Multi-domain heat-fluid-solid coupling simulation prediction method and system for shale gas reservoir thermal recovery

Abstract

The invention belongs to the technical field of heat-fluid-solid coupling simulation, and provides a multi-domain heat-fluid-solid coupling simulation prediction method and system for shale gas reservoir thermal recovery, and the method comprises the steps: obtaining a shale reservoir model; constructing a heat-fluid-solid coupling numerical model considering various nonlinear physical action mechanisms based on the obtained shale reservoir model; solving the shale gas reservoir thermal exploitation process according to the constructed heat-fluid-solid coupling numerical model to obtain a multi-domain dynamic evolution result; and according to the obtained multi-domain dynamic evolution result and a preset dual-drive prediction model, prediction optimization of the gas production behavior of the shale gas reservoir is carried out, and multi-domain heat-fluid-solid coupling simulation prediction for shale gas reservoir thermal recovery is completed.

Description

Technical Field

[0001] This invention belongs to the field of thermal-fluid-structure interaction simulation technology, specifically relating to a multi-domain thermal-fluid-structure interaction simulation prediction method and system for thermal recovery of shale gas reservoirs. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Shale gas, as an important unconventional clean energy source, has attracted much attention due to its abundant reserves, low carbon emissions, and high feasibility of extraction. However, typical shale gas reservoirs generally exhibit low porosity, ultra-low permeability, and complex fracture networks, making it difficult to achieve economically recoverable production using conventional extraction methods. Existing technologies such as horizontal wells and hydraulic fracturing can improve production to some extent, but due to the strong heterogeneity, multi-scale nature, and interactions between multiple nonlinear flow mechanisms and physical processes in shale reservoir pore structure and composition, shale gas production prediction and efficient development still face significant challenges.

[0004] In recent years, thermal enhancement technology, which injects heat into hydraulic fractures or reservoirs to regulate the temperature field, thereby controlling adsorption behavior and improving gas permeability, has gradually become one of the research directions in unconventional resource development. However, the introduction of the temperature field further strengthens the coupling effect between fluid transport, adsorption / desorption behavior, and stress response, making the shale gas reservoir production process exhibit complex characteristics of high coupling across multiple scales, multiple physical fields, and multiple mechanisms. At the macroscopic scale, the systematic modeling methods for the multi-domain thermal-fluid-solid multi-physical field coupling process of shale reservoirs are still imperfect, and related simulation and prediction technologies are not yet mature, mainly in the following two aspects: (1) Insufficient ability to characterize physical processes Shale reservoirs, after hydraulic fracturing, typically consist of an organic matrix domain, an inorganic matrix domain, a natural fracture domain, and a hydraulic fracture domain. This constitutes a four-dimensional media system with significant structural and transport mechanism differences. Furthermore, shale gas extraction is accompanied by the coupling of multiple physical fields, including heat transfer, gas seepage, and rock deformation. However, existing macroscopic numerical models are mostly based on assumptions of monoporous, diporous, triple-porosity, or equivalent media. This simplification of reservoir structure makes it difficult to fully characterize the structural differences and diverse transport mechanisms of the four-dimensional media system in shale reservoirs. Even those studies that have introduced four-dimensional media continuum models often neglect the dynamic response process under the full coupling of heat, fluid, and solid processes. In addition, existing models are insufficient in characterizing the evolution of reservoir properties with stress and temperature variations, as well as the nonlinear seepage within the matrix and the multi-mechanism gas transport.

[0005] (2) The simulation efficiency is low, making it difficult to support rapid prediction and optimization decision-making. Current research on shale gas reservoir development mainly relies on mechanism-driven numerical simulation methods. While these methods can realistically reproduce the physical essence of the thermal extraction process, their strong nonlinearity and tight variable coupling result in large computational loads, slow iterative convergence, and long simulation times under multiple parameter combinations and operating conditions, making it difficult to meet the needs of real-time prediction and scheme optimization. At the same time, existing data-driven methods are not deeply integrated with the mechanism models, lacking an effective mechanism for using large-scale simulation data generated by the mechanism models to train efficient surrogate models, thus hindering the accurate prediction of multi-physics coupling effects. Summary of the Invention

[0006] To address the aforementioned issues, this invention proposes a multi-domain thermo-fluid-solid coupling simulation and prediction method and system for shale gas reservoir thermal exploitation. By constructing a macroscopic-scale numerical model of full thermo-fluid-solid coupling involving organic matrix, inorganic matrix, natural fractures, and hydraulic fractures, a refined characterization of the multi-scale, multi-physics coupling mechanism during shale gas reservoir thermal exploitation is achieved. A data-driven prediction module is introduced, and a rapid mapping relationship is established by training on large-scale high-fidelity simulation data generated based on the mechanism model, thereby significantly improving the efficiency of dynamic gas production prediction and the response speed of engineering decisions.

[0007] According to some embodiments, the first aspect of the present invention provides a multi-domain thermal-fluid-structure coupling simulation and prediction method for thermal recovery of shale gas reservoirs, employing the following technical solution: A multi-domain thermal-fluid-structure interaction simulation and prediction method for thermal recovery of shale gas reservoirs includes: Obtain shale reservoir models; A thermo-fluid-structure interaction numerical model considering multiple nonlinear physical mechanisms was constructed based on the obtained shale reservoir model. The thermal-fluid-solid coupling numerical model was used to solve the thermal exploitation process of shale gas reservoirs, and the multi-domain dynamic evolution results were obtained. Based on the obtained multi-domain dynamic evolution results and the preset dual-drive prediction model, the prediction and optimization of the gas production behavior of shale gas reservoirs are carried out, and the multi-domain thermal-fluid-solid coupling simulation prediction for thermal recovery of shale gas reservoirs is completed.

