Fast calculation method and related device for thermal-fluid coupling field of power equipment

Abstract

The present invention belongs to the technical field of insulation structure design for power equipment and discloses a method and related device for rapidly calculating the thermal-fluid coupling field of power equipment. The method obtains unknown operating conditions; uses a trained KAN neural network proxy model to process the unknown operating conditions to obtain characteristic coefficients corresponding to the unknown operating conditions; and uses calculated orthogonal basis vectors and characteristic coefficients corresponding to the unknown operating conditions to calculate the thermal-fluid coupling field of the power equipment corresponding to the unknown operating conditions. The orthogonal basis vector calculation process includes: establishing a simulation model for the thermal-fluid coupling field of the power equipment and calculating simulation results corresponding to different operating conditions; constructing a sample set based on the simulation results; and performing intrinsic orthogonal decomposition on the sample set to obtain characteristic coefficients and orthogonal basis vectors corresponding to the sample set under different operating conditions. The present invention can meet the needs of rapidly calculating the thermal-fluid coupling field of power equipment, improve computational efficiency, and ensure the accuracy of the calculation results.

Description

Technical Field

[0001] The present invention belongs to the technical field of insulation structure design of electric power equipment, and in particular relates to a method for quickly calculating a thermal-fluid coupling field of electric power equipment and a related device. Background Art

[0002] In recent years, with the large-scale cross-research and application of modern information technologies and energy technologies such as big data, cloud computing, and artificial intelligence, digital technology has developed comprehensively in many fields. In order to promote the digital and intelligent transformation of energy systems, it is necessary to support the construction of new power systems with digital and intelligent power grids, accelerate the development of energy equipment status identification, reliability assessment and fault diagnosis technologies based on artificial intelligence, and enhance on-site perception, computing and data transmission interaction capabilities.

[0003] With the construction and development of smart grids, big data and digital twin technologies have seen initial application in power equipment status monitoring and analysis, fault identification and diagnosis, and the power Internet of Things. The gradual development of various information collection systems and new intelligent data analysis platforms within power grid companies has laid a solid foundation for the development and application of digital technology for power equipment. However, digital technology for power equipment is still primarily at the stage of offline modeling, simulation, and visualization. Problems persist, such as limited sensor distribution, poor timeliness in acquiring field information, and limited intelligent data analysis and processing capabilities. This makes it difficult to achieve comprehensive, real-time, and accurate perception of power equipment status. Further research is urgently needed on core technologies for power equipment digitalization, such as intelligent sensing and digital twins. A digital twin is a comprehensive virtual mapping of a physical entity across time and space, enabling real-time simulation and precise prediction of its status information in a virtual space. Therefore, digital twin technology is key to rapidly and accurately acquiring field information and estimating the lifespan of power equipment.

[0004] However, the low computational efficiency of multi-physics coupled simulation is a serious problem that hinders the visualization and real-time evaluation and analysis of the entire information of power equipment, and is also a difficult problem for the application of digital twin technology in the field of digital power equipment. The core of the digitalization of power equipment is the visualization and evaluation of status data throughout its life cycle. Digital twin technology, with multi-physics simulation technology as one of its core technologies, is the key to the visualization of internal characteristics of current equipment. With the continuous improvement of the digitalization and intelligence level of power equipment, the requirements for the timeliness and accuracy of multi-physics simulation are also increasing. The method of focusing on multi-physics simulation and integrating a small amount of sensor data to correct the simulation results is the general trend. Power equipment is large in size, has a large number of grid nodes, and is subject to the influence of multiple nonlinear physical fields. As a result, the solution of multi-physics simulation is time-consuming and occupies a large amount of computing resources. Currently, offline simulation calculations are mainly used, which makes it difficult to achieve real-time and accurate calculations. Power equipment, such as transformers, GIS (Gas-Insulated Switchgear), or GIL (Gas-Insulated Transmission Line), is characterized by complex structures, numerous components, widely varying material properties, and demanding operating conditions. It is therefore crucial to develop fast multi-physics simulation methods for power equipment and achieve fast, second-level calculations of single and multiple fields within key internal components. This approach can provide technical support for establishing intrinsic safety systems for power equipment. Summary of the Invention

