Method, system, device, and medium for predicting physical fields of thermal mass transport in porous media

Abstract

The application discloses a porous medium heat and mass transport physical field prediction method, system, device and medium. The porous medium heat and mass transport physical field prediction method comprises the following steps: obtaining dimensionally homogeneous input data of a porous medium heat and mass transport process under a target working condition, wherein the dimensionally homogeneous input data at least comprises pore structure, initial boundary value conditions and physical property parameters; taking the dimensionally homogeneous input data as data of an input end of a prediction model, and obtaining a prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model, wherein the prediction model is pre-trained according to a pair of dimensionally homogeneous input and output data obtained from pre-obtained porous medium heat and mass transport physical field data and heat and mass transport control equations, and the pair of dimensionally homogeneous input and output data comprises dimensionally homogeneous input data and dimensionally homogeneous output data under corresponding conditions. By adopting the application, the prediction efficiency of the porous medium heat and mass transport physical field can be effectively improved, and the cost investment of prediction can be reduced.

Description

Technical Field

[0001] This application relates to the field of porous media heat and mass transport technology, and in particular to a method, system, device and medium for predicting the physical field of porous media heat and mass transport. Background Technology

[0002] Thermo-mass transport processes in porous media are widely present in nature and engineering, with important applications in fluid transport within biological tissues, energy resource development and utilization, and aerospace thermal protection. Accurate descriptions of these processes are crucial for the optimized design and precise control of practical applications. Physical field information, such as phase field, velocity field, pressure field, and temperature field, provides a detailed description of the thermo-mass transport process and is a core element for reflecting the intrinsic physical characteristics of the process and conducting downstream optimization design.

[0003] Currently, commonly used methods for obtaining the physical fields of thermal and mass transport in porous media include microscopic model visualization experiments and pore-scale numerical simulations. Microscopic model visualization experiments use etching or additive manufacturing methods to fabricate porous structures and encapsulate them into microscopic models containing fluid inlet and outlet channels. Fluid is introduced into the microscopic model, and fluid phase field and velocity field information are obtained through visualization. However, this method is time-consuming and costly in fabricating the microscopic model, making it difficult to meet the rapid prediction requirements of practical applications. Pore-scale numerical simulations construct a set of algebraic equations through mesh generation, control equations, and initial-boundary conditions, and obtain detailed physical field information between pores through iterative calculations. However, due to the complex solid framework structure of porous media and the complex interface interactions of multiphase fluids, pore-scale numerical simulations are very difficult, with computation times reaching hours or even weeks, making it difficult to meet the rapid prediction requirements of practical applications and support dynamic optimization design. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, system, device, and medium for predicting the physical field of thermal and mass transport in porous media, which can effectively improve the prediction efficiency of the physical field of thermal and mass transport in porous media and reduce the cost of prediction.

[0005] In a first aspect, this application provides a method for predicting the physical field of thermal and mass transport in porous media, including:

[0006] Dimensional input data of the thermal and mass transport process of porous media under target operating conditions are obtained, wherein the dimensional input data includes at least pore structure, initial and boundary conditions and physical property parameters;

[0007] The dimensional input data is used as the input data to the prediction model, so as to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model, wherein:

[0008] The prediction model is pre-trained based on dimensional input-output data pairs obtained from pre-acquired porous medium thermo-mass transport physical field data and thermo-mass transport control equations. The dimensional input-output data pairs include dimensional input data and dimensional output data under conditions corresponding to the dimensional input data.

[0009] In some examples, before using the dimensional input data as input to the prediction model to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target operating condition, the method further includes:

[0010] Obtain physical field data of thermal and mass transport in porous media;

[0011] Based on the physical field data of the porous medium's thermal and mass transport, the dimensional input-output data pair is obtained;

[0012] The architecture of the initial prediction model was designed based on the physics of heat and mass transport.

[0013] In some examples, after designing the architecture of the initial prediction model based on knowledge of thermo-mass transport physics, the following is also included:

[0014] The initial prediction model is trained based on the constraints of the dimensional input-output data pair and the heat and mass transport control equation until the loss between the output of the initial prediction model and the label meets a predetermined condition, thus obtaining the trained prediction model. The label is the dimensional output data in the dimensional input-output data pair under the conditions corresponding to the dimensional input data.

