the above-mentioned machine learning-based shale oil reservoir development simulation method.
[0060] The beneficial effects of the above technical solutions provided by the embodiments of the present invention include at least:
[0061] The machine learning-based shale reservoir development simulation method provided by an embodiment of the present invention first performs an early shale reservoir development simulation based on the finite volume method. When the cumulative simulation time length reaches a first preset time length, a step parameter data set is formed based on the geological parameters and time step obtained from the early shale reservoir development simulation. Then, time step prediction models corresponding to the three phases of oil, water, and gas are trained based on the step parameter data set for subsequent shale reservoir development simulation. In the subsequent shale reservoir development simulation, the maximum residual value is directly used as the input value of the time step prediction model to obtain the time step for the later shale reservoir development simulation, thereby completing the shale reservoir development simulation and obtaining a shale reservoir development simulation result. The finite volume method is used to achieve preliminary simulation of shale reservoir development, establish a preliminary understanding of the dynamic behavior of shale reservoirs, and accumulate data to provide rich training samples for the subsequent time step prediction model. Through the learning and optimization of the machine learning algorithm, the time step can be accurately predicted in the subsequent shale reservoir development simulation. While maintaining the stability of the shale reservoir model and ensuring the accuracy of the numerical simulation, the time step of each step in the shale reservoir development simulation is increased, which can effectively accelerate the progress of subsequent shale reservoir development simulation, reduce the total number of shale reservoir development simulation steps, speed up the simulation speed, and improve the computing efficiency, which is helpful for studying long-term reservoir development, improving the stability of the simulation, and reducing the oscillation in the numerical simulation. It is easy to operate and highly practical.
[0062] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.
[0063] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0064] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0065] Figure 1 A schematic diagram of the steps of a shale oil reservoir development simulation method based on machine learning provided in an embodiment of the present invention;
[0066] Figure 2 A schematic diagram of the steps for simulating shale reservoir development based on a combination of initial geological parameters, initial simulation time step, and preset maximum residual value provided by an embodiment of the present invention;
[0067] Figure 3A schematic diagram showing a comparison of time steps for shale reservoir development simulation using different methods provided in an embodiment of the present invention;
[0068] Figure 4 A schematic diagram of the structure of a shale oil reservoir development simulation device based on machine learning provided in an embodiment of the present application. DETAILED DESCRIPTION
[0069] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.
[0070] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.
[0071] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
[0072] As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.
[0073] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.
[0074] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.
[0075] It should be understood that the size of the serial numbers of the steps in the following embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.
[0076] In order to illustrate the technical solution of the present application, specific embodiments are provided below.
[0077] The inventors found that in the prior art, shale oil reservoirs after hydraulic fracturing exhibit typical multi-scale seepage characteristics, which means that changes in parameters such as permeability and porosity at different spatial scales need to be considered in numerical simulations. The multi-scale nature increases the complexity of the model. Traditional numerical simulation methods often require detailed grid division when dealing with multi-scale problems, resulting in a large model size, which significantly reduces the simulation speed. At the same time, shale oil reservoir development involves multi-phase and multi-component seepage, making numerical simulation particularly complex. Dealing with the dual complexity between fluid and matrix requires a lot of computing resources, so current numerical simulation methods often sacrifice speed while pursuing simulation accuracy. Therefore, how to effectively solve the problem of slow simulation speed and improve simulation efficiency has become an important issue that needs to be addressed in the current research on numerical simulation of shale oil reservoirs. Based on this, the inventors have made the present invention after further research and development, providing a shale oil reservoir development simulation method and device based on machine learning.
[0078] Example 1
[0079] The embodiment of the present invention provides a shale oil reservoir development simulation method based on machine learning, referring to Figure 1 As shown, the method includes:
[0080] S101: Obtaining initial geological parameters, initial simulation time step, and preset maximum residual value combination of the target shale oil reservoir.