[0008] As a further technical limitation, the preset dual-drive prediction model is a dual-drive prediction framework that integrates a mechanistic model and a data-driven approach; specifically: Multi-domain thermal-fluid-solid coupling numerical simulations were conducted under different geological parameters and mining conditions to generate a training dataset covering the parameter space of typical shale gas reservoirs. A data-driven prediction model incorporating physical constraints is built based on the training dataset; A two-way feedback mechanism between the mechanistic model and the data model is established based on the constructed data-driven prediction model.

[0009] Furthermore, the data-driven prediction model includes: The fundamental physical laws of the thermo-fluid-structure interaction process are introduced into the model training process in the form of constraints. Convolutional neural networks (CNNs) were used to extract the spatial distribution features of multiphysics fields within shale reservoirs, while long short-term memory networks (LSTMs) were used to capture temporal evolution features. During model training, the residuals of physical conservation equations are introduced as constraint terms to ensure that the prediction results conform to basic physical laws.

[0010] Furthermore, the bidirectional feedback mechanism includes: Use mechanistic models to provide high-quality training samples for data models; Use data-driven models to quickly predict results for different parameter combinations under working conditions, and use mechanistic models to screen parameters and optimize working conditions. When the data model's prediction results exceed the preset accuracy range, the mechanism model is automatically invoked for supplementary calculations and the training set is updated.

[0011] As a further technical limitation, the process for predicting and optimizing the gas production behavior of the shale gas reservoir is as follows: Based on the trained dual-drive prediction framework, the shale gas reservoir gas production behavior prediction model is invoked. Input the geological parameters, engineering parameters, and thermal enhancement scheme of the shale gas reservoir to be predicted, and quickly output the corresponding evolution process of seepage field, temperature field, stress field, and dynamic curve of gas production. Based on the dual-drive prediction model, the Bayesian optimization algorithm is used to quickly optimize the thermal production enhancement parameters and obtain the optimal combination of development parameters that satisfies the preset objective function. Based on the prediction results of different combinations of development parameters, a multi-scheme comparative analysis was conducted to complete the prediction optimization of the gas production behavior of shale gas reservoirs.

[0012] As a further technical limitation, the shale reservoir model includes an organic matrix domain, an inorganic matrix domain, a natural fracture domain, and a hydraulic fracture domain.

[0013] According to some embodiments, a second aspect of the present invention provides a multi-domain thermal-fluid-structure interaction simulation and prediction system for thermal recovery of shale gas reservoirs, employing the following technical solution: A multi-domain thermal-fluid-structure interaction simulation and prediction system for thermal recovery of shale gas reservoirs includes: The acquisition module is configured to acquire shale reservoir models; The building module is configured to construct a thermal-fluid-structure interaction numerical model based on the acquired shale reservoir model, taking into account multiple nonlinear physical mechanisms. The solver module is configured to solve the thermal exploitation process of shale gas reservoirs based on the constructed thermal-fluid-solid coupling numerical model, and obtain multi-domain dynamic evolution results. The prediction module is configured to predict and optimize the gas production behavior of shale gas reservoirs based on the obtained multi-domain dynamic evolution results and the preset dual-drive prediction model, and complete the multi-domain thermal-fluid-solid coupling simulation prediction for thermal recovery of shale gas reservoirs.

[0014] According to some embodiments, a third aspect of the present invention provides a computer-readable storage medium, employing the following technical solution: A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the multi-domain thermal-fluid-structure coupling simulation and prediction method for shale gas reservoir thermal recovery as described in the first aspect of the present invention.

[0015] According to some embodiments, the fourth aspect of the present invention provides an electronic device, which adopts the following technical solution: An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps in the multi-domain thermal-fluid-structure interaction simulation and prediction method for shale gas reservoir thermal recovery as described in the first aspect of the present invention.

[0016] According to some embodiments, the fifth aspect of the present invention provides a computer program product, which adopts the following technical solution: A computer program product includes software code, wherein the program in the software code performs the steps of the multi-domain thermal-fluid-structure coupling simulation and prediction method for thermal recovery of shale gas reservoirs as described in the first aspect of the present invention.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention achieves a refined characterization of the four-dimensional media structure and multi-mechanism transport processes in shale reservoirs by constructing a multi-domain thermo-fluid-structure interaction numerical model. It explicitly constructs a four-dimensional media system at a macroscopic scale, consisting of an organic matrix domain, an inorganic matrix domain, a natural fracture domain, and a hydraulic fracture domain. The differences between these domains in pore structure, flow mechanisms (such as slip flow, diffusion, and adsorption / desorption), and property evolution (such as stress sensitivity and temperature dependence) are considered separately. This effectively solves the problem of traditional single-porosity, dual-porosity, triple-porosity, or equivalent media models oversimplifying complex reservoir structures, leading to insufficient model representation of site engineering. Furthermore, by introducing thermo-fluid-structure interaction control equations, it accurately describes the dynamic impact of temperature field changes on gas adsorption behavior, seepage capacity, and geostress response, significantly improving the simulation accuracy of multi-physics interaction processes in shale gas reservoirs under thermally enhanced production conditions.

