A method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data

By solving the near-field magnetic field scanning and regularized objective function of strong electromagnetic equipment, a low-frequency radiation field model is constructed, which solves the modeling problem of low-frequency magnetic field radiation interference of strong electromagnetic equipment, and realizes high-precision electromagnetic compatibility performance evaluation and radiation interference predict.

CN114925513BActive Publication Date: 2024-12-06CHONGQING UNIV +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210516453.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-12
Publication Date
2024-12-06
Estimated Expiration
2042-05-12

AI Technical Summary

Technical Problem

In the field of high-tech electromagnetic technology such as aircraft and ships, the low-frequency magnetic field radiation interference of strong electromagnetic equipment is severe, and the existing technology is difficult to effectively model and predict, which affects electromagnetic compatibility performance and lacks a high-precision low-frequency equivalent radiation model.

Method used

By scanning the near-field magnetic field of strong electromagnetic equipment, establishing the near-field magnetic field equation, using regularized objective function and singular value decomposition method to solve the dipole matrix of the magnetic dipole array, and then building a low-frequency radiation field model, using equivalent Dipole array method reconstructs radiation sources.

Benefits of technology

It realizes high-precision modeling of low-frequency radiation interference of strong electromagnetic equipment without relying on the internal structure and circuit topology of the equipment, improving the efficiency of electromagnetic compatibility performance evaluation and radiation interference prediction, and is suitable for ship and aircraft platforms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114925513B_ABST
    Figure CN114925513B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data, comprising the following steps: 1) scanning the near-field magnetic field generated by the strong electromagnetic equipment to establish a near-field magnetic field equation; 2) establishing a regularized objective function according to the near-field magnetic field equation; 3) solving the dipole moment matrix of the magnetic dipole using the regularized objective function; 4) solving the low-frequency radiation field generated by the magnetic dipole array. The present invention obtains the near-field information of strong electromagnetic equipment on platforms such as ships through near-field measurement, wherein the information includes phase and amplitude, and establishes an equivalent dipole array based on the near-field information, and can obtain the magnetic field distribution of a surface at a certain distance from the strong electromagnetic equipment or the magnetic field at a certain point, thereby greatly improving the efficiency of evaluating the radiation interference of strong electromagnetic equipment on platforms such as ships.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of electromagnetic compatibility, and in particular to a method for modeling low-frequency radiation equivalent of strong electromagnetic equipment based on test data. Background Art

[0002] In recent years, with the rapid development of new materials, new power devices and other technologies, high-tech electromagnetic technologies such as integrated power technology and electromagnetic launch technology in the fields of aircraft, ships, etc. have been vigorously launched, and are developing towards all-electric propulsion, multi-electric and even all-electric aircraft technology. In particular, countries around the world are conducting in-depth research on integrated all-electric propulsion systems for ships. Foreign countries have developed various types of integrated all-electric propulsion systems and applied them on various types of ships. The integrated all-electric propulsion system for ships includes many functions such as power generation, transmission, distribution, transformation, dragging, propulsion, energy storage, monitoring and power management. The electromagnetic energy required in new aircraft and ship platforms has greatly increased. It is necessary to realize the transformation, conversion, distribution and use of ultra-large capacity electromagnetic energy in a limited platform space. The complexity of multiple systems and multiple functions has brought serious electromagnetic compatibility problems. The electromagnetic compatibility performance is related to whether the various functional modules of the integrated all-electric propulsion system operate well and whether they are well coordinated with each other, and thus it is also related to whether the entire integrated all-electric propulsion system can have a good operating state and excellent working performance.

[0003] At the same time, with the increase in device switching frequency and the widespread use of digital control and pulse width modulation technology in energy conversion / conversion, on the one hand, the strong electromagnetic equipment on the platform will generate a large amount of broadband interference emissions, and on the other hand, the anti-interference ability of the low-voltage part of the strong electromagnetic equipment will be reduced.

