Electric reactor field-circuit coupling simulation calculation method and system and storage medium

Abstract

The invention discloses an electric reactor field-circuit coupling simulation calculation method and system and a storage medium. The method comprises the steps that an electromagnetic field model and a circuit model of a to-be-analyzed electric reactor are determined; based on the electromagnetic field model and the circuit model, performing resonant frequency analysis and field-circuit interaction prediction by using an artificial intelligence model; determining a field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis result and the field-circuit interaction prediction result; and calculating according to the field-circuit coupling model by using a boundary element method to obtain a calculation result, thereby shortening the calculation time and reducing the consumption of calculation resources under the condition of ensuring the simulation calculation precision.

Description

Technical Field

[0001] The present application relates to the field of simulation technology, and in particular to a method, system and storage medium for simulating and calculating field-circuit coupling of a reactor. Background Art

[0002] In the power system, as an important electrical component, the reactor usually has the functions of controlling reactive power, filtering harmonics, and improving system stability. Specifically, the reactor is widely used in current regulation, filtering, power factor control, etc. through its unique inductance characteristics. Whether in large-scale power transmission and conversion equipment or in subtle electronic circuits, the reactor plays a key role to ensure the stable and safe operation of the system.

[0003] During operation, the reactor will be closely coupled with the surrounding electromagnetic fields, transmission lines and other complex factors. This coupling effect includes the interaction between the reactor and the electrical network and lines, which not only affects the operating efficiency of the power system, but may also cause electromagnetic interference, resonance problems, etc. When the reactor is working, the coupling effect between the inductive reactance of the reactor and the capacitor or other components of the circuit (such as parasitic capacitance, etc.) will significantly change the inductance of the circuit, thereby affecting electrical parameters such as current, voltage and power factor. As the requirements of the power system for high efficiency, low loss and high reliability continue to increase, the design of reactors is increasingly tending towards a more precise and complex direction.

[0004] However, traditional simulation calculation methods for reactor field-circuit coupling often rely on numerical simulation and a large amount of computing resources, resulting in a long calculation time. Summary of the invention

[0005] In order to solve the above technical problems, the embodiments of the present application propose a method, system and storage medium for simulating and calculating field-circuit coupling of a reactor, which can shorten the calculation time and reduce the consumption of computing resources while ensuring the accuracy of simulation calculation.

[0006] In a first aspect, an embodiment of the present application provides a method for simulating and calculating a field-circuit coupling of a reactor, comprising: Determine the electromagnetic field model and circuit model of the reactor to be analyzed; Based on the electromagnetic field model and the circuit model, an artificial intelligence model is used to perform resonance frequency analysis and field-circuit interaction prediction; Determining a field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis results and the field-circuit interaction prediction results; Using the boundary element method, calculating according to the field-circuit coupling model, to obtain calculation results; The training method of the artificial intelligence model includes: Acquire historical simulation data of the sample reactor, wherein the historical simulation data includes historical electromagnetic field data and historical circuit data; Based on the historical simulation data, the deep learning model to be trained is trained by a loss function to obtain the artificial intelligence model, wherein the loss function is based on the resonant frequency corresponding to the sample reactor. Build.

[0007] Optionally, the loss function includes a mean square error loss function, and the formula of the mean square error loss function includes: in, is the loss function, is the number of samples in the historical simulation data, is the true value of the i-th sample in the historical simulation data, is the predicted value of the i-th sample, Represents the frequency represented by the i-th sample The corresponding weight function, and and The smaller the difference between The larger the value of .

[0008] Optionally, based on the historical simulation data, performing model training on the deep learning model to be trained by using a loss function to obtain the artificial intelligence model includes: With the goal of minimizing the loss function, a gradient descent algorithm is used to perform model training on the deep learning model to be trained based on the historical simulation data to obtain the artificial intelligence model.