[0015] In some examples, after designing the architecture of the initial prediction model based on knowledge of thermo-mass transport physics, the following is also included:

[0016] The dimensional input-output data pairs are made dimensionless to determine the dimensionless input-output data pairs;

[0017] The initial prediction model is trained based on the dimensionless input-output data pair and the control equations for heat and mass transport, until the loss between the output of the initial prediction model and the label meets a predetermined condition, thus obtaining the trained prediction model.

[0018] The label is the dimensionless output data in the dimensionless input-output data pair under the condition corresponding to the dimensionless input data.

[0019] In some examples, the prediction result of the porous medium thermal and mass transport physical field corresponding to the target operating condition obtained by the prediction model is the prediction result of the dimensionless physical field of porous medium thermal and mass transport corresponding to the target operating condition. After obtaining the prediction result of the dimensionless physical field of porous medium thermal and mass transport corresponding to the target operating condition by the prediction model, the method further includes:

[0020] Based on the operating parameters of the target operating condition, the dimensionless physical field output by the prediction model is converted into a dimensional physical field to obtain the final prediction result of the thermal and mass transport physical field of the porous medium under the target operating condition.

[0021] In some examples, the dimensionless transformation of the dimensional input-output data pairs to determine the dimensionless input-output data pairs is achieved through the following formula:

[0022]

[0023] Where: x, y, z are spatial coordinates, t is time, u, v, w are the fluid velocity components in the x, y, z directions respectively, p is the fluid pressure, L is the characteristic dimension, U is the characteristic velocity, ρ is the fluid density, and * indicates a dimensionless physical quantity.

[0024] In some examples, after dimensionless transformation of the dimensional input-output data pairs to determine the dimensionless input-output data pairs, the process further includes:

[0025] The dimensionless output data is standardized using the following formula:

[0026]

[0027] Where std represents the standardized variable, the averaging operator in the denominator represents the arithmetic mean of the data after the porous medium volume averaging operation, and Re represents the characteristic Reynolds number.

[0028] Secondly, a system for predicting the physical field of thermal and mass transport in porous media is provided, comprising:

[0029] The acquisition module is used to obtain dimensional input data of the thermal and mass transport process of porous media under target working conditions, wherein the dimensional input data includes at least pore structure, initial boundary conditions and physical property parameters;

[0030] The prediction module is used to take the dimensional input data as the input data of the prediction model, so as to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model. The prediction model is pre-trained based on the dimensional input-output data pair obtained from the pre-obtained porous medium heat and mass transport physical field data and the heat and mass transport control equations. The dimensional input-output data pair includes dimensional input data and dimensional output data under the conditions corresponding to the dimensional input data.

[0031] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the porous medium thermal mass transport physical field prediction method described in the first aspect and any possible implementation of the first aspect.

[0032] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the porous medium thermal-mass transport physical field prediction method described in the first aspect and any possible implementation thereof.

[0033] In the embodiments of this application, dimensional input data of the porous medium thermo-mass transport process under specified operating conditions are used as input data for the prediction model. This allows for rapid prediction of the porous medium thermo-mass transport physical field under the specified operating conditions. This method of using a machine learning model to predict the porous medium thermo-mass transport physical field replaces the experimental and numerical simulation methods used in related technologies. Since a trained model can infer the porous medium thermo-mass transport physical field under specified operating conditions without iteration, the prediction speed is effectively improved. Furthermore, the machine learning model has strong representational capabilities, thus ensuring the prediction accuracy of the physical field. In addition, the training of the machine learning model is driven by both data and the thermo-mass transport control equations, ensuring that the prediction results conform to physical laws. The interpretability and generalization of the machine learning model can be enhanced by explicitly incorporating physical knowledge. Moreover, high-precision pore-scale pressure and temperature fields that match the characteristics of experimental data can be obtained, achieving the effect of indirect experimental measurement of pore-scale pressure and temperature fields. Because experimental methods in related technologies struggle to obtain pore-scale pressure and temperature fields, and numerical simulations suffer from simplification, the accuracy of the simulated pressure and temperature fields is limited. Therefore, the embodiments of this application utilize data and equations jointly to drive the machine learning model training phase, enabling the acquisition of pressure and temperature fields that satisfy the constraints of the phase field, velocity field, and the mass, momentum, and energy conservation equations required for experimental measurements. Depending on the characteristics of different physical processes, the pressure and temperature fields can be acquired separately or simultaneously. Attached Figure Description

