Modeling method and device of transformer circuit, electronic equipment and storage medium

Abstract

The application provides a modeling method and device of a transformer circuit, an electronic device and a storage medium. The method comprises: obtaining coil parameters of a transformer coil; wherein the transformer coil comprises a primary coil and a secondary coil, and the coil parameters comprise the width of the primary coil, the radius of the primary coil, the width of the secondary coil and the radius of the secondary coil; determining element parameters of circuit elements in the transformer circuit according to the coil parameters; wherein the transformer circuit comprises at least two circuit elements; determining a circuit model of the transformer according to a preset correlation and a preset circuit model structure, wherein the preset correlation is the correlation between the element parameters of the circuit elements and the preset circuit elements, and the preset circuit model structure is used to represent the connection relationship between the circuit elements in the circuit model. The application can automatically determine the transformer circuit according to the coil parameters of the transformer, thereby improving the modeling efficiency of the transformer circuit.

Description

Technical Field

[0001] This application relates to circuit technology, and more particularly to a modeling method, apparatus, electronic device, and storage medium for transformer circuits. Background Technology

[0002] Radio frequency integrated circuits use a large number of passive components, such as inductors, transmission lines, and transformers. The on-chip spiral transformer is an important passive component in RFICs.

[0003] Transformers have complex geometries, and at high frequencies, their electrical characteristics are affected by various physical effects and loss mechanisms. Transformer circuit models contain numerous components, requiring multiple iterations to extract parameter values ​​for each component during modeling to find suitable parameters. Changing the transformer coil parameters necessitates further iterations to re-determine these parameters, resulting in low accuracy and efficiency in transformer circuit modeling. Summary of the Invention

[0004] This application provides a method, apparatus, electronic device, and storage medium for modeling transformer circuits, in order to improve the modeling efficiency of transformer circuits.

[0005] Firstly, this application provides a modeling method for transformer circuits, including:

[0006] Obtain the coil parameters of the transformer coil; wherein the transformer coil includes a primary coil and a secondary coil, and the coil parameters include the width of the primary coil, the radius of the primary coil, the width of the secondary coil, and the radius of the secondary coil;

[0007] Based on the coil parameters, the component parameters of the circuit elements in the transformer circuit are determined; wherein, the transformer circuit includes at least two circuit elements;

[0008] The circuit model of the transformer is determined based on the preset correlation and preset circuit model structure. The preset correlation is the correlation between the component parameters of the circuit element and the preset circuit element. The preset circuit model structure is used to represent the connection relationship between each circuit element in the circuit model.

[0009] Secondly, this application provides a modeling apparatus for transformer circuits, comprising:

[0010] A coil parameter acquisition module is used to acquire the coil parameters of a transformer coil; wherein the transformer coil includes a primary coil and a secondary coil, and the coil parameters include the width of the primary coil, the radius of the primary coil, the width of the secondary coil, and the radius of the secondary coil;

[0011] The component parameter determination module is used to determine the component parameters of the circuit elements in the transformer circuit based on the coil parameters; wherein the transformer circuit includes at least two circuit elements;

[0012] The circuit model determination module is used to determine the circuit model of the transformer based on preset correlation relationships and preset circuit model structures. The preset correlation relationships are the component parameters of the circuit elements and the preset correlation relationships between the circuit elements. The preset circuit model structure is used to represent the connection relationships between the circuit elements in the circuit model.

[0013] Thirdly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0014] The memory stores computer-executed instructions;

[0015] The processor executes computer execution instructions stored in the memory to implement the transformer circuit modeling method as described in the first aspect of this application.

[0016] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the transformer circuit modeling method as described in the first aspect of this application.

[0017] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the transformer circuit modeling method as described in the first aspect of this application.

[0018] This application provides a modeling method, apparatus, electronic device, and storage medium for transformer circuits. By obtaining the width and radius of each coil in the transformer, the coil parameters of the transformer coils are obtained. Based on the coil parameters, the component parameters of each circuit element corresponding to the coil parameters are determined. Based on the connection relationships of each circuit element in the transformer circuit and the component parameters of the circuit elements, the transformer circuit is obtained. This solves the problem in existing technologies where, when changing transformer coil parameters, the user has to sequentially extract the component parameters of each circuit element, resulting in low efficiency and accuracy in determining component parameters. By directly obtaining the component parameters of each circuit element from the coil parameters, manpower and time are effectively saved, and the efficiency and accuracy of transformer circuit determination are improved. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0020] Figure 1 A schematic flowchart illustrating a transformer circuit modeling method provided in an embodiment of this application;

[0021] Figure 2 This is a schematic diagram of the device structure of the transformer provided in the embodiments of this application;

[0022] Figure 3 A schematic diagram of the preset circuit model structure of the six-port transformer provided in the embodiments of this application;

[0023] Figure 4 A schematic flowchart illustrating a transformer circuit modeling method provided in an embodiment of this application;

[0024] Figure 5 A structural block diagram of a transformer circuit modeling device provided in an embodiment of this application;

[0025] Figure 6 A structural block diagram of a transformer circuit modeling device provided in an embodiment of this application;

[0026] Figure 7 A structural block diagram of an electronic device provided in an embodiment of this application;

[0027] Figure 8 This is a structural block diagram of an electronic device provided in an embodiment of this application.

[0028] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0030] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0031] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0032] In the description of this application, it should be understood that the terms "first," "second," "third," etc., are used only to distinguish similar objects and are not necessarily used to describe a specific order or sequence, nor should they be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0033] It should be noted that, due to space limitations, this application specification does not exhaustively list all possible implementation methods. Those skilled in the art, after reading this application specification, should be able to deduce that, as long as the technical features do not contradict each other, any combination of technical features can constitute an optional implementation method. The following provides a detailed description of each embodiment.

[0034] With the continuous development of silicon-based CMOS (Complementary Metal Oxide Semiconductor) technology, the feature size of MOS (Metal Oxide Semiconductor) devices is constantly shrinking, the cutoff frequency and performance of transistors are constantly improving, the manufacturing cost is low, and it is easy to integrate with digital integrated circuits, making the advantages of CMOS technology in the radio frequency field increasingly obvious.

[0035] Radio frequency (RF) integrated circuits utilize numerous passive components, such as inductors, transmission lines, and transformers. On-chip helical transformers (OTCs) are crucial passive components in RFICs; therefore, establishing an accurate equivalent circuit model of the transformer is essential.

[0036] Transformers have complex geometries, and at high frequencies, their electrical characteristics are affected by various physical effects and loss mechanisms. Transformer circuit models contain numerous components, and determining the parameter values ​​for each component is a cumbersome process, often requiring multiple iterations with low accuracy. Furthermore, this method can only be used for transformers of fixed dimensions. "Fixed dimensions" can refer to the fixed dimensions of the transformer's coils.