[0004] In addition, the low-frequency magnetic field radiation interference of strong electromagnetic equipment is becoming more and more serious. For this reason, it is necessary to find a general method for each strong electromagnetic equipment in the ship's cabin to construct a modeling method to characterize the low-frequency radiation of each strong electromagnetic equipment, which can be used for ship electromagnetic environment prediction and simulation, so as to take electromagnetic compatibility control measures such as electromagnetic shielding for places with strong radiation interference, so as not to affect the normal operation of other equipment in the cabin; at the same time, using the low-frequency radiation model of strong electromagnetic equipment, the low-frequency magnetic field distribution characteristics of strong electromagnetic equipment in the platform cabin can be predicted, and the layout design and optimization of cabin equipment can be supported under the premise of considering the threshold of low-frequency sensitive equipment. Considering that the specific internal structure, circuit topology results, and electrical connection information of strong electromagnetic equipment are usually unknown, it is necessary to be able to invert and establish a high-precision low-frequency equivalent radiation model of strong electromagnetic equipment based on the near-field measurement information near the strong electromagnetic equipment. Summary of the invention

[0005] The object of the present invention is to provide a method for modeling low-frequency radiation equivalent of strong electromagnetic equipment based on test data, comprising the following steps:

[0006] 1) Scan the near-field magnetic field generated by strong electromagnetic equipment and establish the near-field magnetic field equation;

[0007] The magnetic dipole array includes N magnetic dipoles, wherein the coordinates of the nth magnetic dipole are (x n ,y n , z n ); n=1,2,...,N.

[0008] The number of scanning points for scanning the near-field magnetic field generated by the strong electromagnetic device is M, where the coordinates of the mth near-field magnetic field scanning point are (x m ,y m , z m ), m=1,2,...,M.

[0009] The near-field magnetic field equation is shown below:

[0010] Ax=b (1)

[0011] Among them, matrix A, matrix b and matrix x are as follows:

[0012]

[0013]

[0014] In the formula, represents the x-component of the magnetic dipole moment of the nth magnetic dipole, H nx represents the x-component of the magnetic field at the nth near-field test data point; Represents the conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the x-component of the magnetic field at the n'-th near-field magnetic field scanning point; n'=1,2,...,N;

[0015] Among them, the conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the x-component of the radiated magnetic field at the n'th point is Conversion coefficient between the magnetic dipole moment of the y-component of the n-th magnetic dipole and the x-component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the z component of the nth magnetic dipole and the x component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the y-component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the y component of the nth magnetic dipole and the y component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the z component of the nth magnetic dipole and the y component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the z-component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the y component of the nth magnetic dipole and the z component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the z component of the nth magnetic dipole and the z component of the radiated magnetic field at the n'th point They are as follows:

[0016]

[0017]

[0018]

[0019]

[0020]

[0021]

[0022]

[0023]

[0024]

[0025] in, is the free space wave number, λ is the wavelength, |r-r'|=r, r' is the coordinate vector of magnetic dipole numbered n, and r is the coordinate vector of near-field measurement point n'. λ is the wavelength; r is the modulus of the position vector of the field point; ω is the angular frequency; ε0 is the dielectric constant of air; μ0 is the magnetic permeability of air.

[0026] 2) Establish a regularized objective function based on the near-field magnetic field equation;

[0027] The regularized objective function is as follows:

[0028] min||Ax-b|| 2 +λ 2 ||x|| 2 ,x∈R n (14)

[0029] Among them, λ is the regularization parameter.

[0030] 3) Solve the dipole moment matrix of the magnetic dipole using the regularized objective function;

[0031] The steps of solving the magnetic dipole coordinates using the regularized objective function include:

[0032] 3.1) Calculate the regularization parameter λ;

[0033] The steps to calculate the regularization parameter λ include:

[0034] 3.1.1) Update the regularized objective function (14) and obtain:

[0035]

[0036] Where b i A represents the new matrix after deleting the i-th element from matrix b; (i) Represents the new matrix after deleting the i-th element from matrix A; Represents the new matrix after deleting the i-th element from the matrix x;

[0037] 3.1.2) Introduce the diagonal element t ii , update the regularized objective function (15) and obtain:

[0038]

[0039] Where, t ii is the matrix A(A T A+λ 2 I) -1 A T The diagonal elements of ;