[0034] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0035] Figure 1 This is a flowchart of the method for predicting the physical field of thermal and mass transport in porous media provided in the embodiments of this application;

[0036] Figure 2 This is another flowchart of the method for predicting the physical field of thermal and mass transport in porous media provided in the embodiments of this application;

[0037] Figure 3 This is another flowchart of the method for predicting the physical field of thermal and mass transport in porous media provided in the embodiments of this application;

[0038] Figure 4 This is a structural block diagram of the porous medium thermal and mass transport physical field prediction system provided in the embodiments of this application;

[0039] Figure 5 This is a structural block diagram of the computer device provided in the embodiments of this application. Detailed Implementation

[0040] The present application will now be described in further detail with reference to the embodiments and accompanying drawings. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the application. Furthermore, it should be noted that, for ease of description, only the parts relevant to the application are shown in the accompanying drawings.

[0041] It should be noted that, unless otherwise specified, the embodiments and features of the embodiments in this application can be combined with each other. The present application will now be described in detail with reference to the accompanying drawings and embodiments.

[0042] The following describes in detail, with reference to the accompanying drawings, a method, system, device, and medium for predicting the physical field of thermal and mass transport in porous media according to embodiments of this application.

[0043] This application presents a method, system, equipment, and medium for predicting the physical field of thermal and mass transport in porous media, which solves the problems of high time and high economic cost in obtaining high-precision physical fields of thermal and mass transport in porous media in related technologies.

[0044] Figure 1 This is a flowchart of a method for predicting the physical field of thermal and mass transport in porous media according to an embodiment of this application. Figure 1 As shown, the method for predicting the physical field of thermal and mass transport in porous media according to an embodiment of this application includes the following steps:

[0045] S101: Obtain dimensional input data of the thermal and mass transport process of porous media under the target operating conditions, wherein the dimensional input data includes at least the pore structure, initial and boundary conditions, and physical property parameters.

[0046] The target operating condition is a specified operating condition. That is, the input of the specified operating condition is first preprocessed to determine the dimensional input data of the thermal and mass transport process of the porous medium under the specified operating condition. The dimensional input data includes, but is not limited to, pore structure, initial and boundary conditions, and physical property parameters.

[0047] In one embodiment of this application, preprocessing includes, but is not limited to: cleaning abnormal data, interpolating the pore structure of the porous medium to the required spatial resolution, and processing the binarized pore structure data into a continuous distribution such as Euclidean distance.

[0048] S102: Dimensional input data is used as input data for the prediction model to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model, wherein: the prediction model is pre-trained based on the dimensional input-output data pair obtained from the pre-obtained porous medium heat and mass transport physical field data and the heat and mass transport control equations, the dimensional input-output data pair includes dimensional input data and dimensional output data under the conditions corresponding to the dimensional input data.

[0049] In one embodiment of this application, before using dimensional input data as input to the prediction model to obtain the prediction result of the porous medium's thermal and mass transport physical field corresponding to the target working condition, the method further includes:

[0050] Obtain physical field data of thermal and mass transport in porous media; based on the physical field data of thermal and mass transport in porous media, obtain the dimensional input-output data pair; design the architecture of the initial prediction model based on the physical knowledge of thermal and mass transport. The design of the initial prediction model architecture is guided by the physical knowledge of thermal and mass transport; therefore, the prediction model satisfies certain physical prior constraints, and thus, the output of the prediction model is a physical field prediction result that conforms to physical laws, is interpretable, and has strong generalization.

[0051] Specifically, obtaining high-precision data on the physical fields of thermal and mass transport in porous media is typically achieved through pore-scale experimental methods to acquire phase and velocity field data. Alternatively, existing pore-scale numerical simulation methods can be used to obtain phase, velocity, pressure, temperature, and component concentration field data.

[0052] In the above description, pore-scale experimental methods include, but are not limited to, bright-field observation, particle image velocimetry (PIV), confocal laser scanning microscopy, and stroboscopic X-ray microtomography. Pore-scale numerical simulation methods include, but are not limited to, the volume of fluid (VOF) method, the phase field model (PFM) method, and the lattice Boltzmann method (LBM).