[0018] This invention balances physical realism with computational efficiency, enabling rapid prediction of shale gas production behavior. By integrating a mechanistic model with a data-driven prediction model, a mechanism-data dual-driven prediction model is constructed. While ensuring physical realism, it achieves efficient prediction of gas production dynamics in shale gas reservoirs under multiple operating conditions, effectively improving the inefficiency of traditional numerical simulation calculations for large-scale prediction of reservoir multi-physics responses under various operating conditions.

[0019] This invention can quantitatively assess the impact of thermal injection on gas desorption, seepage, and fracture conductivity in different media domains, revealing the main controlling factors of production capacity under the interaction of heat, fluid, and solid-state processes. This provides technical support for optimizing thermal enhancement parameters and conducting comparative analysis of multiple schemes, demonstrating good engineering applicability and promotional value. Furthermore, its rapid prediction capabilities can be embedded in intelligent well control, digital twins, or oil and gas reservoir management platforms to support dynamic adjustments to development strategies, thereby improving the ultimate recovery rate and economic benefits of shale gas reservoirs. Attached Figure Description

[0020] The accompanying drawings, which form part of this embodiment, are used to provide a further understanding of this embodiment. The illustrative embodiments and their descriptions are used to explain this embodiment and do not constitute an improper limitation of this embodiment.

[0021] Figure 1 This is a flowchart of the multi-domain thermal-fluid-structure interaction simulation and prediction method for thermal recovery of shale gas reservoirs in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the iterative solution process in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the structure of the multilayer neural network in Embodiment 1 of the present invention; Figure 4 This is a schematic diagram of a physical model of a shale gas reservoir in Embodiment 1 of the present invention; Figure 5 This is a comparison chart of the gas production from the multi-domain thermal-fluid-solid coupling simulation of the shale gas reservoir in Embodiment 1 of the present invention and the field measured data. Figure 6 This is a structural block diagram of the multi-domain thermal-fluid-structure coupling simulation and prediction system for shale gas reservoir thermal recovery in Embodiment 2 of the present invention. Detailed Implementation

[0022] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0023] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, 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 invention pertains.

[0024] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0025] In this invention, terms such as "upper," "lower," "left," "right," "front," "back," "vertical," "horizontal," "side," and "bottom" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used only to facilitate the description of the structural relationships of the various components or elements of this invention and do not specifically refer to any component or element in this invention. They should not be construed as limiting the invention.

[0026] In this invention, terms such as "fixed connection," "connected," and "linked" should be interpreted broadly, indicating a fixed connection, an integral connection, or a detachable connection; a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can determine the specific meaning of these terms in this invention based on the specific circumstances, and they should not be construed as limitations on the invention.

[0027] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0028] Example 1 Embodiment 1 of this invention introduces a multi-domain thermal-fluid-structure interaction simulation and prediction method for thermal recovery of shale gas reservoirs.

[0029] like Figure 1 The method shown is a multi-domain thermal-fluid-structure interaction simulation and prediction method for thermal recovery of shale gas reservoirs, including: Obtain shale reservoir models; A thermo-fluid-structure interaction numerical model considering multiple nonlinear physical mechanisms was constructed based on the obtained shale reservoir model. The thermal-fluid-solid coupling numerical model was used to solve the thermal exploitation process of shale gas reservoirs, and the multi-domain dynamic evolution results were obtained. Based on the obtained multi-domain dynamic evolution results and the preset dual-drive prediction model, the prediction and optimization of the gas production behavior of shale gas reservoirs are carried out, and the multi-domain thermal-fluid-solid coupling simulation prediction for thermal recovery of shale gas reservoirs is completed.

[0030] like Figure 2 As shown, the multi-domain thermal-fluid-structure interaction simulation and prediction method for shale gas reservoir thermal recovery includes the following steps: Step S01: Based on the site data and development conditions of the target shale gas reservoir, construct a shale reservoir model. The shale reservoir model includes an organic matrix domain, an inorganic matrix domain, a natural fracture domain, and a hydraulic fracture domain. In this embodiment, the organic matrix domain and the inorganic matrix domain are described using a dual-pore dual-permeability model; the natural fracture domain and the hydraulic fracture domain are described using a discrete fracture model (DFM).

[0031] In this embodiment, the fractures in the natural fracture domain are randomly distributed in space, and the geometric features of the fractures are characterized by statistical parameters or probability distribution functions, and are discretized into several independent fracture units in the shale reservoir geometric model; the geometric features of the fractures include, but are not limited to, fracture length, orientation, dip, areal density or linear density.

[0032] The initial and boundary conditions in this embodiment are set as follows: ; In this context, the initial pressure of the shale reservoir is uniformly distributed and set to the initial pressure value. p 0, This refers to the boundary flow of the organic matrix domain in shale reservoirs. For the boundary flow of the inorganic matrix domain of shale reservoirs, For the flux at the boundary of the fracture domain in shale reservoirs, This indicates that the outer boundary of the shale reservoir is a flow-free boundary. This indicates that the outer boundary of the shale reservoir is an adiabatic boundary, and the hydraulic fracture zone is also included. Set as a constant temperature heat source, with the temperature always at the injection temperature. Horizontal wellbore inner boundary Set to constant bottom hole flowing pressure .