[0081] In step S101, initial geological parameters, an initial simulation time step, and a preset maximum residual value combination are first obtained for the target shale reservoir. The initial geological parameters include, but are not limited to, porosity, absolute permeability, depth, and the viscosities and pressures of the oil, water, and gas phases. A shale reservoir geological model is then established based on the initial geological parameters. The preset maximum residual value combinations include the maximum oil phase residual, the maximum water phase residual, and the maximum gas phase residual.
[0082] Those skilled in the art can implement the steps of establishing a shale reservoir geological model based on the initial geological parameters and setting the value of the combination of the initial simulation time step and the preset maximum residual value according to the detailed description in the prior art, which is not specifically limited in the embodiments of this application. In a specific embodiment, the maximum oil phase residual, the maximum water phase residual, and the maximum gas phase residual are all within the range of 1.0×10 -6 ~1.0×10 -4 .
[0083] S102: Based on the initial geological parameters, the initial simulation time step, and the preset maximum residual value combination, the finite volume method is used to simulate the shale reservoir development until a preset first simulation stop condition is reached, thereby obtaining a step parameter data set and new geological parameters.
[0084] S103: Based on the step parameter data set, a water phase time step prediction model, an oil phase time step prediction model, and a gas phase time step prediction model are obtained through training.
[0085] In the above step S103, the specific implementation process of obtaining the water phase time step prediction model, the oil phase time step prediction model and the gas phase time step prediction model based on the step parameter data set training may include: performing data preprocessing (which may include data cleaning, etc.) and data partitioning (dividing the step parameter data set into training set, validation set and test set), selecting a suitable time step prediction model (such as BP neural network, random forest, etc.) for training, and obtaining the water phase time step prediction model, the oil phase time step prediction model and the gas phase time step prediction model.
[0086] S104: Combining the preset maximum residual value combination with the new geological parameters, inputting them into the oil phase time step prediction model, the water phase time step prediction model and the gas phase time step prediction model respectively, and obtaining a new simulation time step based on the corresponding oil phase time step, water phase time step and gas phase time step.
[0087] S105: Perform shale oil reservoir development simulation based on the new geological parameters and the new simulation time step to obtain updated geological parameters.
[0088] In the above step S105, based on the new geological parameters obtained in the above step S102, the shale oil reservoir development simulation is continued based on the finite volume method within the new simulation time step to obtain updated geological parameters.
[0089] S106: Repeat the steps of obtaining an updated simulation time step according to the updated geological parameters, and performing shale reservoir development simulation according to the updated simulation time step to obtain new updated geological parameters until a preset second simulation stop condition is reached, thereby obtaining a shale reservoir development simulation result.
[0090] In the embodiment of the present application, a preliminary simulation of shale oil reservoir development is achieved through the finite volume method, a preliminary understanding of the dynamic behavior of the shale oil reservoir is established, and data is accumulated at the same time to provide rich training samples for the subsequent time step prediction model. Through the learning and optimization of the machine learning algorithm, the time step is accurately predicted in the subsequent shale oil reservoir development simulation. While maintaining the stability of the shale oil reservoir model and ensuring the accuracy of the numerical simulation, the time step of each step in the shale oil reservoir development simulation is increased, which can effectively accelerate the progress of the subsequent shale oil reservoir development simulation, reduce the total number of shale oil reservoir development simulation steps, speed up the simulation speed, improve the computing efficiency, help to study long-term oil reservoir development, improve the stability of the simulation, reduce the oscillation in the numerical simulation, and is easy to operate and highly practical.
[0091] In the above step S102, based on the initial geological parameters, the initial simulation time step and the preset maximum residual value combination, the finite volume method is used to simulate the shale reservoir development until the preset first simulation stop condition is reached, and the step parameter data set and the new geological parameters are obtained. Figure 2 As shown, the following steps S1021-S1023 are included:
[0092] S1021: Based on the initial geological parameters and the initial simulation time step, the residual equation of the three-component numerical simulation model is established.