[0040] 3.1.3) Use the average value of the diagonal elements instead of t ii , update the regularized objective function (16) and obtain:

[0041]

[0042] The regularized objective function (18) is simplified as follows:

[0043]

[0044] 3.1.4) Using the singular value decomposition method to process the matrix A, we get:

[0045]

[0046] In the formula, σ i is the characteristic root;

[0047] 3.1.5) Calculate the regularization parameter λ, that is:

[0048]

[0049] 3.2) Substitute the regularization parameter λ into the regularization objective function (14) to obtain the dipole moment matrix x of the magnetic dipole, that is:

[0050] x=((A) T A+λ 2 I) -1 (A)T b (21)

[0051] Where I is a unit vector.

[0052] 4) Solve the low-frequency radiation field generated by the magnetic dipole array.

[0053] The steps to solve the low-frequency radiation field generated by the magnetic dipole array on the extrapolated plane include:

[0054] 4.1) Based on the definition of magnetic dipole, establish the spherical coordinate system expression of the electromagnetic field radiated by magnetic dipole;

[0055] The steps to establish the spherical coordinate system expression of the electromagnetic field radiated by the magnetic dipole include:

[0056] 4.1.1) Let the current on the magnetic dipole line element i(t) = I m cosωt=Re[Ie jωt ], the current amplitude is I m ; t is time; ω is angular frequency;

[0057] Establish the components of the electromagnetic field radiated by the magnetic dipole in the spherical coordinate system, namely:

[0058]

[0059] In the formula, is the free space wave number, λ is the wavelength, is the free space wave impedance, r is the modulus of the position vector of the field point; l is the length of the magnetic dipole line element; θ is one of the coordinates of the field point in the spherical coordinate system. In the spherical coordinate system, the coordinates of any point in space are (r, θ, ); is the electric field component; is the magnetic field component;

[0060] 4.1.2) Introducing intermediate parameters f(r) and K e , intermediate parameter K h , intermediate parameter g1(r), intermediate parameter g2(r), intermediate parameter g3(r), simplify formula (22) and get:

[0061]

[0062] In the formula, is the magnetic dipole moment of the magnetic dipole in the z direction.

[0063] Among them, the intermediate parameters f(r) and K e , intermediate parameter K h, intermediate parameters g1(r), intermediate parameters g2(r), and intermediate parameters g3(r) are as follows:

[0064]

[0065] 4.2) The spherical coordinate expression of the electromagnetic field radiated by the magnetic dipole is converted into a rectangular coordinate expression, and the dipole array vector is summed to obtain the low-frequency magnetic field generated by the equivalent dipole array.

[0066] The steps of converting the spherical coordinate system expression of the electromagnetic field radiated by the magnetic dipole into a rectangular coordinate system expression and performing the dipole array vector summation include:

[0067] 4.2.1) Transform equation (24) into a rectangular coordinate system, that is:

[0068]

[0069] 4.2.2) Using the symmetry of the components of the rectangular coordinate system, the magnetic dipole moment M is derived x and M y The corresponding relationship with the magnetic field is calculated from the magnetic dipole M x 、M y and M z The magnetic field generated by the combination of

[0070]

[0071] 4.2.3) The magnetic field generated by each magnetic dipole is vector-summed to obtain the low-frequency radiation field generated by the magnetic dipole array, that is:

[0072]

[0073] In the formula, is the x-component of the magnetic dipole moment of the ith magnetic dipole, is the y component of the magnetic dipole moment of the ith magnetic dipole, is the z component of the magnetic dipole moment of the i-th magnetic dipole;

[0074] Among them, the parameter r i,j As shown below:

[0075]

[0076] x i ,y i 、z i is the coordinate of the ith magnetic dipole, x j ,y j 、z j are the coordinates of the jth near-field measurement point.