[0053] Based on the physical field data of thermal and mass transport in porous media, dimensional input-output data pairs are obtained. Specifically, high-precision physical field data of thermal and mass transport in porous media are preprocessed to determine the preprocessed dimensional input-output data pairs. Dimensional input data includes, but is not limited to, pore structure, initial and boundary conditions, and physical property parameters. Dimensional output data includes physical field data under the corresponding conditions. In this example, dimensional data preprocessing includes anomaly cleaning, interpolating the pore structure and thermal and mass transport physical field data to the required spatial resolution, and processing the binarized pore structure data into a continuous distribution such as Euclidean distance.

[0054] The architecture of the initial prediction model is designed based on the physical knowledge of thermo-mass transport. Specifically, this initial prediction model, also known as the machine learning model architecture, is designed based on this physical knowledge. This machine learning model architecture design includes, but is not limited to: selection of neural network module types, design of the number and size of hidden layers, design of data flow transmission methods, and design of activation functions. The approach to designing the machine learning model architecture based on the physical knowledge of thermo-mass transport includes, but is not limited to: designing activation functions and neural network modules that satisfy physical prior properties, such as symmetry, nonnegativity, positive definiteness, monotonicity, partial differential operators, and fundamental principles of thermodynamics. In a specific example, for the problem of simultaneous prediction of multiple physics fields, a single or multiple decoder combination can be designed, corresponding to simultaneous decoding of multiple physics fields or decoding of each physics field by a specific decoder.

[0055] In one embodiment of this application, after designing the architecture of the initial prediction model based on the physics of heat and mass transport, the method further includes: training the initial prediction model based on the constraints of the dimensional input-output data pair and the heat and mass transport control equations, until the loss between the output of the initial prediction model and the label meets a predetermined condition, thereby obtaining the trained prediction model, wherein the label is the dimensional output data in the dimensional input-output data pair under the conditions corresponding to the dimensional input data.

[0056] like Figure 2As shown, one implementation process of the method for predicting the physical field of thermal and mass transport in porous media is as follows: acquiring high-precision data of the physical field of thermal and mass transport in porous media; preprocessing the high-precision data of the physical field of thermal and mass transport in porous media to determine the dimensional input-output data pairs after preprocessing; designing a machine learning model architecture based on knowledge of thermal and mass transport physics; then training the machine learning model to determine the model parameters according to the dimensional input-output data pairs and the constraints of the thermal and mass transport control equations; finally, predicting the physical field of thermal and mass transport in porous media based on the trained machine learning model.

[0057] According to the embodiment of this application, the method for predicting the physical field of porous media thermo-mass transport uses dimensional input data of the porous media thermo-mass transport process under specified operating conditions as input data to the prediction model. This allows for rapid prediction of the physical field of porous media thermo-mass transport under the specified operating conditions. This method, employing a machine learning model to predict the physical field of porous media thermo-mass transport, replaces the experimental and numerical simulation methods used in related technologies. Since a trained model can infer the physical field of porous media thermo-mass transport under specified operating conditions without iteration, the prediction speed is effectively improved. Furthermore, the machine learning model has strong representational capabilities, thus ensuring the accuracy of the prediction. In addition, the training of the machine learning model is driven by both data and the thermo-mass transport control equations, ensuring that the prediction results conform to physical laws. The interpretability and generalization of the machine learning model can be enhanced by explicitly incorporating physical knowledge. Moreover, it can obtain high-precision pore-scale pressure and temperature fields that match the characteristics of experimental data, achieving the effect of indirect experimental measurement of pore-scale pressure and temperature fields. Because experimental methods in related technologies struggle to obtain pore-scale pressure and temperature fields, and numerical simulations suffer from simplification, the accuracy of the simulated pressure and temperature fields is limited. Therefore, the embodiments of this application utilize data and equations jointly to drive the machine learning model training phase, enabling the acquisition of pressure and temperature fields that satisfy the constraints of the phase field, velocity field, and the mass, momentum, and energy conservation equations required for experimental measurements. Depending on the characteristics of different physical processes, the pressure and temperature fields can be acquired separately or simultaneously.