[0093] In the embodiment of the present application, first, the reservoir space of the shale oil reservoir geological model is divided into discrete control volumes or grid units, which is convenient for the subsequent numerical simulation of shale oil reservoir development using the finite volume method. Then, considering the starting pressure gradient and stress sensitivity effect, the residual equation of the three-component numerical simulation model is established. The starting pressure gradient refers to the pressure gradient required to overcome the initial resistance in the startup stage of oil and gas production. The stress sensitivity effect refers to the fact that during the development of the oil reservoir, the changes in the physical properties of the rock (such as permeability and fracture permeability) are affected by the stress changes in the formation. In order to more accurately simulate the shale oil reservoir development process, the present application introduces the starting pressure gradient and stress sensitivity effect when establishing the residual equation of the mass conservation equation, which can effectively improve the accuracy of the shale oil reservoir development simulation results and make it closer to the actual shale oil reservoir development situation.
[0094] Among them, those skilled in the art can implement the discretization operation of shale oil reservoir development space based on the detailed description of the prior art, which will not be elaborated as the focus of the present invention.
[0095] The residual equations of the three-component numerical simulation model include: the residual equation of the mass conservation equation of the oil component, the residual equation of the mass conservation equation of the water component, and the residual equation of the mass conservation equation of the gas component.
[0096] Based on the porosity, density of the oil component in the oil phase, oil phase saturation and initial simulation time step in the initial geological parameters, the residual equation of the oil component mass conservation equation is established based on the following formula 1:
[0097]
[0098] Where R o is the residual of the oil component mass conservation equation, φ represents the porosity, ρ Oo is the density of the oil component in the oil phase, S o is the oil phase saturation, t is the initial simulation time step, u o is the oil phase flow rate, q O represents the underground mass source and sink of oil components, o represents oil, is the divergence operator, is the partial derivative operator.
[0099] Based on the porosity, water phase density, water phase saturation and initial simulation time step in the initial geological parameters, the residual equation of the water component mass conservation equation is established based on the following formula 2:
[0100]
[0101] Where R w is the residual of the water component mass conservation equation, φ represents the porosity, ρ w is the density of the water phase, S w is the water phase saturation, t is the initial simulation time step, u w is the water phase flow rate, q W represents the underground mass source and sink of water components, w represents water, is the divergence operator, is the partial derivative operator.
[0102] Based on the porosity, density of gas components in the gas phase, gas saturation, and initial simulation time step in the initial geological parameters, the residual equation of the gas component mass conservation equation is established based on the following formula 3:
[0103]
[0104] Where Rg is the residual of the gas component mass conservation equation, φ represents the porosity, ρ Go is the density of the gas component in the gas phase, ρ G and ρ g is the gas phase density, S o is the oil phase saturation, S G is the gas phase saturation, t is the initial simulation time step, u o is the oil phase flow rate, u g is the gas phase flow rate, q G represents the underground mass source and sink of gas components, g represents gas, is the divergence operator, is the partial derivative operator.
[0105] The oil phase flow rate, water phase flow rate, and gas phase flow rate in Formula 1, Formula 2, and Formula 3 are obtained by the following Formula 4:
[0106]
[0107] Where u ph is the flow rate of the pH phase, k is the absolute permeability, k rph is the relative permeability of the pH phase, μ ph is the viscosity of the pH phase, p ph is the pressure of the ph phase, ρ ph is the density of the ph phase, g is the acceleration of gravity; D is the depth; ph represents the category, and the possible values are o, w, and g, where o, w, and g represent oil, water, and gas, respectively.
[0108] S1022: Based on the initial geological parameters and the initial simulation time step, a finite volume method is used to simulate the development of a shale oil reservoir. That is, a numerical method (such as a finite difference method or a finite element method) is used to iteratively solve the residual equations of the three-component numerical simulation model. When the residuals of the mass conservation equation of the oil component are all less than the maximum oil phase residual, the residuals of the mass conservation equation of the water component are all less than the maximum water phase residual, and the residuals of the mass conservation equation of the gas component are all less than the maximum gas phase residual within the initial simulation time step, the first geological parameters and the first time step are obtained. The calculation formula for the first time step is as follows:
[0109]
[0110] Where, Δt new represents the first time step, Δt old represents the initial simulation time step, fac1 and fac2 represent the time increase factor and time decrease factor respectively, dp ph is the maximum pressure change value within the initial simulation time step, dp phmax Increase the maximum allowable pressure change for the initial simulation time step, dSph is the maximum saturation change value within the initial simulation time step, dS phmax The maximum saturation change allowed for increasing the initial simulation time step is ph, which represents the category and can be o, w, or g, where o, w, and g represent oil, water, and gas, respectively.