[0037] In other words, when determining the transformer circuit model, it is necessary to iteratively extract the parameters of each circuit element for a transformer of fixed size in order to find suitable circuit element parameters. If the coil parameters of the transformer are changed, it is necessary to iterate and extract the circuit elements multiple times again to redetermine the parameters of the circuit elements. The modeling accuracy and efficiency of transformer circuits are low, and it is impossible to achieve scalable modeling of transformer size.

[0038] This application provides a modeling method, apparatus, electronic device, and storage medium for transformer circuits, aiming to solve the above-mentioned technical problems in the prior art.

[0039] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0040] Figure 1 This is a flowchart illustrating a transformer circuit modeling method according to an embodiment of this application. The method is executed by a transformer circuit modeling device. Figure 1 As shown, the method includes the following steps:

[0041] S101. Obtain the coil parameters of the transformer coil; wherein, the transformer coil includes a primary coil and a secondary coil, and the coil parameters include the width of the primary coil, the radius of the primary coil, the width of the secondary coil, and the radius of the secondary coil.

[0042] For example, the device structure of a transformer may include a transformer coil, which may include a primary coil and a secondary coil, each of which may consist of a single-turn inductor. The coil parameters of the primary and secondary coils in the transformer may be different; these parameters may be dimensional parameters such as the width and radius of the coil. That is, the coil parameters of the transformer coil may include the width and radius of the primary coil, and the width and radius of the secondary coil. The transformer may include multiple ports, and the voltage and current at different ports may be different. Figure 2 This is a schematic diagram of the transformer's component structure. Figure 2 There are six transformer ports, namely P1, P2, P3, P4, P5 and P6.

[0043] When modeling a transformer circuit, the coil parameters of the transformer coils need to be known. These parameters can be selected from a preset range. For example, the width of the primary coil can be W1, its radius R1, the width of the secondary coil W2, and its radius R2. Users can predetermine the range of coil parameter values ​​and randomly determine each coil parameter within that range, or they can specify the parameters for each coil. The coil parameters can change according to the coil shape. For example, the vector of parameters for a single-turn coil can be [W1, R1, W2, R2], which is currently the most common shape in practical applications. However, this does not exclude multi-turn coils; the vector of parameters for multi-turn coils can be [W1, R1, N1, S1, W2, R2, N2, S2]. Here, N and S represent the number of turns and the spacing, respectively. The coil parameters also involve many details of cross-layer wiring where inner and outer coil traces overlap, and the shape can be quadrilateral, octagonal, or circular, etc.

[0044] S102. Determine the component parameters of the circuit elements in the transformer circuit based on the coil parameters; wherein the transformer circuit includes at least two circuit elements.

[0045] For example, a circuit structure of a transformer circuit is pre-constructed as a preset circuit model structure. This preset circuit model structure can be used to represent the connection relationships between various circuit elements in the transformer circuit; that is, the various circuit elements in the transformer circuit and the connection relationships between them are pre-determined. Modeling the transformer circuit involves determining the component parameters of each circuit element in the transformer circuit, so that the transformer circuit can meet the operating requirements of the transformer under specific coil parameters.

[0046] Figure 3 This is a schematic diagram of a preset circuit model structure for a six-port transformer. The preset circuit model structure for the transformer circuit may include multiple circuit elements, such as resistors, capacitors, and inductors. Figure 3 In this circuit model, R represents resistance, C represents capacitance, L represents inductance, and P1 to P6 represent the six ports of the transformer. The component parameters of each circuit element in the preset circuit model structure are unknown; these parameters can be resistance, capacitance, inductance, etc.

[0047] The relationship between coil parameters and component parameters can be preset, and the component parameters corresponding to the coil parameters can be determined based on the preset relationship.

[0048] The functional relationship between the parameter values ​​of each component and the coil parameters can be preset, and the component parameters of each circuit component can be calculated based on the coil parameters. For example, a neural network model can be preset, with the coil parameters as input and the component parameter values ​​of each circuit component as output.

[0049] The neural network model can be pre-trained. For example, coil parameters to be trained can be pre-collected within the range of coil parameter values ​​to form a vector of coil parameters, which can be represented as [R1, W1, R2, W2]. Under the selected process platform, a six-port electromagnetic simulation platform for the transformer is established. The vector of coil parameters is input into the electromagnetic simulation platform to obtain the six-port S(Scatter) parameters. Then, the component parameters to be trained are collected from the range of component parameter values ​​for each circuit element to form a component parameter vector. A transformer circuit netlist is pre-set, and the component parameter vector is input to obtain the six-port S parameters. The correlation between coil parameters and S parameters can be obtained based on the coil parameters and S parameters; the correlation between component parameters and S parameters can be obtained based on the S parameters and component parameters, thus obtaining the correlation between coil parameters and component parameters. The coil parameters to be trained are input into the pre-built neural network model. If the output result is consistent with the component parameters to be trained, or the difference between the output result and the component parameters to be trained is within a preset difference range, then the neural network model training is considered complete. When modeling a transformer circuit, the determined coil parameters are input into the neural network model, which yields the component parameters of the circuit elements in the transformer circuit.

[0050] S103. Determine the circuit model of the transformer based on the preset correlation and preset circuit model structure. The preset correlation refers to the component parameters of the circuit elements and the preset correlation between the circuit elements. The preset circuit model structure is used to represent the connection relationship between each circuit element in the circuit model.

[0051] For example, the obtained component parameters are the component parameters of each circuit element in the preset circuit model structure, and the obtained component parameters can be represented in vector form. The relationship between the obtained component parameters and each circuit element in the preset circuit model structure is preset, and the circuit element corresponding to each component parameter is determined according to the preset relationship.

[0052] Based on the preset circuit model structure, the composition and circuit structure of the transformer circuit are determined. According to the preset correlation, the component parameter values ​​of each circuit element in the circuit structure are obtained, thus obtaining the final transformer circuit and completing the modeling of the transformer circuit.

[0053] In this embodiment, the circuit model of the transformer is determined according to the preset association relationship and the preset circuit model structure, including: assigning component parameters to the corresponding circuit components according to the preset association relationship to obtain the component parameters of each circuit component in the preset circuit model structure; and obtaining the circuit model of the transformer according to the preset circuit model structure and the component parameters of each circuit component in the preset circuit model structure.