[0058] In one embodiment of this application, data can also be organized in a dimensionless input-output format. For physical similarity problems, only one set of training data is needed, reducing the redundancy of training data. During the prediction phase, the specified operating conditions are converted into dimensionless inputs, and the model predicts the dimensionless physical field, which is then converted back into the corresponding dimensional physical field. This process ensures that a model trained on data of one scale can be applied at different scales, exhibiting scale invariance. Furthermore, since physical prior constraints are introduced into the model design and training phases, the dependence on the amount of training data can be further reduced, achieving both physical similarity and scale-invariant physical field prediction. Specifically, this is combined with... Figure 3 As shown, the method for predicting the physical field of thermal and mass transport in porous media, after designing the architecture of the initial prediction model based on the knowledge of thermal and mass transport physics, further includes: dimensionlessizing the dimensional input-output data pairs to determine dimensionless input-output data pairs; training the initial prediction model based on the constraints of the dimensionless input-output data pairs and the thermal and mass transport control equations until the loss between the output of the initial prediction model and the label meets a predetermined condition, thus obtaining the trained prediction model, wherein the label is the dimensionless output data in the dimensionless input-output data pairs under the conditions corresponding to the dimensionless input data.

[0059] Furthermore, the prediction result of the porous medium thermal and mass transport physical field corresponding to the target operating condition obtained by the prediction model is the prediction result of the dimensionless physical field of porous medium thermal and mass transport corresponding to the target operating condition. After obtaining the prediction result of the dimensionless physical field of porous medium thermal and mass transport corresponding to the target operating condition by the prediction model, the method further includes: converting the dimensionless physical field output by the prediction model into a dimensional physical field according to the operating parameters of the target operating condition, so as to obtain the final prediction result of the porous medium thermal and mass transport physical field under the target operating condition.

[0060] In the above example, the dimensional input-output data pairs are made dimensionless to determine the dimensionless input-output data pairs, which is achieved by the following formula:

[0061]

[0062] Where: x, y, z are spatial coordinates, t is time, u, v, w are the fluid velocity components in the x, y, and z directions, respectively, p is the fluid pressure, L is the characteristic dimension, U is the characteristic velocity, and ρ is the fluid density. * indicates a dimensionless physical quantity.

[0063] After dimensionlessly converting the dimensional input-output data pairs to dimensionless values ​​and determining the dimensionless input-output data pairs, the method further includes: standardizing the dimensionless output data using the following formula:

[0064]

[0065] Where std represents the standardized variable, the averaging operator in the denominator represents the arithmetic mean of the data after the porous medium volume averaging operation, and Re represents the characteristic Reynolds number.

[0066] Specifically, the dimensional input-output data pairs are made dimensionless to determine dimensionless input-output data pairs. The dimensionless output data is then standardized to unify the distribution characteristics of different types of physical field data, ensuring that the influence weights of different types of physical fields on the model loss function and the backpropagation gradient optimization process are consistent. This reduces the difficulty of simultaneous training with multiple physical fields and improves prediction accuracy. By designing an end-to-end model mapping using dimensionless methods and guiding the design of dimensionless training datasets with similarity theory, physical similarity prediction can be achieved while reducing dataset redundancy.

[0067] In the above embodiments, the averaging operator is calculated in the following way:

[0068]

[0069] Where ψ represents the general variable that needs to be processed by the averaging operator. The dimensionless volume of the porous media fluid domain is represented by , N represents the number of data entries in the dataset, and i represents the i-th data entry in the dataset. The characteristic Reynolds number is calculated as follows:

[0070]

[0071] Where μ represents the fluid dynamic viscosity. The standardization of the pressure field ensures that the pressure data variation range is on the same order of magnitude as the velocity, improving the training effect of both the pressure and velocity fields participating in model training. Optionally, the dimensionless training dataset design guided by similarity theory is used. For single-phase flow processes in porous media, the dataset is designed according to two dimensions: dimensionless pore structure and Reynolds number. For two-phase flow processes in porous media, the dataset is designed according to six dimensions: dimensionless pore structure, Reynolds number, capillary number, viscosity ratio, density ratio, and contact angle.

[0072] Based on the dimensionless input / output data, the heat and mass transport control equations, and boundary constraints, the machine learning model is trained to determine its parameters. Specifically, the model loss function design method, driven by both data and domain knowledge, is as follows:

[0073] Loss=(1-α-β)Loss data +αLoss equation +βLoss b.c.