[0111] Among them, those skilled in the art can set the time increase factor and the time reduction factor according to the detailed description in the prior art, and are not specifically limited in the embodiments of the present application. In a specific embodiment, the time increase factor has a value range of 1.2 to 3, and the time reduction factor has a value range of 0.2 to 0.8.
[0112] S1023: Repeat the above process of using the finite volume method to simulate shale reservoir development based on the first geological parameters and the first time step until a preset first simulation stop condition is reached, thereby obtaining a step parameter data set and new geological parameters.
[0113] The preset first simulation stop condition is that the cumulative time length of the shale oil reservoir development simulation reaches the first preset time length. The following steps S10231-S10233 are used to determine whether the progress of the shale oil reservoir development simulation reaches the preset first simulation stop condition:
[0114] S10231: Accumulate the initial simulation time step and all simulation time steps obtained in the iterative solution process to obtain the first cumulative time length of the shale oil reservoir development simulation, and determine whether the first cumulative time length reaches the first preset time length. If so, execute step S10233; if not, execute step S10234.
[0115] In a specific embodiment, the first preset time length ranges from 50 to 100, with the unit being days.
[0116] S10232: Obtain all simulation time steps and their corresponding geological parameters obtained during the iterative solution process, construct a step parameter data set, and use the geological parameters obtained in the last iteration as new geological parameters.
[0117] S10233: Re-execute the steps of performing shale oil reservoir development simulation and iterative solution based on the first geological parameters and the first time step.
[0118] In the above step S104, the preset maximum residual value combination is combined with the new geological parameters and input into the water phase time step prediction model, the oil phase time step prediction model and the gas phase time step prediction model respectively. According to the corresponding oil phase time step, water phase time step and gas phase time step obtained, a new simulation time step is obtained, including:
[0119] The maximum oil phase residual in the preset maximum residual value combination is combined with the new geological parameters and input into the oil phase time step prediction model. The maximum water phase residual in the preset maximum residual value combination is combined with the new geological parameters and input into the water phase time step prediction model. The maximum gas phase residual in the preset maximum residual value combination is combined with the new geological parameters and input into the gas phase time step prediction model to obtain the oil phase time step, water phase time step and gas phase time step, as shown in the following formula 6:
[0120]
[0121] Where, Δt 1max , Δt 2max , Δt 3max represents the oil phase time step, water-gas phase time step and gas phase time step respectively, f1, f2 and f3 represent the oil phase time step prediction model, water phase time step prediction model and gas phase time step prediction model respectively, φ represents porosity, k is absolute permeability, S o 、S w 、S g Represent the saturation of oil phase, water phase and gas phase respectively, p o 、p w 、p g Represent the pressure values of oil phase, water phase and gas phase respectively, q O ,q W ,q G are the underground mass sources and sinks of oil, water and gas components respectively, R omax 、R wmax 、R gmax They represent the maximum oil phase residual, the maximum water phase residual, and the maximum gas phase residual, respectively;
[0122] The minimum value among the oil phase time step, water phase time step and gas phase time step is used as the new simulation time step. The expression is as follows:
[0123] Δt new =min(Δt 1max ,Δt 2max ,Δt 3max ) Formula 7;
[0124] Where, Δt new Indicates the new simulation time step.
[0125] By directly using the preset maximum residual value as the input value of the time step prediction model, the time step of the next step of shale reservoir development simulation is obtained. While maintaining the stability of the shale reservoir model and ensuring the accuracy of numerical simulation, the time step of each step in the shale reservoir development simulation is increased, thereby reducing the total number of shale reservoir development simulation steps and improving computational efficiency. This is helpful for studying long-term reservoir development, improving simulation stability, and reducing oscillations in numerical simulation.