[0009] Optionally, determining the field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis result and the field-circuit interaction prediction result includes: Determine the voltage and current boundary conditions based on the resonant frequency analysis results and the field-circuit interaction prediction results; A field-circuit coupling model of the reactor to be analyzed is determined based on the voltage and current boundary conditions.

[0010] Optionally, the method further comprises: In the case where the calculation result does not converge, modifying the voltage and current boundary conditions according to the calculation result; The modified voltage and current boundary conditions are used as new voltage and current boundary conditions, thereby returning to the step of determining the field-circuit coupling model of the inductor to be analyzed based on the voltage and current boundary conditions, until the obtained calculation results converge, and the converged calculation results are output.

[0011] In a second aspect, an embodiment of the present application provides a reactor field-circuit coupling simulation calculation system, including: A model determination module is used to determine the electromagnetic field model and circuit model of the reactor to be analyzed; An analysis and prediction module, used to perform resonant frequency analysis and field-circuit interaction prediction using an artificial intelligence model based on the electromagnetic field model and the circuit model; A field-circuit coupling module, used to determine a field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis result and the field-circuit interaction prediction result; A calculation module, used for performing calculations according to the field-circuit coupling model using a boundary element method to obtain calculation results; The training method of the artificial intelligence model includes: Acquire historical simulation data of the sample reactor, wherein the historical simulation data includes historical electromagnetic field data and historical circuit data; Based on the historical simulation data, the deep learning model to be trained is trained by a loss function to obtain the artificial intelligence model, wherein the loss function is based on the resonant frequency corresponding to the sample reactor. Build.

[0012] Optionally, the loss function includes a mean square error loss function, and the formula of the mean square error loss function includes: in, is the loss function, is the number of samples in the historical simulation data, is the true value of the i-th sample in the historical simulation data, is the predicted value of the i-th sample, Represents the frequency represented by the i-th sample The corresponding weight function, and and The smaller the difference between The larger the value of .

[0013] Optionally, based on the historical simulation data, performing model training on the deep learning model to be trained by using a loss function to obtain the artificial intelligence model includes: With the goal of minimizing the loss function, a gradient descent algorithm is used to perform model training on the deep learning model to be trained based on the historical simulation data to obtain the artificial intelligence model.

[0014] In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the reactor field-circuit coupling simulation calculation method described in any one of the above items are implemented.

[0015] In summary, the embodiments of the present application have at least the following beneficial effects: By adopting the embodiment of the present application, the electromagnetic field model and the circuit model of the inductor to be analyzed are determined; based on the electromagnetic field model and the circuit model, the resonant frequency analysis and field-circuit interaction prediction are performed using an artificial intelligence model; based on the resonant frequency analysis results and the field-circuit interaction prediction results, the field-circuit coupling model of the inductor to be analyzed is determined; and the boundary element method is used to perform calculations according to the field-circuit coupling model to obtain calculation results, thereby shortening the calculation time and reducing the consumption of computing resources while ensuring the accuracy of simulation calculations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of a method for simulating and calculating a field-circuit coupling of a reactor provided in an embodiment of the present application; Figure 2 It is a schematic diagram of a reactor field-circuit coupling simulation calculation provided in an embodiment of the present application; Figure 3 It is a structural schematic diagram of the reactor field-circuit coupling simulation calculation system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] In the description of the present application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, features defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "multiple" is two or more. In the description of the present application, the term "including" and its variations are open inclusions, i.e., "including but not limited to". The term "based on" means "at least partially based on". The term "according to" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments".

[0019] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0020] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this application have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0021] First, see Figure 1 , shows a schematic flow chart of a reactor field-circuit coupling simulation calculation method provided in an embodiment of the present application, the method comprising steps S101-S104, which are specifically as follows: S101, determining an electromagnetic field model and a circuit model of a reactor to be analyzed; In one example, see Figure 2 , the reactor field model (i.e. the electromagnetic field model of the reactor) and the reactor circuit model (i.e. the circuit model of the reactor) can be built in advance, so that they can be easily called when needed.