[0074] In this context, Loss represents the loss function, and α and β are weighting coefficients. `data` represents data constraints, `equation` represents constraints under the governing equations for heat and mass transport, and `bc` represents boundary condition constraints. Data constraints involve comparing the model output with the labeled data and calculating the loss function using methods such as mean absolute error or mean squared error. Gathering-mass transport governing equation constraints involve substituting the model's output physics field into the governing equations to determine if the output physics field satisfies the fundamental conservation laws and quantitatively calculate the deviation. These constraints include incorporating one or more of the mass conservation equation, momentum conservation equation, energy conservation equation, and component transport equation as regularization terms into the loss function to constrain the model training process.

[0075] For fully connected neural networks, automatic differentiation is used to calculate the derivative terms in the governing equations. For convolutional neural networks, the Sobel operator is used to calculate the derivative terms in the governing equations. Boundary condition constraints involve comparing the boundary values ​​of the model's output physical field with the boundary conditions of the actual physical process, and calculating the loss function using methods such as mean absolute error or mean square error. For second and third type boundary conditions involving derivative calculation, automatic differentiation or the Sobel operator can be used to calculate the boundary derivatives of the model's output physical field. Thermomass transport governing equation constraints and boundary condition constraints do not require labeled data, reducing the model training process's dependence on large-scale training data. Using thermomass transport governing equation constraints ensures that the model's output phase field, velocity field, pressure field, and temperature field satisfy basic conservation laws, reducing the probability of the model predicting non-physical values. When experimental measurements can only provide phase field and velocity field data, introducing governing equation constraints ensures that the model's output pressure and temperature fields match the characteristics of the experimental data, achieving high-precision prediction of multiple types of physical fields.

[0076] The dimensionless physical field of thermal and mass transport in porous media is predicted based on a trained machine learning model. During the prediction phase, the pore structure, initial and boundary conditions, and physical property parameters for a specified operating condition are converted into dimensionless form and input into the trained machine learning model. The model then performs inference operations to output the predicted dimensionless physical field.

[0077] Then, based on the actual operating parameters, the dimensionless physical field is converted into a dimensional physical field to obtain the prediction results required by the user under the specified operating conditions. The dimensional physical field can support subsequent optimization design and dynamic control. The process of converting the dimensionless physical field into a dimensional physical field is the reverse process of dimensionlessization and standardization.

[0078] Therefore, this approach can improve the prediction speed of physical fields, ensure that the prediction results are consistent with physical laws, interpretable, and strongly generalizable, and acquire high-precision pore-scale pressure and temperature fields that match the characteristics of experimental data. It also achieves the effect of indirect experimental measurement of pore-scale pressure and temperature fields. Furthermore, it guarantees physical similarity and scale invariance in physical field prediction, reduces dependence on training data volume, and organizes data in a dimensionless input-output format. For physical similarity problems, only one set of training data is needed, reducing training data redundancy. In the prediction phase, the specified operating conditions are converted into dimensionless inputs, and the model predicts the dimensionless physical field, which is then converted back into the corresponding dimensional physical field. This process ensures that a model trained on data of one scale can be applied at different scales, exhibiting scale invariance. Moreover, since physical prior constraints are introduced during the model design and training phases, the dependence on the amount of training data can be further reduced.

[0079] Figure 4 This is a structural block diagram of a porous medium thermal-mass transport physical field prediction system according to an embodiment of this application. Figure 4 As shown, the porous medium thermal-mass transport physical field prediction system according to an embodiment of this application includes: an acquisition module 410 and a prediction module 420, wherein:

[0080] The acquisition module 410 is used to acquire dimensional input data of the thermal and mass transport process of porous media under the target working condition, wherein the dimensional input data includes at least pore structure, initial boundary conditions and physical property parameters.

[0081] The prediction module 420 is used to take the dimensional input data as the input data of the prediction model, so as to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model. The prediction model is pre-trained based on the dimensional input-output data pair obtained from the pre-obtained porous medium heat and mass transport physical field data and the heat and mass transport control equation. The dimensional input-output data pair includes dimensional input data and dimensional output data under the conditions corresponding to the dimensional input data.