[0033] Step S02: Construct a thermo-fluid-structure interaction numerical model considering multiple nonlinear physical mechanisms in the organic matrix domain, inorganic matrix domain, natural fracture domain, and hydraulic fracture domain. Achieve full coupling of the temperature field, seepage field, and stress field by setting inter-domain coupling terms. The numerical model in this embodiment refers to the multiphysics field governing equations that describe heat transfer, gas seepage, and rock mechanical response, including mechanical equilibrium equations, heat conduction equations, and flow control equations.

[0034] Based on the porous linear elasticity theory, this embodiment considers the volumetric strain induced by gas pressure changes, adsorbed gas desorption, and temperature changes. A deformation control equation for the shale matrix is ​​constructed, and an effective Biot coefficient is introduced to characterize the pore-skeleton interaction between the organic and inorganic matrix domains. The mechanical equilibrium equation is expressed as follows: ; in, Shear modulus Poisson's ratio, , For Biot coefficient, K Bulk modulus The coefficient of thermal expansion is For volume forces, For the maximum adsorption volume strain, This indicates an adsorption strain of 0.5. Constant pore pressure.

[0035] This embodiment takes into account the extremely low porosity and permeability of the shale matrix, and the fact that the reservoir energy is mainly borne by the rock skeleton, under the assumption of local thermal equilibrium ( m = g = Under certain conditions, heat transfer in shale reservoirs is mainly through conduction, while convective heat transfer is relatively weak and can be ignored. The heat conduction equation for shale reservoirs, based on the energy conservation equation, can be expressed as: ; in, T m Indicates the temperature of the rock matrix (solid framework). g This indicates the temperature of the gas inside the pores. Represents the density of a solid. This indicates the specific heat capacity of a solid. This represents the rate of change of temperature over time. Indicates the thermal conductivity of a solid. T Indicates reservoir temperature, Q This is a heat source term.

[0036] In this embodiment, the organic matrix domain considers the coexistence of free gas and adsorbed gas, and the flow control equation is described based on the law of conservation of mass, specifically expressed as follows: ; in, Indicates the mass of the adsorbed gas. Porosity of the organic matrix domain This indicates the gas density within the organic matrix domain. This indicates the apparent permeability of the organic matrix domain. Indicates the gas viscosity within the organic matrix domain. This indicates the gas pressure within the organic matrix domain. This represents the mass exchange term between the organic matrix domain and the inorganic matrix domain.

[0037] In this embodiment, the inorganic matrix domain is considered a continuous matrix, and the organic matrix domain is uniformly embedded in the inorganic matrix domain as discretely distributed "embedded units." The interface between the organic matrix domain and the inorganic matrix domain is defined as a "mass exchange interface," and the mass exchange items between the organic matrix domain and the inorganic matrix domain are defined as follows: By introducing a shape factor Modeling is performed using the Warren-Root pseudosteady-state transport term to describe gas transport from the organic matrix domain to the inorganic matrix domain, which can be expressed as: ; Among them, based on the Bi-Langmuir non-isothermal adsorption model that considers the two-layer adsorption mechanism, the total adsorption capacity per unit mass of reservoir rock in the organic matrix domain can be expressed as: ; in, The gas density under standard conditions. For reservoir rock density, For the volume of Langmuir, Indicates the first i Layer adsorption ratio (first layer) i (The proportion of layer adsorption to total adsorption). Indicates the first i Layer adsorption equilibrium constant.

[0038] In this embodiment, the adsorption equilibrium constant and It can be represented as: ; ; ; in, E 1 and E 2 represents the adsorption energies of the first and second layers, respectively.

[0039] This embodiment considers the dynamic response of the pore structure under the combined effects of temperature and effective stress. The dynamic porosity model of the organic matrix domain, which incorporates stress-sensitive effects, can be expressed as follows: ; in, The bulk modulus of pores. The compressibility coefficient of the organic matrix domain is given. The initial porosity of the organic matrix domain. This represents the initial pressure of the reservoir.

[0040] This embodiment constructs a nanopore apparent permeability model for the organic matrix domain based on multiple nonlinear transport mechanisms such as viscous flow, Knudsen diffusion, and surface diffusion, and considers the real gas effect under non-isothermal flow conditions. This model can be expressed as follows: ; in, The diffusion coefficient of viscous flow is . Knudsen diffusion coefficient , is the surface diffusion coefficient, The tortuosity of the pores in the organic matrix domain.

[0041] In this embodiment, only the presence of free gas is considered in the inorganic matrix domain. The flow control equation based on the law of conservation of mass can be expressed as: ; in, Porosity of the inorganic matrix domain This indicates the gas density within the inorganic matrix domain; This represents the apparent permeability of the inorganic matrix domain. Indicates the gas viscosity within the inorganic matrix domain. This indicates the gas pressure within the inorganic matrix domain. This represents the mass exchange term between the organic matrix domain and the inorganic matrix domain.

[0042] This embodiment considers the dynamic response of the pore structure under the combined effects of temperature and effective stress. The dynamic porosity model of the inorganic matrix domain, which incorporates stress-sensitive effects, can be expressed as: ; Among these, considering Knudsen diffusion, slip flow, and stress-sensitive effects, the apparent dynamic permeability model of the inorganic matrix domain can be expressed as: ; in, The slip coefficient, Represents the dimensionless sparsity coefficient. is the compressibility coefficient of the inorganic matrix domain.

[0043] Knudsen number ( It can be defined as: ; in, The mean free path of the molecules, The characteristic pore radius is denoted by .