[0022] It should be noted that the electromagnetic field model and the circuit model can be used to reflect the performance changes of the reactor to be analyzed under different loads, different frequencies and different operating conditions.

[0023] In one example, a parameterized field model of the reactor can also be established to establish an electromagnetic field model. The three-dimensional model of the reactor can be accurately constructed according to the reactor type, such as air, iron core, high frequency, and actual operation scenario, involving three aspects: geometric modeling, material parameter selection, and boundary condition setting. First, define the geometric shape of the reactor, including the size and relative position of the iron core, coil, and cooling system components, and select the coordinate system according to the geometric symmetry of the reactor, including the rectangular coordinate system and the cylindrical coordinate system. Then, determine the conductivity, magnetic permeability, and relative magnetic permeability of different materials in the reactor; consider the hysteresis effect and saturation magnetic induction intensity of the material, set the hysteresis effect and nonlinear characteristics, and take into account the resistance of the coil winding and the loss caused by the current. Finally, define the current distribution of the winding in the reactor, where the magnetic field in the outer area of ​​the reactor is set to an open boundary and a material with a magnetic permeability close to vacuum according to the operating conditions. In addition, the method may also include: based on the basic characteristics of the inductor element of the inductor, performing a preliminary resonant frequency analysis when establishing the electromagnetic field model, taking into account the significant changes in the relationship between the inductance value and the current and voltage near the resonant frequency, and the model accuracy can be improved through the preliminary resonant frequency analysis results to improve the accuracy of the parameterized field model of the inductor in the circuit.

[0024] In one example, the electromagnetic field model can be constructed by solving Maxwell's equations, where the Maxwell's equations can be expressed as: in, is the electric field, is the magnetic field, is the current density, is the charge density, is the dielectric constant and B is the magnetic induction intensity.

[0025] In this embodiment, the electromagnetic field model of the reactor is established using Maxwell's equations, which can accurately describe the electromagnetic field behavior in the reactor, ensure electromagnetic compatibility, predict performance, and accurately establish the electromagnetic field model of the reactor, providing a basis for field-circuit model coupling.

[0026] In one example, the circuit model may be constructed as follows.

[0027] The circuit part of the reactor is usually composed of components such as inductance and transformer. The relationship between the current and voltage of the inductance part can be described by the following model formula: in, is the voltage across the inductor, is the inductor, is the current in the inductor, is the parasitic resistance in the circuit.

[0028] In actual circuits, there is also parasitic capacitance, which has a resonant frequency with the inductor: in, is the parasitic capacitance.

[0029] Based on this, the functional relationship between the current and voltage of the inductor can be modified as follows: in, It is a parameter related to frequency, indicating that the inductance of the reactor changes with the frequency. In this way, the circuit model can be expressed by the above formula.

[0030] It can be understood that by inputting the current, voltage and other working conditions of the reactor into the above model, the characteristics of the reactor such as the electric field strength and magnetic field strength can be quickly predicted; in addition, the component parameters in the model can also be quickly adjusted according to the electromagnetic behavior of the reactor.

[0031] S102, based on the electromagnetic field model and the circuit model, using an artificial intelligence model to perform resonant frequency analysis and field-circuit interaction prediction; It should be noted that the artificial intelligence model in this embodiment can be a related model that has been pre-trained using a general model training method, so that the model can learn the nonlinear relationship between the electromagnetic field and the circuit, solve the problem that the traditional coupling method requires multiple iterative calculations and numerical simulations, large amount of calculations, and slow calculation speed, and is convenient for quickly and accurately establishing the field-circuit coupling model of the reactor. The artificial intelligence model is suitable for taking the parameters of the electromagnetic field model and the circuit model as input, and performing resonant frequency analysis and field-circuit interaction prediction based on this. Field-circuit interaction can refer to the interaction between the magnetic field (field) and the circuit (circuit).