[0005] In order to solve the problems existing in the prior art, the purpose of the present invention is to provide a method and related devices for quickly calculating the thermal-fluid coupling field of power equipment. The present invention can meet the needs of quickly calculating the thermal-fluid coupling field of power equipment, while improving the calculation efficiency and ensuring the accuracy of the calculation results.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The rapid calculation method of thermal-fluid coupled field of power equipment includes the following steps:

[0008] Obtain unknown working conditions;

[0009] Use the trained KAN neural network proxy model (Kolmogorov-Arnold Network) to process unknown working conditions and obtain the characteristic coefficients corresponding to the unknown working conditions;

[0010] The calculated orthogonal basis vectors and the characteristic coefficients corresponding to the unknown working conditions are used to calculate the thermal-fluid coupling field of the power equipment corresponding to the unknown working conditions; wherein the calculation process of the orthogonal basis vectors includes: establishing a simulation model of the thermal-fluid coupling field of the power equipment; using the simulation model of the thermal-fluid coupling field of the power equipment, calculating the simulation results of the thermal-fluid coupling field of the power equipment corresponding to different working conditions; constructing a sample set through the simulation results of the thermal-fluid coupling field of the power equipment; performing intrinsic orthogonal decomposition on the sample set to obtain the characteristic coefficients and the orthogonal basis vectors of the sample set corresponding to different working conditions.

[0011] Preferably, the simulation result of the thermal-fluid coupling field of the power equipment is temperature or flow velocity.

[0012] Preferably, the input of the KAN neural network agent model is the operating condition, and the output is the characteristic coefficient;

[0013] When the simulation result of the thermal-fluid coupling field of the power equipment is temperature, the number of neurons in the output layer of the KAN neural network proxy model is the same as the dimension of the characteristic coefficient corresponding to the temperature;

[0014] When the simulation result of the thermal-fluid coupling field of the power equipment is the flow velocity, the number of neurons in the output layer of the KAN neural network agent model is the same as the dimension of the characteristic coefficient corresponding to the flow velocity.

[0015] Preferably, the step of using the thermal-fluid coupled field simulation model of the power equipment to calculate the thermal-fluid coupled field simulation results of the power equipment corresponding to different operating conditions includes:

[0016] The finite element numerical calculation method or the finite volume numerical calculation method is used to calculate the thermal-fluid coupling field simulation model of the power equipment to obtain the thermal-fluid coupling field simulation results of the power equipment corresponding to different working conditions.

[0017] Preferably, the step of calculating the thermal-fluid coupling field of the power equipment corresponding to the unknown operating condition using the calculated orthogonal basis vectors and the characteristic coefficients corresponding to the unknown operating condition includes:

[0018] The characteristic coefficient corresponding to the unknown working condition is multiplied by the orthogonal basis vector, and then added with the average value of the sample set to obtain the thermal-fluid coupling field of the power equipment corresponding to the unknown working condition.

[0019] Preferably, the power equipment is a transformer winding, a transformer bushing, a gas-insulated switchgear or a gas-insulated transmission line.

[0020] Preferably: when the power equipment is a transformer winding, the operating conditions include the temperature of the oil channel inlet, the flow rate of the oil channel inlet and the heat source density of the winding;

[0021] When the power equipment is a transformer bushing, a gas-insulated switchgear or a gas-insulated transmission line, the operating conditions include the center current-carrying conductor current and the ambient temperature.

[0022] The present invention also provides a system for implementing the above-mentioned method for rapidly calculating the thermal-fluid coupling field of power equipment, comprising:

[0023] Data acquisition module: used to obtain unknown working conditions;

[0024] The first calculation module is used to process unknown working conditions using the trained KAN neural network proxy model to obtain characteristic coefficients corresponding to the unknown working conditions;

[0025] The second calculation module is used to calculate the thermal-fluid coupling field of the power equipment corresponding to the unknown working conditions using the calculated orthogonal basis vectors and the characteristic coefficients corresponding to the unknown working conditions; wherein the calculation process of the orthogonal basis vectors includes: establishing a simulation model of the thermal-fluid coupling field of the power equipment; using the simulation model of the thermal-fluid coupling field of the power equipment to calculate the simulation results of the thermal-fluid coupling field of the power equipment corresponding to different working conditions; constructing a sample set based on the simulation results of the thermal-fluid coupling field of the power equipment; performing intrinsic orthogonal decomposition on the sample set to obtain the characteristic coefficients of the sample set corresponding to different working conditions and the orthogonal basis vectors.