[0082] The porous media thermo-mass transport physical field prediction system according to embodiments of this application uses dimensional input data of the porous media thermo-mass transport process under specified operating conditions as input data to the prediction model. This allows for rapid prediction of the porous media thermo-mass transport physical field under the specified operating conditions. This method of using a machine learning model to predict the porous media thermo-mass transport physical field replaces the experimental and numerical simulation methods used in related technologies. Since a trained model can infer the porous media thermo-mass transport physical field under specified operating conditions without iteration, the prediction speed is effectively improved. Furthermore, the machine learning model has strong representational capabilities, thus ensuring the prediction accuracy of the physical field. In addition, the training of the machine learning model is driven by both data and the thermo-mass transport control equations, ensuring that the prediction results conform to physical laws. The interpretability and generalization of the machine learning model can be enhanced by explicitly incorporating physical knowledge. Moreover, it can acquire high-precision pore-scale pressure and temperature fields that match the characteristics of experimental data, achieving the effect of indirect experimental measurement of pore-scale pressure and temperature fields. Because experimental methods in related technologies struggle to obtain pore-scale pressure and temperature fields, and numerical simulations suffer from simplification, the accuracy of the simulated pressure and temperature fields is limited. Therefore, the embodiments of this application utilize data and equations jointly to drive the machine learning model training phase, enabling the acquisition of pressure and temperature fields that satisfy the constraints of the phase field, velocity field, and the mass, momentum, and energy conservation equations required for experimental measurements. Depending on the characteristics of different physical processes, the pressure and temperature fields can be acquired separately or simultaneously.

[0083] Specific limitations regarding the porous media thermal and mass transport physical field prediction system can be found in the limitations of the porous media thermal and mass transport physical field prediction method described above, and will not be repeated here. Each module of the aforementioned porous media thermal and mass transport physical field prediction system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0084] In one embodiment, a computer device is provided. Figure 5 This is a structural block diagram of the computer device provided in the embodiments of this application. The computer device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned embodiment of the method for predicting the physical field of porous medium thermal and mass transport. For example, it executes: obtaining dimensional input data of the porous medium thermal and mass transport process under the target operating condition, wherein the dimensional input data includes at least the pore structure, initial boundary conditions, and physical property parameters;

[0085] The dimensional input data is used as the input data to the prediction model, so as to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model, wherein:

[0086] The prediction model is pre-trained based on dimensional input-output data pairs obtained from pre-acquired porous medium thermo-mass transport physical field data and thermo-mass transport control equations. The dimensional input-output data pairs include dimensional input data and dimensional output data under conditions corresponding to the dimensional input data.

[0087] This application also provides a computer-readable storage medium storing a computer program. When the processor executes the computer program, it implements the aforementioned embodiment of the method for predicting the physical field of porous medium thermal and mass transport. For example, it executes: obtaining dimensional input data of the porous medium thermal and mass transport process under target operating conditions, wherein the dimensional input data includes at least pore structure, initial boundary conditions, and physical property parameters;

[0088] The dimensional input data is used as the input data to the prediction model, so as to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model, wherein:

[0089] The prediction model is pre-trained based on dimensional input-output data pairs obtained from pre-acquired porous medium thermo-mass transport physical field data and thermo-mass transport control equations. The dimensional input-output data pairs include dimensional input data and dimensional output data under conditions corresponding to the dimensional input data.

[0090] This application provides a computer program product including instructions that, when executed, cause the method described in this application embodiment to be performed. For example, it can execute... Figure 1 The steps of the method for predicting the physical field of thermal and mass transport in porous media shown are, for example, performing the following: obtaining dimensional input data of the thermal and mass transport process in porous media under the target operating condition, wherein the dimensional input data includes at least the pore structure, initial and boundary conditions, and physical property parameters;

[0091] The dimensional input data is used as the input data to the prediction model, so as to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model, wherein:

[0092] The prediction model is pre-trained based on dimensional input-output data pairs obtained from pre-acquired porous medium thermo-mass transport physical field data and thermo-mass transport control equations. The dimensional input-output data pairs include dimensional input data and dimensional output data under conditions corresponding to the dimensional input data.

[0093] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0094] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0095] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for predicting the physical field of thermal and mass transport in porous media, characterized in that, include: Dimensional input data of the thermal and mass transport process of porous media under target operating conditions are obtained, wherein the dimensional input data includes at least pore structure, initial and boundary conditions and physical property parameters; The dimensional input data is used as the input data to the prediction model, so as to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model, wherein: The prediction model is pre-trained based on dimensional input-output data pairs obtained from pre-acquired physical field data of porous media for heat and mass transport, and constrained by the heat and mass transport control equations. The dimensional input-output data pairs include dimensional input data and dimensional output data under conditions corresponding to the dimensional input data. The prediction model is obtained as follows: Dimensional input-output data pairs are rendered dimensionless to determine dimensionless input-output data pairs; based on the dimensionless input-output data pairs and the heat and mass transport control equations, the initial prediction model is trained until the loss between the output of the initial prediction model and the label meets predetermined conditions, thus obtaining the trained prediction model. The label is the dimensionless output data in the dimensionless input-output data pair under the condition corresponding to the dimensionless input data.