[0126] In step S106, the following steps S1061-S1063 are used to determine whether the progress of the shale reservoir development simulation has reached a preset second simulation stop condition. The preset second simulation stop condition is that the cumulative duration of the shale reservoir development simulation has reached a second preset duration, and the second preset duration must be greater than the first preset duration.
[0127] S1061: Accumulate the initial simulation time step and all simulation time steps obtained during the shale oil reservoir development simulation to obtain a second cumulative time length of the shale oil reservoir development simulation, and determine whether the second cumulative time length reaches a second preset time length. If so, execute step S1063; if not, execute step S1064.
[0128] The second preset time length represents the total duration of the shale oil reservoir development simulation.
[0129] S1062: Obtain shale oil reservoir development simulation results based on all simulation time steps and their corresponding geological parameters obtained during the shale oil reservoir development simulation process.
[0130] S1063: re-execute the above steps of obtaining an updated simulation time step according to the updated geological parameters, and performing shale oil reservoir development simulation according to the updated simulation time step to obtain new updated geological parameters.
[0131] In the embodiment of the present application, during the shale oil reservoir development simulation process, the method further includes:
[0132] The later cumulative time length of the shale oil reservoir development simulation after the water phase time step prediction model, the oil phase time step prediction model, and the gas phase time step prediction model are obtained in step S104 is calculated.
[0133] If the later accumulated time length reaches n times of the first preset time length, all simulation time steps and their corresponding geological parameters obtained during the shale oil reservoir development simulation are obtained to obtain an updated step parameter data set.
[0134] Then, a new water phase time step prediction model, a new oil phase time step prediction model, and a new gas phase time step prediction model are trained based on the updated step parameter dataset.
[0135] Among them, when executing the training based on the updated step parameter data set to obtain a new time step prediction model, those skilled in the art can choose to retrain the new time step prediction model based on the updated step parameter data set in combination with the actual situation, or can perform incremental learning based on the time step prediction model obtained in step S103. This is not specifically limited in the embodiments of the present application. In a specific embodiment, the value range of n is 2 to 6.
[0136] In one specific embodiment, a shale reservoir geological model is established based on certain actual geological parameters. The shale reservoir geological model has a size of 2000 m × 1000 m, a horizontal well located at the center of the model, a depth of 1000 m, 20 fracturing stages, a fracture half-length of 160 m, a model porosity of 6%, a matrix permeability of 0.07 mD, an initial pressure of 53 MPa, a crude oil viscosity of 0.05 mPa·s, a bubble point pressure of 23.2 MPa, and a solution gas-oil ratio of 120. The shale reservoir geological model uses water injection to simulate the injection process of fracturing fluid. The fracturing fluid volume is 40,000 cubic meters. After the fracturing fluid is injected, constant differential pressure production is performed at a bottomhole flowing pressure of 40 MPa. The total simulation time (i.e., the second preset time length mentioned above) is 250 days. Figure 3 To compare the time steps for shale reservoir development simulations using this method (corresponding to the new method in the figure) and the finite volume method alone (corresponding to the original method in the figure), the time steps for the two methods at different time steps are shown. As can be seen from the figure, after using the time step prediction model to predict the next time step in step 17, the time step of this method is significantly improved, completing the simulation in 35 steps, while the original method requires 62 steps, a step reduction of 43.5%.