[0026] The present invention also provides an electronic device, comprising:

[0027] one or more processors;

[0028] a storage device having one or more programs stored thereon;

[0029] When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the method for quickly calculating the thermal-fluid coupled field of an electric power device as described above in the present invention.

[0030] The present invention further provides a storage medium storing a computer program, wherein when the computer program is executed by a processor, the method for quickly calculating the thermal-fluid coupling field of an electric power device as described above is implemented in the present invention.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present invention's method for rapidly calculating the thermal-fluid coupled field of power equipment takes into account that the input and output parameter dimensions of the KAN neural network proxy model should not be excessive, as this will affect training accuracy and efficiency. To improve calculation speed, the sample set data is reduced in dimensionality through proper orthogonal decomposition (i.e., POD decomposition). Rapid calculation of the thermal-fluid coupled field of power equipment is achieved using the KAN neural network proxy model, significantly reducing thermal field computation time while ensuring that the impact on calculation accuracy is within a certain controllable range. Furthermore, in the present invention's method for rapidly calculating the thermal-fluid coupled field of power equipment, the KAN neural network proxy model is used to replace the traditional MLP (Multilayer Perceptron) architecture neural network. The learnable B-spline (Basis Function) in the KAN neural network proxy model is used to replace weight parameters. These B-spline functions can learn patterns between data and, compared to the fixed activation functions of BP (Back Propagation) neural networks, offer greater flexibility and adaptability to multi-physics field learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a grid decomposition diagram of a 220kV transformer winding in an embodiment of the present invention;

[0034] Figure 2 are the eigenvalues ​​and contribution rates of the snapshot matrix of the 220 kV transformer winding temperature field in an embodiment of the present invention;

[0035] Figure 3 are the eigenvalues ​​and contribution rates of the snapshot matrix of the 220 kV transformer winding velocity field in an embodiment of the present invention;

[0036] Figure 4 This is a cloud diagram of the temperature field of a 220kV transformer winding calculated using a finite element simulation method in an embodiment of the present invention;

[0037] Figure 5 This is a cloud diagram of the temperature field of a 220kV transformer winding obtained by using the method for quickly calculating the thermal-fluid coupling field of power equipment according to an embodiment of the present invention;

[0038] Figure 6 This is a cloud diagram of the 220kV transformer winding velocity field calculated using the finite element simulation method in an embodiment of the present invention;

[0039] Figure 7 This is a cloud diagram of the 220kV transformer winding velocity field obtained by using the method for rapid calculation of thermal-fluid coupling field of power equipment according to an embodiment of the present invention;

[0040] Figure 8This is a comparison diagram of the reduced-order model errors of a 220kV transformer winding temperature field test set obtained by using a BP neural network and a method for quickly calculating the thermal-fluid coupled field of power equipment according to an embodiment of the present invention;

[0041] Figure 9 This is a comparison diagram of the reduced-order model errors of the 220kV transformer winding velocity field test set obtained by using the BP neural network and the fast calculation method for the thermal-fluid coupled field of the power equipment of the present invention in an embodiment of the present invention.

[0042] In the figure, 1-oil channel inlet, 2-first oil channel outlet, 3-second oil channel outlet, 4-medium voltage winding, 5-high voltage winding, 6-baffle, 7-transformer oil. DETAILED DESCRIPTION

[0043] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0044] The method for quickly calculating the thermal-fluid coupling field of electric power equipment of the present invention comprises the following steps:

[0045] Step 1: Establish a thermal-fluid coupling field simulation model for power equipment;

[0046] Step 2: Using a finite element or finite volume numerical calculation method on the thermal-fluid coupled field simulation model of the power equipment established in step 1, calculate multiple sets of thermal-fluid coupled field simulation results of the power equipment corresponding to different operating conditions; wherein the thermal-fluid coupled field simulation results of the power equipment are temperatures or flow rates; and the data of the multiple sets of different operating conditions are obtained by a Latin hypercube sampling method;

[0047] Step 3: construct a sample set based on the thermal-fluid coupled field simulation results of the power equipment obtained in step 2, where the sample set is a sample matrix;

[0048] Step 4: Perform POD decomposition (i.e., proper orthogonal decomposition) on the sample set to obtain the characteristic coefficients and orthogonal basis vectors of the sample set corresponding to different working conditions;

[0049] Step 5: Using the working conditions as input and the characteristic coefficients as output, a KAN neural network proxy model is constructed. The mapping relationship between the working conditions and the characteristic coefficients can be obtained through the KAN neural network proxy model; the constructed KAN neural network proxy model is trained to obtain a trained KAN neural network proxy model;

[0050] Step 6: Input the unknown working condition into the KAN neural network agent model trained in step 5 to obtain the characteristic coefficient corresponding to the unknown working condition.