[0032] S103, determining a field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis result and the field-circuit interaction prediction result; It should be noted that when the reactor is working, the resonant frequency is usually closely related to the operating frequency of the system. Near the resonant frequency, the coupling effect between the inductive reactance of the reactor and the capacitor or other components of the circuit (such as parasitic capacitance, etc.) will significantly change the inductance of the circuit, thereby affecting electrical parameters such as current, voltage, and power factor. Therefore, in this embodiment, the resonant frequency analysis results can be used to assist the field-circuit interaction prediction results, thereby efficiently constructing a more accurate field-circuit coupling model.

[0033] S104, using the boundary element method to perform calculations according to the field-circuit coupling model to obtain calculation results.

[0034] The training method of the artificial intelligence model includes: Acquire historical simulation data of the sample reactor, wherein the historical simulation data includes historical electromagnetic field data and historical circuit data; Based on the historical simulation data, the deep learning model to be trained is trained by a loss function to obtain the artificial intelligence model, wherein the loss function is based on the resonant frequency corresponding to the sample reactor. Build.

[0035] In an optional implementation, the loss function includes a mean square error loss function, and the formula of the mean square error loss function includes: in, is the loss function, is the number of samples in the historical simulation data, is the true value of the i-th sample in the historical simulation data, is the predicted value of the i-th sample, Represents the frequency represented by the i-th sample The corresponding weight function, and and The smaller the difference between The larger the value of .

[0036] It should be noted that the real value and the predicted value in this embodiment can be reflected in parameters such as voltage, current, and electromagnetic field.

[0037] In this embodiment, different weights can be assigned to the prediction errors at different frequencies, and the maximum weight will appear at the resonant frequency. The closest frequency makes the weight smaller and the corresponding loss error smaller when the frequency is far away from the resonant frequency, so as to ensure better fitting of the impact of frequencies close to the resonant frequency on the voltage and current of the reactor, thereby learning the influence mechanism of the resonant frequency on the electromagnetic field and circuit model, so that the artificial intelligence model can predict the change of inductance value with resonant frequency, as well as the distribution of voltage and current at different frequencies.

[0038] In an optional implementation, based on the historical simulation data, the deep learning model to be trained is trained by using a loss function to obtain the artificial intelligence model, including: With the goal of minimizing the loss function, a gradient descent algorithm is used to perform model training on the deep learning model to be trained based on the historical simulation data to obtain the artificial intelligence model.

[0039] In one example, the specific process of the gradient descent algorithm may include: (1) Forward propagation: Calculate the network's predicted value.

[0040] (2) Calculate the loss function: Calculate the loss function based on the predicted value and the actual value.

[0041] (3) Back propagation: Calculate the gradient of the loss function for each parameter in the model through the chain rule.

[0042] (4) Update parameters: Update the parameters according to the formula of the gradient descent method. For each parameter (for example, including weights and biases), the update rule is: .in, are the neural network parameters (weights and biases), is the learning rate, which depends on the step size of each update, is the loss function with respect to the parameters The gradient of , The parameters after updating and the parameters before updating are represented in sequence.

[0043] In an optional implementation, determining the field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis result and the field-circuit interaction prediction result includes: Determine the voltage and current boundary conditions based on the resonant frequency analysis results and the field-circuit interaction prediction results; A field-circuit coupling model of the reactor to be analyzed is determined based on the voltage and current boundary conditions.

[0044] In this embodiment, since the aforementioned artificial intelligence model has the ability to analyze resonant frequency and predict field-circuit interaction, it can predict the mutual influence of voltage, current and electromagnetic field in the circuit, and then obtain more accurate voltage and current boundary conditions based on the resonant frequency analysis results and field-circuit interaction prediction results.

[0045] In one example, simulation tools (such as Ansys Maxwell + Simplorer, ComsolMultiphysics, etc.) can be used to perform field-circuit coupling simulation based on the resonant frequency analysis results and field-circuit interaction prediction results to simulate reasonable voltage and current boundary conditions, and configure the field-circuit coupling model accordingly.