[0137] Example 2
[0138] Based on the same inventive concept, the embodiment of the present invention also provides a shale oil reservoir development simulation device based on machine learning, referring to Figure 4 As shown, the device includes:
[0139] The first acquisition module 101 is used to obtain the initial geological parameters, initial simulation time step and preset maximum residual value combination of the target shale oil reservoir;
[0140] A first simulation module 102 is configured to perform a shale reservoir development simulation using a finite volume method based on the initial geological parameters, the initial simulation time step, and the preset maximum residual value combination until a preset first simulation stop condition is reached, thereby obtaining a step parameter data set and new geological parameters;
[0141] A first training module 103 is configured to train a water phase time step prediction model, an oil phase time step prediction model, and a gas phase time step prediction model based on the step parameter data set;
[0142] A first prediction module 104 is configured to combine the preset maximum residual value combination with the new geological parameters, and input them into the oil phase time step prediction model, the water phase time step prediction model, and the gas phase time step prediction model, respectively, to obtain a new simulation time step based on the corresponding oil phase time step, water phase time step, and gas phase time step;
[0143] A second simulation module 105 is configured to perform shale reservoir development simulation based on the new geological parameters and the new simulation time step to obtain updated geological parameters;
[0144] The second acquisition module 106 is used to repeatedly execute the steps of obtaining an updated simulation time step based on the updated geological parameters, performing a shale reservoir development simulation based on the updated simulation time step, and obtaining new updated geological parameters until a preset second simulation stop condition is reached, thereby obtaining a shale reservoir development simulation result.
[0145] Example 3
[0146] Based on the same inventive concept, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the machine learning-based shale oil reservoir development simulation method described in the above-mentioned embodiment 1 is implemented.
[0147] Example 4
[0148] Based on the same inventive concept, an embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the shale oil reservoir development simulation method based on machine learning as described in the above-mentioned embodiment 1 is implemented.
[0149] Example 5
[0150] Based on the same inventive concept, an embodiment of the present invention also provides a computer program product containing instructions. When the computer program product is run on a computer device, the computer device executes the shale oil reservoir development simulation method based on machine learning as described in the above embodiment 1.
[0151] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) containing computer-usable program code.
[0152] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0153] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0155] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.
Claims
1. A shale oil reservoir development simulation method based on machine learning, characterized in that: include: Obtaining the initial geological parameters, initial simulation time step, and preset maximum residual value combination of the target shale oil reservoir; Based on the initial geological parameters, the initial simulation time step, and the preset maximum residual value, a finite volume method is used to perform a shale reservoir development simulation until a preset first simulation stop condition is reached, thereby obtaining a step parameter data set and new geological parameters; Based on the step parameter data set, a water phase time step prediction model, an oil phase time step prediction model, and a gas phase time step prediction model are trained; Combining the preset maximum residual value combination with the new geological parameters, the results are input into the oil phase time step prediction model, the water phase time step prediction model, and the gas phase time step prediction model, respectively, and obtaining a new simulation time step based on the corresponding oil phase time step, water phase time step, and gas phase time step; Performing shale oil reservoir development simulation based on the new geological parameters and the new simulation time step to obtain updated geological parameters; Repeat the steps of obtaining an updated simulation time step according to the updated geological parameters, performing a shale reservoir development simulation according to the updated simulation time step, and obtaining new updated geological parameters until a preset second simulation stop condition is reached, thereby obtaining a shale reservoir development simulation result.
2. The method according to claim 1, wherein The preset maximum residual value combination includes: maximum oil phase residual, maximum water phase residual, and maximum gas phase residual; The combination of the preset maximum residual value and the new geological parameters is input into the water phase time step prediction model, the oil phase time step prediction model, and the gas phase time step prediction model, respectively, and a new simulation time step is obtained according to the corresponding oil phase time step, water phase time step, and gas phase time step, including: The maximum oil phase residual, the maximum water phase residual, and the maximum gas phase residual are combined with the new geological parameters and input into the oil phase time step prediction model, the water phase time step prediction model, and the gas phase time step prediction model, respectively. Based on the following formula, the oil phase time step, the water phase time step, and the gas phase time step are obtained: Where Δt 1max , Δt 2max , Δt 3max represents the oil phase time step, water-gas phase time step and gas phase time step respectively, f1, f2 and f3 represent the oil phase time step prediction model, water phase time step prediction model and gas phase time step prediction model respectively, φ represents porosity, k is absolute permeability, S o 、S w 、S g Represent the saturation of oil phase, water phase and gas phase respectively, p o 、p w 、p g Represent the pressure values of oil phase, water phase and gas phase respectively, q O ,q W ,q G are the underground mass sources and sinks of oil, water and gas components respectively, R omax 、R wmax 、R gmax They represent the maximum oil phase residual, the maximum water phase residual, and the maximum gas phase residual, respectively; The minimum value among the oil phase time step, the water phase time step and the gas phase time step is taken as the new simulation time step.