[0051] Step 7: Using the characteristic coefficient corresponding to the unknown operating condition obtained in step 6 and the orthogonal basis vectors obtained in step 4, the thermal-fluid coupling field of the power equipment corresponding to the unknown operating condition is calculated. The specific calculation process is as follows:

[0052] Multiply the characteristic coefficient corresponding to the unknown operating condition obtained in step 6 by the orthogonal basis vector obtained in step 4, and then add the multiplication result to the average value of the sample set constructed in step 3 to obtain the thermal-fluid coupling field of the power equipment corresponding to the unknown operating condition.

[0053] In the above solution of the present invention, when the simulation result of the thermal-fluid coupling field of the power equipment is temperature, the operating conditions are used as input and the characteristic coefficients are used as output. When constructing the KAN neural network proxy model, the number of neurons in the output layer of the KAN neural network proxy model is the same as the dimension of the characteristic coefficient corresponding to the temperature;

[0054] When the simulation result of the thermal-fluid coupling field of the power equipment is the flow velocity, the operating conditions are used as input and the characteristic coefficients are used as output. When the KAN neural network proxy model is constructed, the number of neurons in the output layer of the KAN neural network proxy model is the same as the dimension of the characteristic coefficients corresponding to the flow velocity.

[0055] In the above solution of the present invention, the power equipment may be a transformer winding, a transformer bushing, a GIS or a GIL;

[0056] When the power equipment is a transformer winding, the operating conditions include the temperature of the oil channel inlet, the flow rate of the oil channel inlet, and the heat source density of the winding;

[0057] When the power equipment is a transformer bushing, GIS or GIL, the operating conditions include the current of the central current-carrying conductor and the ambient temperature.

[0058] Example 1

[0059] This embodiment uses the above method to quickly calculate the thermal-fluid coupling field of a 220 kV transformer winding, including the following steps:

[0060] Step 1): Use CAD software to establish a thermal two-dimensional finite element analysis model of a 220kV transformer, and divide the thermal two-dimensional finite element analysis model into grids. The thermal two-dimensional finite element analysis model grid has a total of 289415 nodes and 246955 units. Figure 1 As shown; the thermal performance parameters of the material required for calculation are set for the thermal two-dimensional finite element analysis model, wherein the thermal performance parameters include thermal conductivity and convection heat transfer coefficient.

[0061] Step 2): Set the constraints required for the calculation. Since the 220kV transformer winding under study is an oil natural circulation structure, the oil flow inside the winding completely relies on the natural convection generated by the combined action of thermal expansion and contraction of the fluid and gravity. In the thermal two-dimensional finite element analysis model of the 220kV transformer winding, the flow velocity boundary and pressure boundary are set at the oil channel inlet 1 and the oil channel outlet (including the first oil channel outlet 2 and the second oil channel outlet 3), respectively. The outer boundary of the fluid except the oil channel inlet 1, the first oil channel outlet 2 and the second oil channel outlet 3 is set to adiabatic, and the thermal-fluid coupling field simulation model of the 220kV transformer winding used for calculation is obtained.

[0062] Step 3): Using the Latin Hypercube sampling method, we selected the four most significant factors affecting winding temperature: oil temperature at oil channel inlet 1, oil flow rate at oil channel inlet 1, heat source density of high-voltage winding 5, and heat source density of medium-voltage winding 4. This formed a sample space. We generated 100 sets of training data and 30 sets of test data.

[0063] Step 4): Use the simulation software Fluent to simulate the above 100 sets of training data to obtain 100 sets of temperature field result data and flow rate field result data of 220kV transformer windings. According to the temperature field result data and flow rate field result data, a temperature snapshot matrix and a flow rate field snapshot matrix are formed respectively. POD decomposition is performed on the temperature snapshot matrix and the flow rate field snapshot matrix respectively to obtain the corresponding eigenvalues ​​and contribution rates, as shown in the following example: Figure 2 and Figure 3 shown.