[0077] It is worth noting that electric dipoles and magnetic dipoles are the most basic idealized radiation units. The present invention uses an array of electric dipoles or magnetic dipoles to equivalently represent the overall radiation characteristics of the radiation source. According to the radiation electromagnetic field expression of a single electric dipole or magnetic dipole, that is, the relationship between the electric dipole moment or magnetic dipole moment and the electric field or magnetic field, vector summation is used to obtain the radiation electromagnetic field expression of the electric dipole array or magnetic dipole array. Therefore, the relationship between the near-field test data near the strong electromagnetic device and the electric / magnetic dipole array can be expressed as a matrix equation Ax=b, where the coefficient matrix A is related to the coordinates of the near-field measurement and the coordinates of the dipole, the unknown quantity x is the dipole moment of the dipole array, and b is the near-field information obtained by the near-field measurement, including amplitude and phase information. The dipole moment of the dipole array can be obtained by solving the matrix equation. After obtaining the dipole moment, a radiation source model of a strong electromagnetic device based on the equivalent dipole method is obtained.

[0078] The technical effect of the present invention is undoubted, and the present invention has the following advantages:

[0079] 1) The present invention can establish a high-precision equivalent radiation model of a strong electromagnetic device based on near-field measurement information without knowing the specific internal structure, circuit topology results and electrical connections of the strong electromagnetic device. The model can be used to evaluate the radiation interference state of the strong electromagnetic device, and then as a low-frequency radiation interference source, it can be used to predict the distribution of low-frequency radiation fields and low-frequency electromagnetic interference of strong electromagnetic devices on platforms such as ships and aircraft. The present invention has wide applicability and can be used for radiation interference evaluation and analysis on platforms such as ships and aircraft.

[0080] 2) The present invention provides a high-precision modeling method for the low-frequency radiation equivalent dipole array of strong electromagnetic equipment based on test data. The method first obtains the low-frequency radiation near-field information of strong electromagnetic equipment on platforms such as ships through measurement, and establishes an equivalent dipole array radiation model of the radiation field through Tikhonov regularization and generalized cross-validation method; within a certain error, at the position of interest, the equivalent magnetic dipole array radiation model can obtain the same low-frequency magnetic field distribution as the actual strong electromagnetic equipment.

[0081] 3) The present invention obtains near-field information of strong electromagnetic equipment on platforms such as ships through near-field measurement, wherein the information includes phase and amplitude. By using this near-field information to establish an equivalent dipole array, the magnetic field distribution of a surface at a certain distance from the strong electromagnetic equipment or the magnetic field at a certain point can be obtained, thereby greatly improving the efficiency of evaluating the radiation interference of strong electromagnetic equipment on platforms such as ships. The present invention does not rely on the internal structure, circuit topology results, and electrical connections of the strong electromagnetic equipment. This method can be used to obtain an equivalent model of its radiation source, and has strong applicability, is not limited to a certain type of strong electromagnetic equipment, and is particularly suitable for mastering test data of radiation from strong electromagnetic equipment;

[0082] 4) The present invention can not only obtain a high-precision two-dimensional radiation source equivalent model of a strong electromagnetic device, but also a high-precision three-dimensional radiation source equivalent model of a strong electromagnetic device. The present invention provides a least square method combined with Tikhonov regularization and solves the ill-conditioned matrix problem that occurs in the process of reconstructing the radiation source using the equivalent dipole method, which greatly improves the accuracy of the solution; the present invention can achieve high-precision reconstruction of the radiation source equivalent model in the low-frequency range (50Hz~1kHz). BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 This is a flowchart of Tikhonov regularization based on generalized cross-validation method; Figure 2 This is the model of a three-phase synchronous motor in Maxwell3D (the three planes from top to bottom are the extrapolation plane, the measurement plane, and the dipole array plane); Figure 3 (a)-(d) are the magnetic field distribution at 400 mm obtained from simulation data; Figure 4 (a)-(d) are the magnetic field distribution at 400 mm obtained by the equivalent dipole method; Figure 5 is the relative error of the magnetic field;

[0084] Figure 6 is the relative position of the measuring plane and the extrapolated plane (where the inner part is the measuring plane and the outer part is the extrapolated plane); Figure 7 Number the surface for reconstructing the radiated magnetic field; Figure 8 (a)-(d) are the magnetic field distribution of surface 1 obtained from simulation data; Fig. 9 (a)-(d) are the magnetic field distributions of surface 1 obtained by the equivalent dipole method; Fig.10 is the relative error of the magnetic field of surface 1;