[0054] Specifically, the component parameters are represented in vector form. For example, the obtained component parameter vector is [L10, R10, L11, R11, Cox1, Rsub1, Csub1, L20, R20, L21, R21, Cox2, Rsub2, Csub2, L30, R30, L31, R31, Cox3, Rsub3, Csub3, L40, R40, L41, R41, Cox4, Rsub4, Csub4, C12, C13, C14, C16, C23, C24, C26, C34, C35, C45, M12, M34, M13, M24, M14, M23], where L10, L11, L20, L21, L30, L31, L40, and L41 are... Figure 3 The component parameters of the inductors in the preset circuit model structure; R10, R11, Rsub1, R20, R21, Rsub2, R30, R31, Rsub3, R40, R41, and Rsub4 are... Figure 3 The component parameters of the resistors in the preset circuit model structure; Cox1, Csub1, Cox2, Csub2, Cox3, Csub3, Cox4, Csub4, C12, C13, C14, C16, C23, C24, C26, C34, C35 and C45 are... Figure 3 The component parameters of the capacitors in the preset circuit model structure are as follows: M12 is the mutual inductance between L10 and L20; M34 is the mutual inductance between L30 and L40; M13 is the mutual inductance between L10 and L30; M24 is the mutual inductance between L20 and L40; M14 is the mutual inductance between L10 and L40; and M23 is the mutual inductance between L20 and L30.

[0055] Based on the preset association relationships, component parameters are assigned to the corresponding circuit components, thus determining the component parameters of each circuit component. Based on the preset circuit model structure, the connection relationships between each circuit component are determined. Based on the component parameters of each circuit component in the preset circuit model structure, the final circuit model of the transformer is obtained, enabling the simulation of an actual transformer that meets the coil parameter conditions.

[0056] The advantage of this setup is that it assigns the obtained component parameters to the corresponding circuit components, avoiding errors in determining the component parameters of each circuit component in the preset circuit model structure. This enables the transformer circuit to meet the working requirements of the transformer under the condition of the coil parameters, realizes automatic modeling of the transformer, and allows the transformer coil parameters to be changed at any time, realizing scalable modeling of the transformer size and improving the modeling accuracy and efficiency of the transformer circuit.

[0057] This application provides a method for modeling a transformer circuit. By obtaining the width and radius of each coil in the transformer, the coil parameters of the transformer coils are obtained. Based on the coil parameters, the component parameters of each circuit element corresponding to the coil parameters are determined. Based on the connection relationships of each circuit element in the transformer circuit and the component parameters of the circuit elements, the transformer circuit is obtained. This solves the problem in the prior art where, when changing the transformer coil parameters, the user has to sequentially extract the component parameters of each circuit element, resulting in low efficiency and accuracy in determining the component parameters. By directly obtaining the component parameters of each circuit element from the coil parameters, manpower and time are effectively saved, and the efficiency and accuracy of transformer circuit determination are improved.

[0058] Figure 4 This is a flowchart illustrating a modeling method for a transformer circuit provided in an embodiment of this application. This embodiment is an optional embodiment based on the above embodiment.

[0059] In this embodiment, the component parameters of the circuit elements in the transformer circuit are determined based on the coil parameters. This can be further refined as follows: the coil parameters are input into a preset component parameter determination model, and the component parameters of the circuit elements in the transformer circuit are output. The component parameter determination model has a preset functional relationship between the coil parameters and the component parameters.

[0060] like Figure 4 As shown, the method includes the following steps:

[0061] S401. Obtain the coil parameters of the transformer coil; wherein, the transformer coil includes a primary coil and a secondary coil, and the coil parameters include the width of the primary coil, the radius of the primary coil, the width of the secondary coil, and the radius of the secondary coil.

[0062] For example, this step can refer to step S101 above, and will not be repeated here.

[0063] S402. Input the coil parameters into the preset component parameter determination model and output the component parameters of the circuit elements in the transformer circuit; wherein, the component parameter determination model has a preset functional relationship between the coil parameters and the component parameters.

[0064] For example, a component parameter determination model is pre-trained, which can be a neural network model. This model is used to obtain component parameters based on coil parameters. The model input can be coil parameters in vector form, and the output can be component parameters in vector form. The model pre-defines a functional relationship between the coil parameters and the component parameters. This functional relationship serves to determine the component parameters based on the coil parameters.

[0065] In this embodiment, inputting coil parameters into a preset component parameter determination model and outputting component parameters of circuit elements in the transformer circuit includes: inputting coil parameters into a preset first neural network model and outputting an intermediate feature vector of the transformer corresponding to the coil parameters; wherein, the first neural network model has a preset functional relationship between coil parameters and intermediate feature vector, and the intermediate feature vector is used to represent transformer characteristics; inputting the intermediate feature vector into a preset second neural network model and outputting component parameters of circuit elements in the transformer circuit corresponding to the intermediate feature vector; wherein, the second neural network model has a preset functional relationship between intermediate feature vector and component parameters.

[0066] Specifically, the component parameter determination model may include two neural network models: a first neural network model and a second neural network model. The first neural network model and the second neural network model are concatenated to form the component parameter determination model. Concatenating the first neural network model and the second neural network model can mean using the output of the first neural network as the input of the second neural network model.

[0067] The coil parameters in vector form are input into the first neural network model. This model has a pre-defined functional relationship between the coil parameters and the intermediate feature vector. The intermediate feature vector can be calculated based on the coil parameters. This vector represents the transformer characteristics, which can include voltage, current, mutual inductance, and capacitance data at each port. These characteristics can be calculated based on the S-parameters, Y-parameters (admittance), and Z-parameters (impedance) at each port. For example, the S-parameters can be determined first based on the coil parameters, then converted into Y and Z parameters. The real and imaginary parts of these parameters are then used as elements in the intermediate feature vector.

[0068] After obtaining the intermediate feature vector, the intermediate feature vector is input into the second neural network model. The second neural network model has a preset functional relationship between the intermediate feature vector and the component parameters. Based on the intermediate feature vector, the component parameters in vector form are obtained and output.

[0069] The advantage of this setup is that by splicing the two neural network models together, a model with defined component parameters is obtained, which facilitates independent training of the two neural network models and thus obtains the input and output data required for independent training of the two networks.

[0070] In this embodiment, before inputting the coil parameters into a preset first neural network model and outputting the intermediate feature vector of the transformer corresponding to the coil parameters, the method further includes: determining the coil parameter vector to be trained based on the value range of the coil parameters of the transformer coil; determining the scattering parameters of the transformer based on the coil parameter vector to be trained and a preset electromagnetic simulation platform of the transformer; determining the intermediate feature vector of the transformer based on the scattering parameters of the transformer; inputting the coil parameter vector to be trained into the preset first neural network model, and if the difference between the output of the first neural network model and the intermediate feature vector is within a preset first difference range, then the training of the first neural network model is determined to be complete.

[0071] Specifically, the first neural network model is pre-trained by determining the values ​​of multiple coil parameters within the range of values ​​for each coil parameter, thus obtaining a vector of coil parameters to be trained. This vector of coil parameters is a vector form of coil parameters, and each vector can include a width value, a radius value, a width value, and a radius value of the secondary coil. These multiple coil parameter vectors are then used as the training vectors for the first neural network model.