[0064] Step 5): 100 sets of working conditions obtained by simulation are input into the KAN neural network proxy model. The output of the KAN neural network proxy model is the characteristic coefficients corresponding to different sets of working conditions. Through training and learning, based on the KAN neural network proxy model, a mapping relationship between different construction working conditions and characteristic coefficients is established. The KAN neural network proxy model is trained 10,000 epochs (i.e., 10,000 times), and the batch size is set to 32 and the Adam optimizer is used. After training, a trained KAN neural network proxy model is obtained.

[0065] Step 6): Under working conditions (i.e. the temperature of oil channel inlet 1 T in =323.1K, flow rate at oil channel inlet 1 v =0.127m / s, power of high voltage winding 5 Q h =155120 W / m 3 , the power of medium voltage winding 4 Q m =91191 W / m 3), the KAN neural network agent model trained in step 5) is used to predict the temperature field distribution and flow velocity field distribution of the 220kV transformer winding.

[0066] from Figures 4 to 7 It can be seen that the predicted temperature results (quick calculation) near the 220kV transformer winding using the KAN neural network proxy model of the present invention and the temperature results calculated by finite element simulation are both between 314K-335K, and the flow velocity field results are between 0m / s-0.165m / s. The physical field distribution is basically consistent, and the error is mainly concentrated in the position near the first oil channel outlet 2 of the winding. Figure 8 and Figure 9 It can be seen that the average relative error of the temperature field predicted by the BP neural network is mostly between 0.5% and 1.1%, and the maximum absolute error is mostly between 0.1K and 1.2K, while the average relative error of the temperature field predicted by the KAN neural network proxy model of the present invention is mostly between 0.48% and 0.9%, and the maximum absolute error is mostly between 0.1K and 0.4K. The average relative error of the flow velocity field predicted by the BP neural network is mostly between 2% and 7.5%, and the maximum absolute error is mostly between 0.0002m / s and 0.00125m / s, while the average relative error of the temperature field predicted by the KAN neural network proxy model of the present invention is mostly between 2% and 7%, and the maximum absolute error is mostly between 0.0001m / s and 0.0004m / s. It can be seen that the overall prediction errors of the temperature field and flow velocity field of the KAN neural network proxy model of the present invention are smaller than those of the temperature field and flow velocity field of the BP neural network, verifying the rationality and accuracy of the fast calculation method of the thermal-fluid coupling field of power equipment based on the KAN neural network proxy model of the present invention.

[0067] In addition, an embodiment of the present invention further provides a system for implementing the above-mentioned method for rapidly calculating the thermal-fluid coupling field of an electric power device of the present invention, the system comprising:

[0068] Data acquisition module: used to obtain unknown working conditions;

[0069] A first calculation module is configured to process the unknown working condition using the trained KAN neural network proxy model to obtain a characteristic coefficient corresponding to the unknown working condition;

[0070] The second calculation module is used to calculate the thermal-fluid coupling field of the power equipment corresponding to the unknown operating conditions using the calculated orthogonal basis vectors and the characteristic coefficients corresponding to the unknown operating conditions; wherein the calculation process of the orthogonal basis vectors includes: establishing a simulation model of the thermal-fluid coupling field of the power equipment; using the simulation model of the thermal-fluid coupling field of the power equipment to calculate the simulation results of the thermal-fluid coupling field of the power equipment corresponding to different operating conditions; constructing a sample set based on the simulation results of the thermal-fluid coupling field of the power equipment; performing intrinsic orthogonal decomposition on the sample set to obtain the characteristic coefficients and orthogonal basis vectors of the sample set corresponding to different operating conditions.

[0071] The embodiments of the present invention also provide corresponding electronic devices and computer-readable storage media for implementing the solutions provided by the embodiments of the present invention.

[0072] The device includes a memory and a processor, the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the method for rapid calculation of the thermal-fluid coupling field of the power equipment described in any embodiment of the present application.

[0073] The storage medium stores a computer program, wherein when the computer program is executed by the processor, the method for quickly calculating the thermal-fluid coupling field of an electric power device described in any embodiment of the present application is implemented.