[0044] In this embodiment, only free gas is considered in the natural and hydraulic fracture domains. Based on the law of conservation of mass, the flow control equation can be expressed as: ; in, Indicates the width of the fracture within a natural or hydraulic fracture domain. This indicates the porosity of natural or hydraulically fractured domains. Indicates the permeability of natural or hydraulic fracture domains. This indicates the gas viscosity within a natural or hydraulically fractured region. This indicates the gas pressure within the natural fracture domain and the hydraulic fracture domain. This indicates the gas density within natural and hydraulic fracture domains.

[0045] This embodiment considers the Klinkenberg effect and stress sensitivity effect. The apparent dynamic permeability model of the natural fracture domain is as follows: ; in, The slip coefficient of the natural fracture domain. is the compressibility coefficient of the natural fracture domain.

[0046] The apparent dynamic permeability model of the hydraulic fracture domain in this embodiment is as follows: ; Considering the non-ideal nature of gases, based on the real gas law, the gas density can be expressed as: ; in, The molar mass of the gas. R The fraction represents the ideal gas.

[0047] This embodiment uses the Mahmoud formula to calculate the compressibility factor at the corresponding pressure and temperature. Specifically, it can be expressed as: ; in, , T c and p c This indicates the critical temperature and critical pressure.

[0048] This implementation considers the effects of real gases, including gas viscosity. The calculation expression can be represented as: ; in, , , These are the fitting coefficients.

[0049] Step S03: Based on the multi-domain thermo-fluid-solid coupling numerical model, model parameters and initial and boundary conditions, the thermal exploitation process of shale gas reservoir is numerically solved to obtain the dynamic evolution results of temperature field, pressure field, stress field and gas production. In this embodiment, step S03 includes the following steps: Step S301: The reservoir model is meshed using a finite element structured mesh. Based on this, the multi-domain thermal-fluid-structure interaction control equations are spatially discretized, and the continuous partial differential control equations of each computational domain are transformed into a set of algebraic equations in the form of nodal unknowns. Step S302: Use the fully implicit time integration method to discretize the discretized algebraic equations in time to improve the stability of numerical calculation; Step S303: Construct a coupled solution matrix containing three unknown variables: pressure, temperature, and displacement, and use the Newton-Raphson iterative method to solve the matrix nonlinearly; Step S304: In the solution process, an adaptive time step control strategy is introduced, and the preprocessed conjugate gradient method (PCG) combined with algebraic multigrid (AMG) technology is used to accelerate the solution process of the linear equation system. Step S305: Output the results of the multi-domain thermo-fluid-solid coupling numerical simulation, including the evolution of the temperature field, pressure field, stress field, and gas production of the shale gas reservoir over time.

[0050] The general form of spatial discretization in this embodiment can be expressed as: ; Where M is the mass matrix; K is the stiffness matrix; U is the vector of nodal unknowns, including temperature, pressure and displacement variables; and F is the equivalent load vector.

[0051] This embodiment uses the fully implicit backward Euler method for time discretization, for any field variable. At time step n+1 The value at that location can be represented as: ; This embodiment uses the Newton-Raphson iterative method to nonlinearly solve the matrix, resulting in the residual vector R(U) = 0. k In the next iteration: ; ; R(U) can be defined as: ; in, The Jacobian matrix, whose specific form reflects the coupling relationship between the pressure, temperature, and displacement fields, can be expressed as: ; in, This indicates the impact of stress on oneself. , These represent the effects of temperature change and displacement on the pressure field, respectively. The effect of pressure changes on the temperature field. This indicates the effect of temperature on itself. This indicates the effect of displacement on the temperature field. , This indicates the effect of pressure and temperature on the displacement field. This indicates the effect of displacement on itself.

[0052] Step S0S4: Construct a dual-drive prediction model that integrates mechanistic model and data-driven approach; Step S04 in this embodiment mainly includes the following steps: Step S401: Conduct a series of multi-domain thermal-fluid-solid coupling numerical simulations under different geological parameters and mining conditions to generate a training dataset covering the parameter space of typical shale gas reservoirs; Step S402: Construct a data-driven prediction model incorporating physical constraints based on the training dataset. The prediction model includes: (1) Introduce the basic physical laws of the thermo-fluid-structure interaction process into the model training process in the form of constraints; (2) Use convolutional neural networks (CNN) to extract features from the spatial distribution data of pressure field, temperature field and stress field output by the mechanism model to obtain the spatial correlation features of multiple physical fields inside the reservoir; use long short-term memory network (LSTM) to model the sequence data of the above features changing over time in order to capture the temporal evolution of multiple physical fields during the exploitation of shale gas reservoirs. (3) The error between the prediction result and the calculation result of the mechanism model is used as the data-driven loss term, and the physical conservation equation residual is introduced as the physical constraint term, thereby improving the physical authenticity and generalization ability of the prediction result.

[0053] Step S403: Establish a two-way feedback mechanism between the mechanistic model and the data model, which mainly includes: (1) Use mechanistic models to provide high-quality training samples for data models; (2) Use data-driven models to quickly predict different parameter combinations, thereby assisting the mechanism model in parameter selection and operating condition optimization; (3) When the error between the prediction result of the data-driven model and the calculation result of the mechanism model exceeds the preset accuracy threshold When necessary, the mechanism model is automatically invoked to perform supplementary calculations and update the training set, thereby updating and training the data-driven model.

[0054] Step S05: Based on the framework established in step S04, realize rapid prediction and optimization of gas production behavior in shale gas reservoirs.