3. The method according to claim 1, wherein Also includes: In the process of obtaining an updated simulation time step according to the updated geological parameters and performing a shale reservoir development simulation according to the updated simulation time step to obtain new updated geological parameters, calculating a later cumulative time length of the shale reservoir development simulation after training the water phase time step prediction model, the oil phase time step prediction model, and the gas phase time step prediction model; If the later accumulated time length reaches a multiple of the first preset time length, all simulation time steps and their corresponding geological parameters obtained in the shale oil reservoir development simulation are obtained to obtain an updated step parameter data set; A new water phase time step prediction model, a new oil phase time step prediction model and a new gas phase time step prediction model are trained based on the updated step parameter data set.
4. The method according to claim 2, wherein The finite volume method is used to perform shale reservoir development simulation based on the initial geological parameters, the initial simulation time step, and the preset maximum residual value combination until a preset first simulation stop condition is reached, thereby obtaining a step parameter data set and new geological parameters, including: Based on the initial geological parameters and the initial simulation time step, a residual equation of the three-component numerical simulation model is established; the residual equation of the three-component numerical simulation model includes a residual equation of the oil component mass conservation equation, a residual equation of the water component mass conservation equation, and a residual equation of the gas component mass conservation equation; Based on the initial geological parameters and the initial simulation time step, the residual equation of the three-component numerical simulation model is solved using the finite volume method. When the residuals of the oil component mass conservation equation within the initial simulation time step are all less than the maximum oil phase residual, the residuals of the water component mass conservation equation are all less than the maximum water phase residual, and the residuals of the gas component mass conservation equation are all less than the maximum gas phase residual, the first geological parameters are obtained, and the first time step is obtained based on the following formula: Where Δt new represents the first time step, Δt old represents the initial simulation time step, fac1 and fac2 represent the time increase factor and time decrease factor respectively, dp ph is the maximum pressure change value within the initial simulation time step, dp phmax Increase the maximum allowable pressure change value for the initial simulation time step, dS ph is the maximum saturation change value within the initial simulation time step, dS phmax The maximum saturation change value allowed for increasing the initial simulation time step, ph represents the category, and the possible values are o, w, and g, where o, w, and g represent oil, water, and gas, respectively; The above process of simulating shale reservoir development based on the first geological parameter and the first time step is repeatedly executed until a preset first simulation stop condition is reached, thereby obtaining the step parameter data set and the new geological parameters.
5. The method according to claim 4, wherein The preset first simulation stop condition is that the cumulative time length of the shale oil reservoir development simulation reaches a first preset time length; Whether the progress of shale oil reservoir development simulation reaches the preset first simulation stop condition is determined by the following method: determining a first cumulative time length for shale reservoir development simulation based on the initial simulation time step and all simulation time steps obtained in the iterative solution process, and determining whether the first cumulative time length reaches a first preset time length; If yes, obtain all simulation time steps and their corresponding geological parameters obtained in the iterative solution process, construct the step parameter data set, and use the geological parameters obtained in the last iteration as the new geological parameters; If not, the above steps of performing shale oil reservoir development simulation based on the first geological parameter and the first time step and iteratively solving the problem are repeated.
6. The method according to claim 5, wherein The preset second simulation stop condition is that the cumulative time length of the shale oil reservoir development simulation reaches a second preset time length; the second preset time length is greater than the first preset time length; Whether the progress of the shale oil reservoir development simulation has reached the preset second simulation stop condition is determined by the following method: determining a second cumulative time length of the shale reservoir development simulation based on the initial simulation time step and all simulation time steps obtained during the shale reservoir development simulation, and determining whether the second cumulative time length reaches a second preset time length; If so, obtaining the shale oil reservoir development simulation result based on all simulation time steps and their corresponding geological parameters obtained during the shale oil reservoir development simulation process; If not, the above steps of obtaining an updated simulation time step according to the updated geological parameters are re-executed, and shale oil reservoir development simulation is performed according to the updated simulation time step to obtain new updated geological parameters.