[0074] Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A fast calculation method for thermal-fluid coupling field of power equipment, characterized by: The process includes the following: Obtain unknown working conditions; Using the trained KAN neural network proxy model to process unknown operating conditions to obtain characteristic coefficients corresponding to the unknown operating conditions; wherein the KAN neural network proxy model is constructed with the operating conditions as input and the characteristic coefficients as output; The characteristic coefficient corresponding to the unknown operating condition is multiplied by the calculated orthogonal basis vector, and then added with the average value of the sample set to calculate the thermal-fluid coupling field of the power equipment corresponding to the unknown operating condition; wherein, the calculation process of the orthogonal basis vector includes: establishing a simulation model of the thermal-fluid coupling field of the power equipment; using the simulation model of the thermal-fluid coupling field of the power equipment to calculate the simulation results of the thermal-fluid coupling field of the power equipment corresponding to different operating conditions; constructing a sample set based on the simulation results of the thermal-fluid coupling field of the power equipment; performing intrinsic orthogonal decomposition on the sample set to obtain the characteristic coefficients and the orthogonal basis vectors of the sample set corresponding to different operating conditions.

2. The method for rapid calculation of thermal-fluid coupling field of power equipment according to claim 1, characterized in that: The simulation results of thermal-fluid coupled fields of power equipment are temperature or flow velocity.

3. The method for rapid calculation of thermal-fluid coupling field of power equipment according to claim 2, characterized in that: When the simulation result of the thermal-fluid coupling field of the power equipment is temperature, the number of neurons in the output layer of the KAN neural network proxy model is the same as the dimension of the characteristic coefficient corresponding to the temperature; When the simulation result of the thermal-fluid coupling field of the power equipment is the flow velocity, the number of neurons in the output layer of the KAN neural network agent model is the same as the dimension of the characteristic coefficient corresponding to the flow velocity.

4. The method for rapid calculation of thermal-fluid coupling field of power equipment according to claim 2, characterized in that: The step of using the thermal-fluid coupled field simulation model of the power equipment to calculate the thermal-fluid coupled field simulation results of the power equipment corresponding to different operating conditions includes: The finite element numerical calculation method or the finite volume numerical calculation method is used to calculate the thermal-fluid coupling field simulation model of the power equipment to obtain the thermal-fluid coupling field simulation results of the power equipment corresponding to different working conditions.

5. The method for rapid calculation of thermal-fluid coupling field of power equipment according to claim 1, characterized in that: The power equipment is transformer windings, transformer bushings, gas-insulated switchgear or gas-insulated transmission lines.

6. The method for rapid calculation of thermal-fluid coupling field of power equipment according to claim 5, characterized in that: When the power equipment is a transformer winding, the operating conditions include the temperature at the oil channel inlet, the flow rate at the oil channel inlet, and the heat source density of the winding; When the power equipment is a transformer bushing, a gas-insulated switchgear or a gas-insulated transmission line, the operating conditions include the center current-carrying conductor current and the ambient temperature.

7. Rapid calculation system for thermal-fluid coupling field of power equipment, characterized by: include: Data acquisition module: used to obtain unknown working conditions; A first calculation module is configured to process unknown operating conditions using a trained KAN neural network proxy model to obtain characteristic coefficients corresponding to the unknown operating conditions; wherein the KAN neural network proxy model is constructed with the operating conditions as input and the characteristic coefficients as output; The second calculation module: multiply the characteristic coefficient corresponding to the unknown working condition by the calculated orthogonal basis vector, and add the average value of the sample set to calculate the thermal-fluid coupling field of the power equipment corresponding to the unknown working condition; wherein, the calculation process of the orthogonal basis vector includes: establishing a simulation model of the thermal-fluid coupling field of the power equipment; using the simulation model of the thermal-fluid coupling field of the power equipment to calculate the simulation results of the thermal-fluid coupling field of the power equipment corresponding to different working conditions; constructing a sample set based on the simulation results of the thermal-fluid coupling field of the power equipment; performing intrinsic orthogonal decomposition on the sample set to obtain the characteristic coefficients and the orthogonal basis vectors of the sample set corresponding to different working conditions.

8. An electronic device, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method for quickly calculating the thermal-fluid coupled field of an electric power device according to any one of claims 1 to 6.

9. A storage medium, characterized in that: A computer program is stored thereon, wherein when the computer program is executed by a processor, the method for quickly calculating the thermal-fluid coupling field of an electric power equipment as claimed in any one of claims 1 to 6 is implemented.

Images