[0085] Fig.11 (a)-(d) are the magnetic field distribution of surface 2 obtained from simulation data; Fig.12 (a)-(d) are the magnetic field distributions of surface 2 obtained by the equivalent dipole method; Fig.13 is the relative error of the magnetic field of surface 2; Fig.14 (a) -(d) are the magnetic field distribution of surface 3 obtained by simulation data; Fig.15 (a)-(d) are the magnetic field distributions of surface 3 obtained by the equivalent dipole method;

[0086] Fig.16 is the relative error of the magnetic field of surface 3; Fig.17 (a)-(d) are the magnetic field distribution of surface 4 obtained by simulation data; Fig.18 (a)-(d) are the magnetic field distributions of surface 4 obtained by the equivalent dipole method; Fig.19 is the relative error of the magnetic field of surface 4; Fig. 20 (a)-(d) are the magnetic field distribution of surface 5 obtained by simulation data;

[0087] Fig.21 (a)-(d) are the magnetic field distributions of surface 5 obtained by the equivalent dipole method; Fig. 22 is the relative error of the magnetic field of surface 5; Fig.23 (a)-(d) are the magnetic field distribution of surface 6 obtained by simulation data; Fig.24 (a) -(d) are the magnetic field distributions of surface 6 obtained by the equivalent dipole method; Fig.25 is the relative error of the magnetic field on surface 6. DETAILED DESCRIPTION

[0088] The present invention is further described below in conjunction with the embodiments, but it should not be understood that the above subject matter of the present invention is limited to the following embodiments. Without departing from the above technical ideas of the present invention, various substitutions and changes are made according to the common technical knowledge and customary means in the art, which should all be included in the protection scope of the present invention.

[0089] Embodiment 1:

[0090] See also Figures 1 to 25 , a method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data, comprising the following steps:

[0091] 1) Scan the near-field magnetic field generated by strong electromagnetic equipment and establish the near-field magnetic field equation;

[0092] The magnetic dipole array includes N magnetic dipoles, and the magnetic dipoles are placed along the z-axis, wherein the coordinate of the nth magnetic dipole is (x n ,y n , z n ); n=1,2,...,N.

[0093] The number of scanning points for scanning the near-field magnetic field generated by the strong electromagnetic device is M, where the coordinates of the mth near-field magnetic field scanning point are (x m ,y m , z m ), m=1,2,...,M.

[0094] The near-field magnetic field equation is shown below:

[0095] Ax=b (1)

[0096] Among them, matrix A, matrix b and matrix x are as follows:

[0097]

[0098]

[0099] In the formula, represents the x-component of the magnetic dipole moment of the nth magnetic dipole, H nx represents the x-component of the magnetic field at the nth near-field test data point; Represents the conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the x-component of the magnetic field at the n'-th near-field magnetic field scanning point; n'=1,2,...,N;

[0100] Among them, the conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the x-component of the radiated magnetic field at the n'th point is Conversion coefficient between the magnetic dipole moment of the y-component of the n-th magnetic dipole and the x-component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the z component of the nth magnetic dipole and the x component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the y-component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the y component of the nth magnetic dipole and the y component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the z component of the nth magnetic dipole and the y component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the z-component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the y component of the nth magnetic dipole and the z component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the z component of the nth magnetic dipole and the z component of the radiated magnetic field at the n'th point They are as follows:

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[0107]

[0108]

[0109]

[0110] in, is the free space wave number, λ is the wavelength, |r-r'|=r, r' is the coordinate vector of magnetic dipole No. 2, and r is the coordinate vector of near-field measurement point 1. λ is the wavelength; r is the modulus of the position vector of the field point; ω is the angular frequency; ε0 is the dielectric constant of air; μ0 is the magnetic permeability of air.

[0111] 2) Establish a regularized objective function based on the near-field magnetic field equation;

[0112] The regularized objective function is as follows:

[0113] min||Ax-b|| 2 +λ 2 ||x|| 2 ,x∈R n (14)

[0114] Where λ is the regularization parameter. n is the set of real numbers.