2. The method for predicting the physical field of thermal and mass transport in porous media according to claim 1, characterized in that, Before using the dimensional input data as input to the prediction model to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition, the method further includes: Obtain physical field data of thermal and mass transport in porous media; Based on the physical field data of the porous medium's thermal and mass transport, the dimensional input-output data pair is obtained; The architecture of the initial prediction model was designed based on the physics of heat and mass transport.

3. The method for predicting the physical field of thermal and mass transport in porous media according to claim 2, characterized in that, After designing the architecture of the initial prediction model based on the physics of heat and mass transport, it also includes: The initial prediction model is trained based on the constraints of the dimensional input-output data pair and the heat and mass transport control equation until the loss between the output of the initial prediction model and the label meets a predetermined condition, thus obtaining the trained prediction model. The label is the dimensional output data in the dimensional input-output data pair under the conditions corresponding to the dimensional input data.

4. The method for predicting the physical field of thermal and mass transport in porous media according to claim 1, characterized in that, in, The prediction result of the porous medium thermal and mass transport physical field corresponding to the target operating condition obtained through the prediction model is the prediction result of the dimensionless physical field of porous medium thermal and mass transport corresponding to the target operating condition. After obtaining the prediction result of the porous medium thermal and mass transport dimensionless physical field corresponding to the target operating condition through the prediction model, the method further includes: Based on the operating parameters of the target operating condition, the dimensionless physical field output by the prediction model is converted into a dimensional physical field to obtain the final prediction result of the thermal and mass transport physical field of the porous medium under the target operating condition.

5. The method for predicting the physical field of thermal and mass transport in porous media according to claim 1 or 4, characterized in that, The step of dimensionlessizing the dimensional input-output data pairs to determine the dimensionless input-output data pairs is achieved through the following formula: in: x , y , z For spatial coordinates, t For time, u , v , w They are respectively x , y , z The fluid velocity component in the direction, p For fluid pressure, L For feature size, U Characteristic velocity, ρ The fluid density is represented by *, which indicates a dimensionless physical quantity.

6. The method for predicting the physical field of thermal and mass transport in porous media according to claim 5, characterized in that, After dimensionless transformation of the dimensional input-output data pairs to determine the dimensionless input-output data pairs, the process further includes: The dimensionless output data is standardized using the following formula: Where std represents the standardized variable, and the averaging operator in the denominator represents the arithmetic mean of the data after the porous media volume averaging operation. Re This represents the characteristic Reynolds number.

7. A system for predicting the physical field of thermal and mass transport in porous media, characterized in that, include: The acquisition module is used to obtain dimensional input data of the thermal and mass transport process of porous media under target working conditions, wherein the dimensional input data includes at least pore structure, initial boundary conditions and physical property parameters; The prediction module is used to take the dimensional input data as input to the prediction model, so as to obtain the prediction result of the porous medium heat and mass transport physical field corresponding to the target working condition through the prediction model. The prediction model is pre-trained based on dimensional input-output data pairs obtained from pre-acquired porous medium heat and mass transport physical field data, and constrained by the heat and mass transport control equations. The dimensional input-output data pairs include dimensional input data and dimensional output data under conditions corresponding to the dimensional input data. The prediction model is obtained as follows: Dimensional input-output data pairs are rendered dimensionless to determine dimensionless input-output data pairs; based on the dimensionless input-output data pairs and the heat and mass transport control equations, the initial prediction model is trained until the loss between the output of the initial prediction model and the label meets predetermined conditions, thus obtaining the trained prediction model. The label is the dimensionless output data in the dimensionless input-output data pair under the condition corresponding to the dimensionless input data.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for predicting the physical field of thermal and mass transport in porous media according to any one of claims 1-6.

9. A computationally readable storage medium, comprising a memory and a computer program stored on the memory and executable on a processor, characterized in that, When the program is executed by the processor, it implements the method for predicting the physical field of thermal and mass transport in porous media according to any one of claims 1-6.

Images