[0055] Step S05 in this embodiment includes the following steps: (1) Based on the dual-drive prediction model trained in step S04, call the shale gas reservoir gas production behavior prediction model; (2) Input the geological parameters, engineering parameters and thermal production enhancement scheme of the shale gas reservoir to be predicted, and quickly output the corresponding seepage field, temperature field, stress field evolution process and gas production dynamic curve; (3) Based on the dual-drive prediction model, the Bayesian optimization algorithm is used to quickly optimize the thermal production enhancement parameters (such as injection temperature, injection rate and injection time) to obtain the optimal combination of development parameters that meet the preset objective function. (4) Conduct multi-scheme comparative analysis based on the prediction results of different combinations of development parameters to provide a basis for the selection and decision-making of thermal exploitation schemes for shale gas reservoirs.

[0056] The training dataset in this embodiment includes pressure field, temperature field, stress field, and production dynamics data under different reservoir conditions and different thermal enhancement schemes.

[0057] The data-driven prediction model with physical constraints introduced in this embodiment is constructed using a multi-layer feedforward neural network structure, including an input layer, at least one hidden layer, and an output layer. The number of nodes in each layer is adaptively determined according to the dimension of the input parameters and the prediction target. A schematic diagram of its network structure is shown below. Figure 3 As shown.

[0058] In this embodiment, the number of input layer nodes m This is used to receive an input vector consisting of geological parameters, operating condition parameters, and initial state parameters of the shale gas reservoir. The input vector can be represented as: ; The input features in this embodiment include, but are not limited to: initial geological pressure, initial formation temperature, porosity of each domain, permeability of each domain, adsorbed gas content, Langmuir parameters, Poisson's ratio, elastic modulus, Biot coefficient, thermal conductivity, specific heat capacity, coefficient of thermal expansion, fracture density, organic matter content, bottom hole pressure, hot injection temperature, and production time.

[0059] The multilayer feedforward neural network in this embodiment includes at least one hidden layer, and the number of nodes in the hidden layer is set to... l The connection weights between the input layer and the hidden layer are represented as follows: ; The corresponding bias vector is expressed as: ; The output of the hidden layer node can be represented as: ; in, () is the nonlinear activation function of the hidden layer, which can preferably be the ReLU function.

[0060] The number of output layer nodes in this embodiment is: q This is used to output the target prediction results for shale gas reservoirs. The connection weight between the hidden layer and the output layer is represented as follows: ; The threshold corresponding to the output layer is represented as follows: ; The output of the output layer node is represented as follows: ; in, () is the activation function of the output layer, which can preferably be a linear function, used to output the characteristic quantities of the pressure field, temperature field, stress field, or gas production prediction results.

[0061] In this embodiment, the results obtained from multiple sets of multi-domain thermo-fluid-structure interaction numerical simulations are used as the desired output value. Construct a joint loss function that includes data error terms and physical constraint terms: ; Among them, the data error term Represented as: ; Among them, the physical constraint term is constructed by introducing the residuals of the heat conduction equation, the mass conservation equation, and the mechanical equilibrium equation, and is used to constrain the model prediction results to satisfy the basic physical conservation laws. These are the weighting coefficients.

[0062] This embodiment uses the Adam optimization algorithm to train the neural network, and adjusts the weights from the input layer to the hidden layer according to the joint loss function. and weights from hidden layer to output layer The model is iteratively updated to gradually bring its output closer to the expected value. After multiple rounds of sample training, a converged data-driven prediction model is obtained for the rapid prediction of gas production behavior in shale gas reservoirs.

[0063] The error between the prediction results of the data-driven model and the calculation results of the mechanism model in this embodiment can be expressed as: ; Among them, when the error Exceeding the preset accuracy threshold At that time, that is It automatically calls the mechanism model to perform supplementary calculations under the corresponding parameter conditions, generates new training samples and updates the training dataset, and retrains the data-driven model to improve prediction accuracy and physical consistency.

[0064] The following study takes a typical shale gas storage block as the research object. The overall geometric dimensions of the target reservoir are 1100m×290m×90m. A physical model of the shale gas reservoir is constructed based on the symmetry assumption. The physical model of the shale gas reservoir occupies one-quarter of the entire reservoir geometry, and the simulation domain size is 550m×145m×90m. The model includes 400 randomly oriented natural fractures with an average length of 40m within the entire reservoir domain, and 14 hydraulic fractures perpendicular to the horizontal well with a spacing of 30.5m. The half-length of each hydraulic fracture is 47.2m. The boundary conditions are set as follows: a maximum horizontal stress of 41.6 MPa is applied along the X-direction boundary, a minimum horizontal stress of 37.3 MPa is applied along the Y-direction boundary, and the remaining boundaries are subject to rolling constraints. All outer boundaries are set as flow-free boundaries, and a bottomhole flowing pressure of 3.69 MPa is applied as an inner boundary condition at the horizontal wellbore. The initial temperature field is 352 K, the outer boundaries are adiabatic, and thermal enhancement is achieved by injecting a 478 K high-temperature fluid. The established shale gas reservoir physical model is as follows: Figure 4 As shown.

[0065] This embodiment applies a multi-domain thermo-fluid-solid coupled numerical simulation method to a shale gas storage block. Based on field test data, experimental data, and reservoir parameters obtained from published literature, relevant modeling characteristic parameters for the organic matrix domain, inorganic matrix domain, natural fracture domain, and hydraulic fracture domain are determined. On this basis, multiple sets of numerical simulation studies are conducted under different geological parameters and thermal extraction conditions, and the results are fitted and validated with field production data. The numerical simulation results are then output, such as... Figure 5 As shown in the figure. The basic physical properties of the shale reservoir and fluids, as well as the characteristic parameters of each domain, are represented in Tables 1 and 2.