[0072] Before training the first neural network model, an electromagnetic simulation platform for the transformer is established under a selected process platform, which can be a six-port electromagnetic simulation platform. The selected process platform can be a 40nm CMOS process platform. Based on the limitations of the selected process characteristics and the actual usage requirements of the transformer, the range of values ​​for the transformer coil parameters is determined. Within this range, the coil parameters are varied to generate a series of datasets with different coil parameter sizes. The number of values ​​for each coil parameter can be changed based on the network training accuracy requirements for the dataset size. The coil parameters of different transformers constitute different coil parameter vectors, i.e., [R1, W1, R2, W2]. Under the specified process platform, the electromagnetic simulation platform is calibrated, and all coil parameter vectors in the simulation dataset are used to obtain the multi-port S-parameters of the transformer; for example, the six-port S-parameters can be obtained. After obtaining the S-parameters, the multi-port S-parameters are processed to obtain intermediate feature vectors.

[0073] The coil parameter vector used to generate the S-parameters is used as the coil parameter vector to be trained. Based on the coil parameter vector to be trained and the corresponding intermediate feature vectors, the first neural network model is trained. Using the coil parameter vector as input and the intermediate feature vector as output, a portion of the coil parameter vectors to be trained and their corresponding intermediate feature vectors are selected as the training set, and the remaining samples are used as the test set. The parameters of the first neural network model and the size of the dataset are adjusted until the difference between the output of the first neural network model and the corresponding intermediate feature vector is within a preset first difference range. At this point, the training of the first neural network model is considered complete, thus establishing the functional relationship between the coil parameters and the intermediate feature vectors. The number of hidden layers in the first neural network model can be adjusted, and activation functions, loss functions, and optimizers can be selected. The learning rate and number of iterations can be determined, and training continues until the error requirement is met.

[0074] The advantage of this setup is that, through the electromagnetic simulation platform, the correct S-parameters corresponding to the coil parameter vector can be determined, which facilitates obtaining the correct intermediate feature vector based on the correct S-parameters and improves the training accuracy of the first neural network model.

[0075] In this embodiment, before inputting the intermediate feature vector into the preset second neural network model and outputting the component parameters of the circuit elements in the transformer circuit corresponding to the intermediate feature vector, the method further includes: determining the component parameter vector to be trained based on the value range of the component parameters of the circuit elements in the transformer circuit; determining the scattering parameters of the transformer based on the component parameter vector to be trained and the preset transformer circuit netlist; determining the intermediate feature vector of the transformer based on the scattering parameters of the transformer; inputting the intermediate feature vector into the preset second neural network model; and determining that the training of the second neural network model is complete if the difference between the output of the second neural network model and the component parameter vector to be trained is within the preset second difference range.

[0076] Specifically, the second neural network model is pre-trained by pre-setting the parameter value ranges for each circuit element. Within these ranges, multiple parameter values ​​for each circuit element can be determined, resulting in a vector of parameter values ​​to be trained. This vector represents the parameter values ​​in vector form, and each vector can contain the parameter values ​​for each circuit element within the pre-defined circuit model structure. For example, the pre-defined circuit model structure may include circuit elements such as inductors, resistors, capacitors, and mutual inductors. These multiple parameter vectors are then used as the training vectors for the second neural network model.

[0077] Before training the second neural network model, a transformer circuit netlist is pre-compiled. This netlist is a text file that represents the connection relationships of various circuit components in the pre-defined circuit model structure. Based on the constraints of the selected process characteristics and the size range of the transformer coils, the parameter value ranges for each circuit component are determined. Within these ranges, the parameter values ​​are transformed to generate a dataset of component parameters. Multiple component parameter vectors can be obtained from this dataset. The number of possible values ​​for each component parameter can be adjusted based on the network training accuracy requirements of the dataset size. The netlist is used to simulate the aforementioned component parameter vectors, obtaining the multi-port S-parameters of the transformer corresponding to each component parameter vector. Data processing is then performed on the S-parameters to obtain intermediate feature vectors.

[0078] The component parameter vectors used to generate S-parameters are used as the component parameter vectors to be trained. The second neural network model is trained based on these component parameter vectors and the corresponding intermediate feature vectors. Using the intermediate feature vectors as input and the component parameter vectors to be trained as output, a subset of intermediate feature vectors and their corresponding component parameter vectors are selected as the training set, and the remaining samples as the test set. The parameters of the second neural network model and the size of the dataset are adjusted until the difference between the output of the second neural network model and the corresponding component parameter vector to be trained is within a preset second difference range. At this point, the training of the second neural network model is considered complete, thus establishing the functional relationship between the intermediate feature vectors and the component parameters. The number of hidden layers in the second neural network model can be adjusted, and activation functions, loss functions, and optimizers can be selected. The learning rate and number of iterations can be determined, and training continues until the error requirements are met.

[0079] The advantage of this setup is that by pre-setting the netlist, the correct S-parameters corresponding to the component parameter vectors can be determined, which makes it easier to obtain the correct intermediate feature vectors based on the correct S-parameters and improve the training accuracy of the second neural network model.

[0080] It is worth noting that the first neural network model and the second neural network model are trained independently. Therefore, the S-parameters obtained through the electromagnetic simulation platform may not be the same as the S-parameters obtained through the netlist. That is, the intermediate feature vectors when training the first neural network model may not be the same as the intermediate feature vectors when training the second neural network model.

[0081] In this embodiment, the scattering parameters include scattering parameter matrices at at least two frequency points within a preset operating frequency range. Determining the intermediate feature vector of the transformer based on its scattering parameters includes: converting the scattering parameter matrices at at least two frequency points into admittance parameter matrices; determining the equivalent capacitance value of each transformer port based on the admittance parameter matrices at at least two frequency points, and using the equivalent capacitance value of each transformer port at at least two frequency points as elements of the intermediate feature vector; obtaining a local admittance parameter matrix from the admittance parameter matrix based on a preset port, and converting the local admittance parameter matrix into an impedance parameter matrix; wherein the local admittance parameter matrix includes elements of the admittance parameter matrix of the preset port; and determining the mutual inductance data between the preset ports based on the impedance parameter matrix, and using the mutual inductance data as elements of the intermediate feature vector.

[0082] Specifically, the processing procedure for the S-parameters obtained from the electromagnetic simulation platform is the same as that for the S-parameters obtained from the netlist. The S-parameters of each port of the transformer are obtained through simulation using the electromagnetic simulation platform or netlist. The S-parameters are multi-dimensional matrices containing multiple frequency points. If the transformer has six ports, the S-parameters are a six-dimensional matrix containing multiple frequency points. For example, if the transformer's device simulation frequency is 0.5 to 100 GHz with a step size of 0.5 GHz, there are 200 frequency points, and the S-parameters are a 6×6×200 matrix, meaning each frequency point corresponds to a 6×6 matrix. Based on the transformer's operating frequency range, at least two frequency points can be selected; for example, three frequency points can be selected: low frequency, mid frequency, and high frequency. The S-parameters at each frequency point are then converted into Y-parameters and Z-parameters, both of which have the same dimensions as the S-parameters.