7. The method according to claim 4, wherein The residual equation of the three-component numerical simulation model is established based on the initial geological parameters and the initial simulation time step, including: Based on the initial geological parameters and the initial simulation time step, the residual equation of the oil component mass conservation equation is established based on the following formula: Where R o is the residual of the oil component mass conservation equation, φ represents the porosity, ρ Oo is the density of the oil component in the oil phase, S o is the oil phase saturation, t is the initial simulation time step, u o is the oil phase flow rate, q O represents the underground mass source and sink of oil components, o represents oil, is the divergence operator, is the partial derivative operator; Based on the initial geological parameters and the initial simulation time step, the residual equation of the water component mass conservation equation is established based on the following formula: Where R w is the residual of the water component mass conservation equation, φ represents the porosity, ρ w is the density of the water phase, S w is the water phase saturation, t is the initial simulation time step, u w is the water phase flow rate, q W represents the underground mass source and sink of water components, w represents water, is the divergence operator, is the partial derivative operator; Based on the initial geological parameters and the initial simulation time step, the residual equation of the gas component mass conservation equation is established based on the following formula: Where R g is the residual of the gas component mass conservation equation, φ represents the porosity, ρ Go is the density of the gas component in the oil phase, ρ G and ρ g is the gas phase density, S o is the oil phase saturation, S G is the gas phase saturation, t is the initial simulation time step, u o is the oil phase flow rate, u g is the gas phase flow rate, q G represents the underground mass source and sink of gas components, g represents gas, is the divergence operator, is the partial derivative operator.
8. The method according to claim 7, wherein The flow rates of the oil phase, water phase, and gas phase are calculated based on the following formula: Where u ph is the flow rate of the pH phase, k is the absolute permeability, k rph is the relative permeability of the pH phase, μ ph is the viscosity of the pH phase, p ph is the pressure of the ph phase, ρ ph is the density of the ph phase, g is the acceleration of gravity; D is the depth; ph represents the category, and the possible values are o, w, and g, where o, w, and g represent oil, water, and gas, respectively.
9. A shale oil reservoir development simulation device based on machine learning, characterized in that: include: The first acquisition module is used to obtain the initial geological parameters, initial simulation time step and preset maximum residual value combination of the target shale oil reservoir; a first simulation module, configured to perform a shale reservoir development simulation using a finite volume method based on the initial geological parameters, the initial simulation time step, and the preset maximum residual value combination, until a preset first simulation stop condition is reached, thereby obtaining a step parameter data set and new geological parameters; A first training module is used to train a water phase time step prediction model, an oil phase time step prediction model, and a gas phase time step prediction model based on the step parameter data set; a first prediction module, configured to combine the preset maximum residual value combination with the new geological parameters, and input the combined values into the oil phase time step prediction model, the water phase time step prediction model, and the gas phase time step prediction model, respectively, and obtain a new simulation time step based on the corresponding oil phase time step, water phase time step, and gas phase time step; A second simulation module is used to perform shale oil reservoir development simulation based on the new geological parameters and the new simulation time step to obtain updated geological parameters; The second acquisition module is used to repeatedly execute the steps of obtaining an updated simulation time step according to the updated geological parameters, and performing shale oil reservoir development simulation according to the updated simulation time step to obtain new updated geological parameters until a preset second simulation stop condition is reached, thereby obtaining a shale oil reservoir development simulation result.
10. A computer-readable storage medium storing instructions, which, when executed on a terminal, causes the terminal to execute the shale oil reservoir development simulation method based on machine learning as described in any one of claims 1 to 8.
11. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for simulating shale oil reservoir development based on machine learning as described in any one of claims 1 to 8 is implemented.
12. A computer program product comprising instructions, which, when executed on a computer device, causes the computer device to execute the shale oil reservoir development simulation method based on machine learning according to any one of claims 1 to 8.