[0066] Table 1. Basic physical properties of shale reservoirs and fluids. parameter numerical values unit Reservoir dimensions (length × width × height) 1100×290×90 m Simulated area size 550×145×90 m Simulation cycle 1600 d initial temperature 352 K Heat injection temperature 478 K Initial pressure 20.34 MPa Bottom hole pressure 3.69 MPa Horizontal well length 426.7 m Young's modulus 34 GPa Maximum horizontal ground stress 41.6 MPa Minimum horizontal stress 37.3 MPa Poisson's ratio 0.2 Shale formation density 2580 <![CDATA[kg / m 3 ]]> Shale bulk modulus 8 GPa coefficient of thermal expansion 0.00003 1 / K Langmuir volume <![CDATA[2.72×10 -3 ]]> <![CDATA[m 3 / kg]]> Langmuir pressure <![CDATA[4.48×10 6 ]]> Pa Table 2 Multi-domain characteristic parameters of shale reservoirs

[0067] This embodiment constructs a multi-domain thermo-fluid-structure interaction (TFI) numerical model, achieving a refined characterization of the four-dimensional media structure and multi-mechanism transport processes in shale reservoirs. It explicitly constructs a four-dimensional media system at a macroscopic scale, consisting of an organic matrix domain, an inorganic matrix domain, a natural fracture domain, and a hydraulic fracture domain. The model considers the differences in pore structure, flow mechanisms (such as slip flow, diffusion, and adsorption / desorption), and property evolution (such as stress sensitivity and temperature dependence) among the different domains. This effectively solves the problem of traditional single-porosity, dual-porosity, triple-porosity, or equivalent media models oversimplifying complex reservoir structures, leading to insufficient model representation of site engineering. Furthermore, by introducing TFI governing equations, it accurately describes the dynamic impact of temperature field changes on gas adsorption behavior, permeability, and geostress response, significantly improving the simulation accuracy of multi-physics interaction processes in shale gas reservoirs under thermally enhanced production conditions.

[0068] This embodiment balances physical realism with computational efficiency, enabling rapid prediction of shale gas production behavior. By integrating the mechanism model with the data-driven prediction model, a mechanism-data dual-driven prediction model is constructed. While ensuring physical realism, it achieves efficient prediction of gas production dynamics of shale gas reservoirs under multiple operating conditions, effectively improving the inefficiency of traditional numerical simulation calculations for large-scale prediction of reservoir multi-physics responses under various operating conditions.

[0069] This embodiment can quantitatively assess the impact of thermal injection on gas desorption, seepage, and fracture conductivity in different media domains, revealing the main controlling factors of production capacity under the interaction of heat, fluid, and solid-state processes. This provides technical support for optimizing thermal enhancement parameters and conducting comparative analysis of multiple schemes, demonstrating good engineering applicability and promotional value. Furthermore, its rapid prediction capability can be embedded in intelligent well control, digital twins, or oil and gas reservoir management platforms to support dynamic adjustments to development strategies, thereby improving the ultimate recovery rate and economic benefits of shale gas reservoirs.

[0070] Example 2 Embodiment 2 of the present invention introduces a multi-domain thermal-fluid-structure coupling simulation and prediction system for thermal recovery of shale gas reservoirs.

[0071] like Figure 6 The multi-domain thermal-fluid-structure interaction simulation and prediction system for shale gas reservoir thermal recovery shown includes: The acquisition module is configured to acquire shale reservoir models; The building module is configured to construct a thermal-fluid-structure interaction numerical model based on the acquired shale reservoir model, taking into account multiple nonlinear physical mechanisms. The solver module is configured to solve the thermal exploitation process of shale gas reservoirs based on the constructed thermal-fluid-solid coupling numerical model, and obtain multi-domain dynamic evolution results. The prediction module is configured to predict and optimize the gas production behavior of shale gas reservoirs based on the obtained multi-domain dynamic evolution results and the preset dual-drive prediction model, and complete the multi-domain thermal-fluid-solid coupling simulation prediction for thermal recovery of shale gas reservoirs.

[0072] The detailed steps are the same as those of the multi-domain thermal-fluid-structure coupling simulation and prediction method for shale gas reservoir thermal recovery provided in Example 1, and will not be repeated here.

[0073] Example 3 Embodiment 3 of the present invention provides a computer-readable storage medium.

[0074] A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the multi-domain thermal-fluid-structure coupling simulation and prediction method for shale gas reservoir thermal recovery as described in Embodiment 1 of the present invention.

[0075] The detailed steps are the same as those of the multi-domain thermal-fluid-structure coupling simulation and prediction method for shale gas reservoir thermal recovery provided in Example 1, and will not be repeated here.

[0076] Example 4 Embodiment 4 of the present invention provides an electronic device.

[0077] An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, it implements the steps in the multi-domain thermal-fluid-structure coupling simulation and prediction method for shale gas reservoir thermal recovery as described in Embodiment 1 of the present invention.

[0078] The detailed steps are the same as those of the multi-domain thermal-fluid-structure coupling simulation and prediction method for shale gas reservoir thermal recovery provided in Example 1, and will not be repeated here.

[0079] Example 5 Embodiment 5 of the present invention provides a computer program product.