[0083] The S-parameter matrices at different frequencies are converted into Y-parameter matrices. The elements of the Y-parameter matrix are complex numbers, with real numbers representing the real part and imaginary numbers representing the imaginary part. Both the real and imaginary parts of each element in the Y-parameter matrix can be used as elements of the intermediate eigenvector. Ya can be used to represent the Y-parameter matrix after the S-parameter transformation. If the transformer has six ports, for a selected frequency, the elements in the Ya-parameter matrix are Ya... ij i and j represent the transformer ports, with i taking values ​​from 1 to 6 and j taking values ​​from 1 to 6. The parameter matrix of Ya can be represented as:

[0084] Ya = [Ya] 11 Ya 12 Ya 13 Ya 14 Ya 15 Ya 16

[0085] Ya 21 Ya 22 Ya 23 Ya24 Ya 25 Ya 26

[0086] Ya 31 Ya 32 Ya 33 Ya 34 Ya 35 Ya 36

[0087] Ya 41 Ya 42 Ya 43 Ya 44 Ya 45 Ya 46

[0088] Ya 51 Ya 52 Ya 53 Ya 54 Ya 55 Ya 56

[0089] Ya 61 Ya 62 Ya 63 Ya 64 Ya 65 Ya 66 ]

[0090] The real and imaginary parts of each element in the selected Ya parameter matrix at each frequency point are used as part of the intermediate feature vector. A Yj can also be defined to represent the feature data of the substrate parasitic network at each port; the formula for calculating Yj is:

[0091]

[0092] In the admittance parameter matrix at at least two frequency points, the sum of each column's elements is determined, and the real and imaginary parts of the sum of each column's elements are used as elements of the intermediate eigenvector. That is, Yj is calculated at each frequency point, and the real and imaginary parts of each Yj at each frequency point are used as elements of the intermediate eigenvector. For example, if three frequency points are selected, and six Yj can be calculated for each frequency point, then there are a total of 18 Yj.

[0093] Based on the Y-parameter matrix (Ya) at each frequency point, the equivalent capacitance value of each transformer port is determined, and the equivalent capacitance value of each port at each frequency point is used as an element of the intermediate feature vector. The calculation formula for the equivalent capacitance value can be preset, and the equivalent capacitance value is determined according to the preset formula. Since the equivalent capacitance value is calculated from the imaginary part of the elements in Ya, it is not a complex number and can be directly used as an element in the intermediate feature vector.

[0094] The six-port transformer can be configured to have ports P5 and P6 grounded, thus becoming a four-port transformer. The ungrounded ports are the preset ports, i.e., ports P1 to P4. Elements from the Ya matrix where i and j are 1 to 4 are taken to construct the four-port Y-parameter matrix Yb. Yb is the local admittance parameter matrix. Transforming Yb into the Z-parameter matrix Zb, Zb can be expressed as:

[0095] Zb = [Zb] 11 Zb 12 Zb 13 Zb 14

[0096] Zb 21 Zb 22 Zb 23 Zb 24

[0097] Zb 31 Zb 32 Zb 33 Zb 34

[0098] Zb 41 Zb 42 Zb 43 Zb 44 ]

[0099] Based on the obtained impedance parameter matrix Zb, and using a preset formula for determining mutual inductance characteristic data, the mutual inductance data between preset ports is calculated, and this mutual inductance data is used as elements of the intermediate feature vector. For example, if the preset ports are P1 to P4, the mutual inductance characteristic data between each port could be the mutual inductance characteristic data between P1 and P2, P1 and P3, P1 and P4, P3 and P4, etc.

[0100] The state of each port can also be preset, for example, opening ports P5 and P6 and short-circuiting ports P2 and P4. The Ya matrix is ​​converted into a 6-dimensional Z-parameter matrix Za. Elements with values ​​i = 1 to 4 and j = 1 to 4 are taken from Za to obtain the four-port Z-parameter matrix Zc. The Zc matrix is ​​converted into a four-port Y-parameter matrix Yc. Elements with values ​​i = 1 and 3 and j = 1 and 3 are taken from Yc to form the two-port Y-parameter matrix Yd. Yd is converted into a two-port Z-parameter matrix Zd, which can be represented as:

[0101] Zd = [Zd] 11 Zd 13

[0102] Zd 31 Zd 33 ]

[0103] Based on the Zd matrix and a pre-defined formula for calculating feature parameters, the feature parameters are determined and used as elements in the intermediate feature vector. For example, five feature parameters are pre-defined: Lp, Ls, Qp, Qs, and k. The formula for calculating these feature parameters is as follows:

[0104]

[0105]

[0106]

[0107]

[0108]

[0109] Where w is the selected frequency.

[0110] The advantage of this setup is that by determining multiple elements in the intermediate feature vector based on the S-parameters, it is easier to adjust each element in the intermediate feature vector. This allows the first and second neural network models to be trained based on the intermediate feature vector, thereby training appropriate functional relationships between coil parameters and intermediate feature vectors, as well as between intermediate feature vectors and component parameters, thus improving the accuracy of the transformer circuit model.

[0111] In this embodiment, the equivalent capacitance value of each transformer port is determined based on the admittance parameter matrix at at least two frequency points, including: determining each transformer port as a first port, and determining the port that is not short-circuited with the first port as a second port; determining the element corresponding to the first port when the second port is short-circuited from the admittance parameter matrix as the target admittance parameter element; adding the imaginary parts of the target admittance parameter elements and dividing them by the frequency point value of the corresponding frequency point to obtain the equivalent capacitance value of the first port at at least two frequency points.

[0112] Specifically, the admittance parameter matrix at at least two frequency points is a Ya matrix. The equivalent capacitance value of each port at each frequency point is determined based on the elements of the Ya matrix. Each port is sequentially designated as the first port, and the port not short-circuited to the first port is designated as the second port. For example, Figure 3 In the given equation, if the first port is P1, then the second ports are P3, P4, and P6; if the first port is P2, then the second ports are P3, P4, and P6; if the first port is P3, then the second ports are P1, P2, and P5; if the first port is P4, then the second ports are P1, P2, and P5; if the first port is P5, then the second ports are P3, P4, and P6; and if the first port is P6, then the second ports are P1, P2, and P5.

[0113] Ya ij This represents the value of the Y parameter corresponding to port j when port i is short-circuited, i.e., Ya. ij Let Ya be the element corresponding to j when port i is short-circuited in the Ya matrix. ij The value is the current at port i divided by the voltage at port j. When the second port in the Ya matrix is ​​short-circuited, the element corresponding to the first port is determined as the target admittance parameter element. The imaginary parts of multiple target admittance parameter elements corresponding to the first port are added together, and the sum is divided by the selected frequency value to obtain the equivalent capacitance value of the first port at each frequency. After obtaining the sum and dividing by the frequency value, the negative of the result can be determined as the equivalent capacitance value of the first port. For example, for... Figure 3 The formula for calculating the equivalent capacitance Cj of port P1 in the figure is as follows:

[0114]

[0115] Where C1 is the equivalent capacitance value of the first port.