[0046] In an optional embodiment, the method further includes: In the case where the calculation result does not converge, modifying the voltage and current boundary conditions according to the calculation result; The modified voltage and current boundary conditions are used as new voltage and current boundary conditions, thereby returning to the step of determining the field-circuit coupling model of the inductor to be analyzed based on the voltage and current boundary conditions, until the obtained calculation results converge, and the converged calculation results are output.

[0047] In this embodiment, since the influence of the resonant frequency on the circuit inductance and the resulting current and voltage fluctuations are taken into account, it is ensured that the boundary conditions can adapt to the frequency changes in each iterative calculation step, thereby improving the accuracy and efficiency of the calculation. In addition, according to the modification of the boundary conditions, the new voltage and current boundary conditions can be used to quickly update the reactor field-circuit coupling model, avoiding the complex process of recalculating the electromagnetic field model and / or circuit model each time.

[0048] In one example, the boundary element method may use a boundary element algorithm (BEM) based on resonant frequency analysis to deal with the dynamic coupling form between the reactor and the circuit considering the resonant frequency. The formula of the boundary element algorithm includes: in, is the frequency of the resonant frequency , on-site The field quantity (electric field or magnetic field) at time t, is the boundary surface, is the boundary condition function, indicating that The source strength (current, voltage) at the location includes the time (characterized by time t) and frequency The dynamic effects associated with is the frequency-dependent Green's function.

[0049] In this embodiment, the boundary element algorithm has the following advantages: (1) High efficiency: The boundary element method can reduce the solution dimension of a three-dimensional problem to two dimensions, thereby greatly reducing the amount of calculation. It has significant computational advantages, especially when dealing with reactors with complex geometric shapes.

[0050] (2) Accuracy: The boundary element method can handle the coupling problem between electromagnetic fields and circuits with high precision and is suitable for large-scale power systems and complex electrical equipment.

[0051] (3) Flexibility: The boundary element method is applicable to various types of boundary conditions and can cope well with various coupling forms between reactors and circuits, such as static, dynamic, time domain and frequency domain.

[0052] Based on the above corresponding embodiments, the present application can achieve the following beneficial effects: 1. Through geometric modeling, material parameter selection, and boundary condition setting of the reactor in actual operation, the electromagnetic field model that meets the actual operating conditions of the reactor can be quickly established. The parameterized circuit model of the reactor that meets the actual operation of the reactor can be established through the external circuit characteristics of the reactor in operation. The accuracy of the electromagnetic field model and the circuit model can be improved through resonant frequency analysis, reflecting the performance changes of the reactor under different loads, different frequencies, and different operating conditions.

[0053] 2. Using artificial intelligence models (e.g. Figure 2 The recurrent neural network deep learning model shown in ) uses a large amount of simulation data to learn the complex nonlinear relationship between the electromagnetic field model and the circuit model of the reactor, and applies it to the field-circuit model coupling process to learn the influence mechanism of the resonant frequency on the field-circuit coupling model, realize automatic interactive prediction and calculation of the field-circuit model, and obtain more accurate voltage and current boundary conditions, so as to quickly and accurately establish the field-circuit coupling model of the reactor.

[0054] 3. Use the boundary element algorithm to solve the reactor field-circuit coupling model taking into account the influence of the resonant frequency. Modify the voltage and current boundary conditions of the reactor according to the calculation results, quickly update the reactor field-circuit coupling model and perform calculations until convergence. This avoids the complex process of recalculating the electromagnetic field model and greatly reduces the amount of calculation, so as to realize the rapid calculation of the reactor field-circuit coupling.

[0055] On the second aspect, accordingly, the embodiments of the present application also provide a reactor field-circuit coupling simulation calculation system, which can implement all the processes of the reactor field-circuit coupling simulation calculation method provided in the above embodiments.