[0080] A computer program product includes software code, wherein the program in the software code performs the steps of the multi-domain thermal-fluid-structure coupling simulation and prediction method for thermal recovery of shale gas reservoirs as described in Embodiment 1 of the present invention.

[0081] The detailed steps are the same as those of the multi-domain thermal-fluid-structure coupling simulation and prediction method for shale gas reservoir thermal recovery provided in Example 1, and will not be repeated here.

[0082] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0083] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations 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, 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, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0084] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0086] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0087] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

[0088] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. A multi-domain thermal-fluid-structure interaction simulation and prediction method for thermal recovery of shale gas reservoirs, characterized in that, include: Obtain shale reservoir models; A thermo-fluid-structure interaction numerical model considering multiple nonlinear physical mechanisms was constructed based on the obtained shale reservoir model. The thermal-fluid-solid coupling numerical model was used to solve the thermal exploitation process of shale gas reservoirs, and the multi-domain dynamic evolution results were obtained. Based on the obtained multi-domain dynamic evolution results and the preset dual-drive prediction model, the prediction and optimization of the gas production behavior of shale gas reservoirs are carried out, and the multi-domain thermal-fluid-solid coupling simulation prediction for thermal recovery of shale gas reservoirs is completed.

2. The multi-domain thermal-fluid-structure interaction simulation and prediction method for shale gas reservoir thermal recovery as described in claim 1, characterized in that, The preset dual-drive prediction model is a dual-drive prediction framework that integrates a mechanism model and a data-driven approach; specifically: Multi-domain thermal-fluid-solid coupling numerical simulations were conducted under different geological parameters and mining conditions to generate a training dataset covering the parameter space of typical shale gas reservoirs. A data-driven prediction model incorporating physical constraints is built based on the training dataset; A two-way feedback mechanism between the mechanistic model and the data model is established based on the constructed data-driven prediction model.

3. The multi-domain thermal-fluid-structure interaction simulation and prediction method for shale gas reservoir thermal recovery as described in claim 2, characterized in that, The data-driven prediction model includes: The fundamental physical laws of the thermo-fluid-structure interaction process are introduced into the model training process in the form of constraints. A convolutional neural network was used to extract the spatial distribution characteristics of multi-physics fields in shale reservoirs, and a long short-term memory network was used to capture the temporal evolution characteristics. During model training, the residuals of physical conservation equations are introduced as constraint terms to ensure that the prediction results conform to basic physical laws.

4. The multi-domain thermal-fluid-structure interaction simulation and prediction method for shale gas reservoir thermal recovery as described in claim 2, characterized in that, The two-way feedback mechanism includes: Use mechanistic models to provide high-quality training samples for data models; Use data-driven models to quickly predict results for different parameter combinations under working conditions, and use mechanistic models to screen parameters and optimize working conditions. When the data model's prediction results exceed the preset accuracy range, the mechanism model is automatically invoked for supplementary calculations and the training set is updated.

5. The multi-domain thermal-fluid-structure interaction simulation and prediction method for shale gas reservoir thermal recovery as described in claim 1, characterized in that, The process of predicting and optimizing the gas production behavior of the shale gas reservoir is as follows: Based on the trained dual-drive prediction framework, the shale gas reservoir gas production behavior prediction model is invoked. Input the geological parameters, engineering parameters, and thermal enhancement scheme of the shale gas reservoir to be predicted, and quickly output the corresponding evolution process of seepage field, temperature field, stress field, and dynamic curve of gas production. Based on the dual-drive prediction model, the Bayesian optimization algorithm is used to quickly optimize the thermal production enhancement parameters and obtain the optimal combination of development parameters that satisfies the preset objective function. Based on the prediction results of different combinations of development parameters, a multi-scheme comparative analysis was conducted to complete the prediction optimization of the gas production behavior of shale gas reservoirs.

6. The multi-domain thermal-fluid-structure interaction simulation and prediction method for shale gas reservoir thermal recovery as described in claim 1, characterized in that, The shale reservoir model includes an organic matrix domain, an inorganic matrix domain, a natural fracture domain, and a hydraulic fracture domain.

7. A multi-domain thermal-fluid-structure interaction simulation and prediction system for thermal recovery of shale gas reservoirs, characterized in that, include: The acquisition module is configured to acquire shale reservoir models; The building module is configured to construct a thermal-fluid-structure interaction numerical model based on the acquired shale reservoir model, taking into account multiple nonlinear physical mechanisms. The solver module is configured to solve the thermal exploitation process of shale gas reservoirs based on the constructed thermal-fluid-solid coupling numerical model, and obtain multi-domain dynamic evolution results. The prediction module is configured to predict and optimize the gas production behavior of shale gas reservoirs based on the obtained multi-domain dynamic evolution results and the preset dual-drive prediction model, and complete the multi-domain thermal-fluid-solid coupling simulation prediction for thermal recovery of shale gas reservoirs.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the multi-domain thermal-fluid-structure interaction simulation and prediction method for thermal recovery of shale gas reservoirs as described in any one of claims 1-6.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the steps of the multi-domain thermal-fluid-structure interaction simulation and prediction method for thermal recovery of shale gas reservoirs as described in any one of claims 1-6.

10. A computer program product, comprising software code, characterized in that, The program in the software code executes the steps of the multi-domain thermal-fluid-structure interaction simulation and prediction method for thermal recovery of shale gas reservoirs as described in any one of claims 1-6.

Images