[0116] The beneficial effect of this setting is that by using Cj at different frequency points as part of the intermediate feature vector, the functional relationship between the coil parameters and the intermediate feature vector, as well as the functional relationship between the intermediate feature vector and the component parameters, is fully considered, thereby improving the training accuracy of the model and thus improving the determination accuracy of the transformer circuit model.

[0117] In this embodiment, determining the mutual inductance data between preset ports based on the impedance parameter matrix includes: determining the impedance parameter elements between every two preset ports based on the impedance parameter matrix; dividing the imaginary part of the impedance parameter elements between every two preset ports by the frequency value of the corresponding frequency point to obtain the mutual inductance data between every two preset ports at at least two frequency points.

[0118] Specifically, based on the obtained Z-parameter matrix Zb, the mutual inductance data between preset ports is calculated. For example, if the preset ports are P1 to P4, the mutual inductance characteristic data between each port can be the mutual inductance characteristic data between P1 and P2, the mutual inductance characteristic data between P1 and P3, the mutual inductance characteristic data between P1 and P4, the mutual inductance characteristic data between P3 and P4, etc. Figure 3 The circuit in the diagram has a symmetrical structure, with P2 and P1 being symmetrical. Therefore, it is possible to determine the mutual inductance characteristics between P1 and P3, as well as between P1 and P4, without needing to determine the mutual inductance characteristics between P2 and P3, or between P2 and P4.

[0119] There are at least two preset ports. When determining the mutual inductance characteristic data between each pair of preset ports, the impedance parameter elements between the two preset ports can be obtained from the Zb matrix. Dividing the imaginary part of the impedance parameter elements between these two preset ports by the frequency value of each selected frequency point yields the mutual inductance characteristic data Mij between each pair of preset ports at each frequency point. For example, the formula for determining the preset mutual inductance characteristic data can be:

[0120]

[0121] M12 represents the mutual inductance characteristic data between ports P1 and P2.

[0122] The advantage of this setting is that by considering the mutual inductance feature data between the two preset ports as the intermediate feature vector, the functional relationship between the coil parameters and the intermediate feature vector, as well as the functional relationship between the intermediate feature vector and the component parameters, can be fully taken into account, thereby improving the training accuracy of the model and thus improving the determination accuracy of the transformer circuit model.

[0123] S403. Determine the circuit model of the transformer based on the preset correlation and preset circuit model structure. The preset correlation refers to the component parameters of the circuit elements and the preset correlation between the circuit elements. The preset circuit model structure is used to represent the connection relationship between each circuit element in the circuit model.

[0124] For example, this step can refer to step S103 above, and will not be repeated here.

[0125] This application provides a method for modeling a transformer circuit. By obtaining the width and radius of each coil in the transformer, the coil parameters of the transformer coils are obtained. Based on the coil parameters, the component parameters of each circuit element corresponding to the coil parameters are determined. Based on the connection relationships of each circuit element in the transformer circuit and the component parameters of the circuit elements, the transformer circuit is obtained. This solves the problem of low efficiency and accuracy in determining component parameters caused by the sequential extraction of component parameters for each circuit element in the prior art. By directly obtaining the component parameters of each circuit element from the coil parameters, manpower and time are effectively saved, and the efficiency and accuracy of transformer circuit determination are improved.

[0126] Figure 5 This is a structural block diagram of a transformer circuit modeling apparatus provided in an embodiment of this application. For ease of explanation, only the parts relevant to the embodiments of this disclosure are shown. (Refer to...) Figure 5 The device includes: a coil parameter acquisition module 501, a component parameter determination module 502, and a circuit model determination module 503.

[0127] The coil parameter acquisition module 501 is used to acquire the coil parameters of the transformer coil; wherein the transformer coil includes a primary coil and a secondary coil, and the coil parameters include the width of the primary coil, the radius of the primary coil, the width of the secondary coil, and the radius of the secondary coil;

[0128] The component parameter determination module 502 is used to determine the component parameters of the circuit elements in the transformer circuit based on the coil parameters; wherein the transformer circuit includes at least two circuit elements;

[0129] The circuit model determination module 503 is used to determine the circuit model of the transformer based on preset association relationships and preset circuit model structures. The preset association relationships are the component parameters of the circuit elements and the preset association relationships between the circuit elements. The preset circuit model structure is used to represent the connection relationships between the circuit elements in the circuit model.

[0130] Figure 6 This application provides a structural block diagram of a transformer circuit modeling device according to an embodiment of the present application. Figure 5 Based on the illustrated embodiments, as Figure 6 As shown, the circuit model determination module 503 includes a parameter allocation unit 5031 and a model determination unit 5032.

[0131] The parameter allocation unit 5031 is used to allocate the component parameters to the corresponding circuit components according to the preset association relationship, so as to obtain the component parameters of each circuit component in the preset circuit model structure.

[0132] The model determination unit 5032 is used to obtain the circuit model of the transformer based on the preset circuit model structure and the component parameters of each circuit element in the preset circuit model structure.

[0133] In one example, the component parameter determination module 502 is specifically used for:

[0134] The coil parameters are input into a preset component parameter determination model, and the component parameters of the circuit elements in the transformer circuit are output; wherein, the component parameter determination model has a preset functional relationship between the coil parameters and the component parameters.

[0135] In one example, component parameter determination module 502 includes:

[0136] The first model output unit is used to input the coil parameters into a preset first neural network model and output an intermediate feature vector of the transformer corresponding to the coil parameters; wherein, the first neural network model has a preset functional relationship between the coil parameters and the intermediate feature vector, and the intermediate feature vector is used to represent the transformer characteristics;

[0137] The second model output unit is used to input the intermediate feature vector into a preset second neural network model and output the component parameters of the circuit elements in the transformer circuit corresponding to the intermediate feature vector; wherein, the second neural network model has a preset functional relationship between the intermediate feature vector and the component parameters.

[0138] In one example, the device further includes:

[0139] The coil parameter vector acquisition module is used to determine the coil parameter vector to be trained based on the value range of the coil parameters of the transformer coil before inputting the coil parameters into the preset first neural network model and outputting the intermediate feature vector of the transformer corresponding to the coil parameters.

[0140] The electromagnetic simulation module is used to determine the scattering parameters of the transformer based on the parameter vector of the coil to be trained and the preset electromagnetic simulation platform of the transformer.

[0141] An intermediate vector determination module is used to determine the intermediate feature vector of the transformer based on the scattering parameters of the transformer.

[0142] The first model training module is used to input the coil parameter vector to be trained into a preset first neural network model. If the difference between the output of the first neural network model and the intermediate feature vector is within a preset first difference range, then the training of the first neural network model is determined to be complete.