[0056] See also Figure 3 , shows a schematic diagram of the structure of a reactor field-circuit coupling simulation computing system provided in an embodiment of the present application, the reactor field-circuit coupling simulation computing system comprising: A model determination module 301 is used to determine the electromagnetic field model and circuit model of the reactor to be analyzed; An analysis and prediction module 302 is used to perform resonance frequency analysis and field-circuit interaction prediction based on the electromagnetic field model and the circuit model using an artificial intelligence model; A field-circuit coupling module 303 is used to determine a field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis result and the field-circuit interaction prediction result; The calculation module 304 is used to use the boundary element method to perform calculations according to the field-circuit coupling model to obtain calculation results.

[0057] The training method of the artificial intelligence model includes: Acquire historical simulation data of the sample reactor, wherein the historical simulation data includes historical electromagnetic field data and historical circuit data; Based on the historical simulation data, the deep learning model to be trained is trained by a loss function to obtain the artificial intelligence model, wherein the loss function is based on the resonant frequency corresponding to the sample reactor. Build.

[0058] In an optional implementation, the loss function includes a mean square error loss function, and the formula of the mean square error loss function includes: in, is the loss function, is the number of samples in the historical simulation data, is the true value of the i-th sample in the historical simulation data, is the predicted value of the i-th sample, Represents the frequency represented by the i-th sample The corresponding weight function, and and The smaller the difference between The larger the value of .

[0059] In an optional implementation, based on the historical simulation data, the deep learning model to be trained is trained by using a loss function to obtain the artificial intelligence model, including: With the goal of minimizing the loss function, a gradient descent algorithm is used to perform model training on the deep learning model to be trained based on the historical simulation data to obtain the artificial intelligence model.

[0060] In an optional implementation, determining the field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis result and the field-circuit interaction prediction result includes: Determine the voltage and current boundary conditions based on the resonant frequency analysis results and the field-circuit interaction prediction results; A field-circuit coupling model of the reactor to be analyzed is determined based on the voltage and current boundary conditions.

[0061] In an optional implementation, the system further includes an iteration module, wherein the iteration module is configured to: In the case where the calculation result does not converge, modifying the voltage and current boundary conditions according to the calculation result; The modified voltage and current boundary conditions are used as new voltage and current boundary conditions, thereby returning to the step of determining the field-circuit coupling model of the inductor to be analyzed based on the voltage and current boundary conditions, until the obtained calculation results converge, and the converged calculation results are output.

[0062] In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the reactor field-circuit coupling simulation calculation method described in any one of the above items are implemented.

[0063] In summary, the embodiments of the present application have at least the following beneficial effects: By adopting the embodiment of the present application, the electromagnetic field model and the circuit model of the inductor to be analyzed are determined; based on the electromagnetic field model and the circuit model, the resonant frequency analysis and field-circuit interaction prediction are performed using an artificial intelligence model; based on the resonant frequency analysis results and the field-circuit interaction prediction results, the field-circuit coupling model of the inductor to be analyzed is determined; and the boundary element method is used to perform calculations according to the field-circuit coupling model to obtain calculation results, thereby shortening the calculation time and reducing the consumption of computing resources while ensuring the accuracy of simulation calculations.

[0064] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by means of software plus the necessary hardware platform, and of course it can also be implemented entirely by hardware. Based on such an understanding, all or part of the contribution of the technical solution of the present application to the background technology can be embodied in the form of a software product, and the computer software product can be stored in a storage medium, such as ROM (Read-Only Memory) / RAM (Random Access Memory), a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application or some parts of the embodiments.

[0065] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications are also considered to be within the scope of protection of the present application.