[0143] In one example, the device further includes:

[0144] The component parameter vector acquisition module is used to determine the component parameter vector to be trained based on the value range of the component parameters of the circuit elements in the transformer circuit before inputting the intermediate feature vector into the preset second neural network model and outputting the component parameters of the circuit elements in the transformer circuit corresponding to the intermediate feature vector.

[0145] The scattering parameter determination module is used to determine the scattering parameters of the transformer based on the parameter vector of the component to be trained and the preset transformer circuit netlist.

[0146] An intermediate vector determination module is used to determine the intermediate feature vector of the transformer based on the scattering parameters of the transformer.

[0147] The second model training module is used to input the intermediate feature vector into a preset second neural network model. If the difference between the output of the second neural network model and the parameter vector of the element to be trained is within a preset second difference range, then the training of the second neural network model is determined to be complete.

[0148] In one example, the scattering parameters include scattering parameter matrices at at least two frequency points within a preset operating frequency range;

[0149] The intermediate vector determination module is specifically used for:

[0150] Based on the transformer's scattering parameters, the intermediate feature vector of the transformer is determined, including:

[0151] Convert the scattering parameter matrix at at least two frequency points into the admittance parameter matrix;

[0152] Based on the admittance parameter matrix at at least two frequency points, determine the equivalent capacitance value of each transformer port, and determine the equivalent capacitance value of each transformer port at at least two frequency points as the element of the intermediate feature vector;

[0153] The admittance parameters of a preset port are obtained from the admittance parameter matrix and used as a local admittance parameter matrix. The local admittance parameter matrix is ​​then converted into an impedance parameter matrix. The local admittance parameter matrix includes elements of the admittance parameter matrix of the preset port.

[0154] Based on the impedance parameter matrix, the mutual inductance data between preset ports is determined, and the mutual inductance data is used as an element of the intermediate feature vector.

[0155] In one example, the intermediate vector determination module is also specifically used for:

[0156] Each transformer port is designated as the first port, and the port that is not short-circuited to the first port is designated as the second port;

[0157] The elements corresponding to the first port when the second port is short-circuited are determined from the admittance parameter matrix, and these are the target admittance parameter elements.

[0158] The imaginary parts of the target admittance parameter elements are added together and divided by the frequency value of the corresponding frequency point to obtain the equivalent capacitance value of the first port at at least two frequency points.

[0159] In one example, the intermediate vector determination module is also specifically used for:

[0160] Based on the impedance parameter matrix, determine the impedance parameter elements between every two preset ports;

[0161] Divide the imaginary part of the impedance parameter element between each pair of preset ports by the frequency value of the corresponding frequency point to obtain the mutual inductance data between each pair of preset ports at at least two frequency points.

[0162] Figure 7 A structural block diagram of an electronic device provided in an embodiment of this application, such as... Figure 7 As shown, the electronic device includes: a memory 71 and a processor 72; the memory 71 is a memory for storing executable instructions of the processor 72.

[0163] The processor 72 is configured to perform the method provided in the above embodiments.

[0164] The electronic device also includes a receiver 73 and a transmitter 74. The receiver 73 is used to receive instructions and data sent by other devices, and the transmitter 74 is used to send instructions and data to external devices.

[0165] Figure 8 This is a block diagram illustrating an electronic device according to an exemplary embodiment. The device may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness device, personal digital assistant, etc.

[0166] Device 800 may include one or more of the following components: processing component 802, memory 804, power supply component 806, multimedia component 808, audio component 810, input / output (I / O) interface 812, sensor component 814, and communication component 816.

[0167] Processing component 802 typically controls the overall operation of device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

[0168] Memory 804 is configured to store various types of data to support the operation of device 800. Examples of this data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0169] Power supply component 806 provides power to various components of device 800. Power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to device 800.

[0170] Multimedia component 808 includes a screen that provides an output interface between the device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0171] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0172] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0173] Sensor assembly 814 includes one or more sensors for providing status assessments of various aspects of device 800. For example, sensor assembly 814 may detect the on / off state of device 800, the relative positioning of components such as the display and keypad of device 800, changes in the position of device 800 or a component of device 800, the presence or absence of user contact with device 800, the orientation or acceleration / deceleration of device 800, and temperature changes of device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0174] Communication component 816 is configured to facilitate wired or wireless communication between device 800 and other devices. Device 800 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0175] In an exemplary embodiment, device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0176] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, which can be executed by a processor 820 of device 800 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0177] A non-transitory computer-readable storage medium, wherein when the instructions in the storage medium are executed by the processor of a terminal device, the terminal device is able to execute the aforementioned modeling method for the transformer circuit of the terminal device.

[0178] This application also discloses a computer program product, including a computer program that, when executed by a processor, implements the method described in this embodiment.

[0179] Various embodiments of the systems and technologies described above in this application can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0180] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or electronic device.

[0181] In the context of this application, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0182] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0183] The systems and technologies described herein can be implemented in computing systems that include back-end components (e.g., as data electronic devices), or computing systems that include middleware components (e.g., application electronic devices), or computing systems that include front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0184] Computer systems can include client and electronic devices. Clients and electronic devices are generally geographically separated and typically interact via communication networks. The client-electronic device relationship is created by computer programs running on the respective computers and having a client-electronic device relationship with each other. The electronic device can be a cloud electronic device, also known as a cloud computing electronic device or cloud host, a host product within the cloud computing service system, addressing the shortcomings of traditional physical hosts and VPS services ("Virtual Private Server," or simply "VPS") in terms of management difficulty and weak business scalability. The electronic device can also be an electronic device in a distributed system or an electronic device incorporating blockchain technology. It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application is achieved, and this is not limited herein.