Claims

1. A method for simulating and calculating field-circuit coupling of a reactor, characterized in that: include: Determine the electromagnetic field model and circuit model of the reactor to be analyzed; Based on the electromagnetic field model and the circuit model, an artificial intelligence model is used to perform resonance frequency analysis and field-circuit interaction prediction; Determining a field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis results and the field-circuit interaction prediction results; Using the boundary element method, calculating according to the field-circuit coupling model, to obtain calculation results; The training method of the artificial intelligence model includes: Acquire historical simulation data of the sample reactor, wherein the historical simulation data includes historical electromagnetic field data and historical circuit data; Based on the historical simulation data, the deep learning model to be trained is trained by a loss function to obtain the artificial intelligence model, wherein the loss function is based on the resonant frequency corresponding to the sample reactor. Build.

2. The reactor field-circuit coupling simulation calculation method according to claim 1, characterized in that: The loss function includes a mean square error loss function, and the formula of the mean square error loss function includes: in, is the loss function, is the number of samples in the historical simulation data, is the true value of the i-th sample in the historical simulation data, is the predicted value of the i-th sample, Represents the frequency represented by the i-th sample The corresponding weight function, and and The smaller the difference between The larger the value of .

3. The reactor field-circuit coupling simulation calculation method according to claim 1, characterized in that: The method of performing model training on the deep learning model to be trained based on the historical simulation data by using a loss function to obtain the artificial intelligence model includes: With the goal of minimizing the loss function, a gradient descent algorithm is used to perform model training on the deep learning model to be trained based on the historical simulation data to obtain the artificial intelligence model.

4. The reactor field-circuit coupling simulation calculation method according to claim 1, characterized in that: The determining of the field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis result and the field-circuit interaction prediction result includes: Determine the voltage and current boundary conditions based on the resonant frequency analysis results and the field-circuit interaction prediction results; A field-circuit coupling model of the reactor to be analyzed is determined based on the voltage and current boundary conditions.

5. The method for simulating and calculating the field-circuit coupling of a reactor according to claim 4, characterized in that: The method further comprises: In the case where the calculation result does not converge, modifying the voltage and current boundary conditions according to the calculation result; The modified voltage and current boundary conditions are used as new voltage and current boundary conditions, thereby returning to the step of determining the field-circuit coupling model of the inductor to be analyzed based on the voltage and current boundary conditions, until the obtained calculation results converge, and the converged calculation results are output.

6. A reactor field-circuit coupling simulation calculation system, characterized in that: include: A model determination module is used to determine the electromagnetic field model and circuit model of the reactor to be analyzed; An analysis and prediction module, used to perform resonant frequency analysis and field-circuit interaction prediction using an artificial intelligence model based on the electromagnetic field model and the circuit model; A field-circuit coupling module, used to determine a field-circuit coupling model of the reactor to be analyzed based on the resonant frequency analysis result and the field-circuit interaction prediction result; A calculation module, used for performing calculations according to the field-circuit coupling model using a boundary element method to obtain calculation results; The training method of the artificial intelligence model includes: Acquire historical simulation data of the sample reactor, wherein the historical simulation data includes historical electromagnetic field data and historical circuit data; Based on the historical simulation data, the deep learning model to be trained is trained by a loss function to obtain the artificial intelligence model, wherein the loss function is based on the resonant frequency corresponding to the sample reactor. Build.

7. The reactor field-circuit coupling simulation calculation system according to claim 6, characterized in that: The loss function includes a mean square error loss function, and the formula of the mean square error loss function includes: in, is the loss function, is the number of samples in the historical simulation data, is the true value of the i-th sample in the historical simulation data, is the predicted value of the i-th sample, Represents the frequency represented by the i-th sample The corresponding weight function, and and The smaller the difference between The larger the value of .

8. The reactor field-circuit coupling simulation calculation system according to claim 6, characterized in that: The method of performing model training on the deep learning model to be trained based on the historical simulation data by using a loss function to obtain the artificial intelligence model includes: With the goal of minimizing the loss function, a gradient descent algorithm is used to perform model training on the deep learning model to be trained based on the historical simulation data to obtain the artificial intelligence model.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the reactor field-circuit coupling simulation calculation method according to any one of claims 1 to 5 is implemented.

Images