[0185] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0186] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method of modeling a transformer circuit, characterized by, The method comprises the following steps: obtaining coil parameters of a transformer coil; wherein the transformer coil comprises a primary coil and a secondary coil, and the coil parameters comprise a width of the primary coil, a radius of the primary coil, a width of the secondary coil, and a radius of the secondary coil; determining element parameters of circuit elements in the transformer circuit according to the coil parameters, comprising: inputting the coil parameters into a preset element parameter determination model to output the element parameters of the circuit elements in the transformer circuit, wherein a functional relationship between the coil parameters and the element parameters is preset in the element parameter determination model; wherein the transformer circuit comprises at least two circuit elements; determining a circuit model of the transformer according to a preset correlation relationship and a preset circuit model structure, wherein the preset correlation relationship is a correlation relationship between the element parameters of the circuit elements and preset circuit elements, and the preset circuit model structure is used to represent a connection relationship between the circuit elements in the circuit model; inputting the coil parameters into a preset element parameter determination model to output the element parameters of the circuit elements in the transformer circuit, comprising: inputting the coil parameters into a preset first neural network model to output an intermediate feature vector of the transformer corresponding to the coil parameters; wherein a functional relationship between the coil parameters and the intermediate feature vector is preset in the first neural network model, and the intermediate feature vector is used to represent a transformer characteristic; inputting the intermediate feature vector into a preset second neural network model to output the element parameters of the circuit elements in the transformer circuit corresponding to the intermediate feature vector; wherein a functional relationship between the intermediate feature vector and the element parameters is preset in the second neural network model; before inputting the coil parameters into the preset first neural network model to output the intermediate feature vector of the transformer corresponding to the coil parameters, further comprising: determining a coil parameter vector to be trained according to a coil parameter value range of the transformer coil; determining a scattering parameter of the transformer according to the coil parameter vector to be trained and a preset electromagnetic simulation platform of the transformer; determining the intermediate feature vector of the transformer according to the scattering parameter of the transformer; inputting the coil parameter vector to be trained into the preset first neural network model, and if a difference between an output of the first neural network model and the intermediate feature vector is within a preset first difference range, determining that the first neural network model is trained; the scattering parameter comprises scattering parameter matrices at at least two frequency points in a preset working frequency range, and the intermediate feature vector of the transformer is determined according to the scattering parameter of the transformer, comprising: converting the scattering parameter matrices at the at least two frequency points into admittance parameter matrices; determining equivalent capacitance values of each transformer port according to the admittance parameter matrices at the at least two frequency points, and determining the equivalent capacitance values of each transformer port at the at least two frequency points as elements of the intermediate feature vector. Obtaining the admittance parameter of the preset port from the admittance parameter matrix as a local admittance parameter matrix, and converting the local admittance parameter matrix into an impedance parameter matrix; wherein the local admittance parameter matrix includes the admittance parameter matrix element of the preset port; According to the impedance parameter matrix, the mutual inductance data between the preset ports is determined, and the mutual inductance data is determined as an element of the intermediate feature vector.

2. The method of claim 1, wherein, Before inputting the intermediate feature vector into a preset second neural network model and outputting the element parameter of the circuit element in the transformer circuit corresponding to the intermediate feature vector, the method further comprises: According to the element parameter value range of the circuit element in the transformer circuit, the element parameter vector to be trained is determined; According to the element parameter vector to be trained and the preset transformer circuit netlist, the scattering parameter of the transformer is determined; According to the scattering parameter of the transformer, the intermediate feature vector of the transformer is determined; The intermediate feature vector is input into a preset second neural network model, and if the difference between the output of the second neural network model and the element parameter vector to be trained is within a preset second difference range, it is determined that the training of the second neural network model is completed.

3. The method of claim 1, wherein, According to the admittance parameter matrix at least two frequency points, the equivalent capacitance value of each transformer port is determined, comprising: Each transformer port is determined as a first port, and a port not short-circuited with the first port is determined as a second port; The element corresponding to the first port when the second port is short-circuited is determined from the admittance parameter matrix as a target admittance parameter element; The imaginary part of the target admittance parameter element is added and divided by the frequency point value of the corresponding frequency point to obtain the equivalent capacitance value of the first port at least two frequency points.

4. The method of claim 1, wherein, According to the impedance parameter matrix, the mutual inductance data between the preset ports is determined, comprising: According to the impedance parameter matrix, the impedance parameter element between each two preset ports is determined; The imaginary part of the impedance parameter element between each two preset ports is respectively divided by the frequency point value of the corresponding frequency point to obtain the mutual inductance data between each two preset ports at least two frequency points.

5. The method of claim 1, wherein, According to the preset correlation and the preset circuit model structure, the circuit model of the transformer is determined, comprising: According to the preset correlation, the element parameter is assigned to the corresponding circuit element to obtain the element parameter of each circuit element in the preset circuit model structure; According to the preset circuit model structure and the element parameter of each circuit element in the preset circuit model structure, the circuit model of the transformer is obtained.

6. A modeling apparatus for a transformer circuit, characterized by Comprise: The coil parameter acquisition module is used for acquiring the coil parameter of the transformer coil; wherein the transformer coil includes a primary coil and a secondary coil, and the coil parameter includes the width of the primary coil, the radius of the primary coil, the width of the secondary coil, and the radius of the secondary coil; The element parameter determination module is configured to input the coil parameter into a preset element parameter determination model, and output an element parameter of a circuit element in the transformer circuit; wherein the transformer circuit comprises at least two circuit elements, and the element parameter determination model is preset with a functional relationship between the coil parameter and the element parameter; The circuit model determination module is configured to determine a circuit model of the transformer according to a preset correlation relationship and a preset circuit model structure, wherein the preset correlation relationship is a correlation relationship between the element parameter of the circuit element and a preset circuit element, and the preset circuit model structure is used to represent a connection relationship between circuit elements in the circuit model; The element parameter determination module comprises: The first model output unit is configured to input the coil parameter into a preset first neural network model, and output an intermediate feature vector of the transformer corresponding to the coil parameter; wherein the first neural network model is preset with a functional relationship between the coil parameter and the intermediate feature vector, and the intermediate feature vector is used to represent a transformer characteristic; The second model output unit is configured to input the intermediate feature vector into a preset second neural network model, and output an element parameter of a circuit element in the transformer circuit corresponding to the intermediate feature vector; wherein the second neural network model is preset with a functional relationship between the intermediate feature vector and the element parameter; Before the coil parameter is input into the preset first neural network model and the intermediate feature vector of the transformer corresponding to the coil parameter is output, the method further comprises: determining a coil parameter vector to be trained according to a coil parameter value range of a transformer coil; determining a scattering parameter of the transformer according to the coil parameter vector to be trained and a preset electromagnetic simulation platform of the transformer; determining the intermediate feature vector of the transformer according to the scattering parameter of the transformer; inputting the coil parameter vector to be trained into the preset first neural network model, and determining that the first neural network model is trained if a difference between an output of the first neural network model and the intermediate feature vector is within a preset first difference range; The scattering parameter comprises scattering parameter matrices at at least two frequency points in a preset working frequency range, and the intermediate feature vector of the transformer is determined according to the scattering parameter of the transformer, comprising: converting the scattering parameter matrices at the at least two frequency points into admittance parameter matrices; determining equivalent capacitance values of each transformer port according to the admittance parameter matrices at the at least two frequency points, and determining the equivalent capacitance values of each transformer port at the at least two frequency points as elements of the intermediate feature vector; obtaining an admittance parameter of a preset port from the admittance parameter matrix as a local admittance parameter matrix, converting the local admittance parameter matrix into an impedance parameter matrix; wherein the local admittance parameter matrix comprises an admittance parameter matrix element of the preset port; determining mutual inductance data between preset ports according to the impedance parameter matrix, and determining the mutual inductance data as elements of the intermediate feature vector.

7. An electronic device, comprising: comprise: a processor, and a memory connected with the processor in communication; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the modeling method of the transformer circuit according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by the processor to implement the modeling method of the transformer circuit according to any one of claims 1-5.

9. A computer program product, characterised in that, The computer program is executed by the processor to implement the modeling method of the transformer circuit according to any one of claims 1-5.

Images