[0115] 3) Solve the dipole moment matrix of the magnetic dipole using the regularized objective function;

[0116] The steps of solving the magnetic dipole coordinates using the regularized objective function include:

[0117] 3.1) Calculate the regularization parameter λ;

[0118] The steps to calculate the regularization parameter λ include:

[0119] 3.1.1) Update the regularized objective function (14) and obtain:

[0120]

[0121] Where b i A represents the new matrix after deleting the i-th element from matrix b; (i) Represents the new matrix after deleting the i-th element from matrix A; Represents the new matrix after deleting the i-th element from the matrix x;

[0122] 3.1.2) Introduce the diagonal element t ii , update the regularized objective function (15) and obtain:

[0123]

[0124] In the formula, t ii is the matrix A(A T A+λ 2 I) -1 A T The diagonal elements of ;

[0125] 3.1.3) Use the average value of the diagonal elements instead of t ii , update the regularized objective function (16) and obtain:

[0126]

[0127] The regularized objective function (18) is simplified as follows:

[0128]

[0129] 3.1.4) Using the singular value decomposition method to process the matrix A, we get:

[0130]

[0131] In the formula, σ i is the characteristic root;

[0132] 3.1.5) Calculate the regularization parameter λ, that is:

[0133]

[0134] 3.2) Substitute the regularization parameter λ into the regularization objective function (14) to obtain the dipole moment matrix x of the magnetic dipole, that is:

[0135] x=((A) T A+λ 2 I) -1 (A) T b (21)

[0136] Where I is a unit vector.

[0137] 4) Solve the low-frequency radiation field generated by the magnetic dipole array.

[0138] The steps to solve the low-frequency radiation field generated by the magnetic dipole array on the extrapolated plane include:

[0139] 4.1) Based on the definition of magnetic dipole, establish the spherical coordinate system expression of the electromagnetic field radiated by magnetic dipole;

[0140] The steps to establish the spherical coordinate system expression of the electromagnetic field radiated by the magnetic dipole include:

[0141] 4.1.1) Let the current on the magnetic dipole line element i(t) = I m cosωt=Re[Ie jωt ], the current amplitude is I m ; t is time; ω is angular frequency;

[0142] Establish the components of the electromagnetic field radiated by the magnetic dipole in the spherical coordinate system, namely:

[0143]

[0144] In the formula, is the free space wave number, λ is the wavelength, is the free space wave impedance, r is the modulus of the position vector of the field point; l is the length of the magnetic dipole line element; θ is one of the coordinates of the field point in the spherical coordinate system. In the spherical coordinate system, the coordinates of any point in space are (r, θ, ); is the electric field component; is the magnetic field component;

[0145] 4.1.2) Introducing intermediate parameters f(r) and K e , intermediate parameter K h , intermediate parameter g1(r), intermediate parameter g2(r), intermediate parameter g3(r), simplify formula (22) and get:

[0146]

[0147] In the formula, is the magnetic dipole moment of the magnetic dipole in the z direction.

[0148] Among them, the intermediate parameters f(r) and K e , intermediate parameter K h , intermediate parameters g1(r), intermediate parameters g2(r), and intermediate parameters g3(r) are as follows:

[0149]

[0150] 4.2) The spherical coordinate expression of the electromagnetic field radiated by the magnetic dipole is converted into a rectangular coordinate expression, and the dipole array vector is summed to obtain the low-frequency magnetic field generated by the equivalent dipole array.

[0151] The steps of converting the spherical coordinate system expression of the electromagnetic field radiated by the magnetic dipole into a rectangular coordinate system expression and performing the dipole array vector summation include:

[0152] 4.2.1) Transform equation (24) into a rectangular coordinate system, that is:

[0153]

[0154] 4.2.2) Using the symmetry of the components of the rectangular coordinate system, the magnetic dipole moment M is derived x and M y The corresponding relationship with the magnetic field is calculated from the magnetic dipole M x 、M y and M z The magnetic field generated by the combination of

[0155]

[0156] 4.2.3) The magnetic field generated by each magnetic dipole is vector-summed to obtain the low-frequency radiation field generated by the magnetic dipole array, that is:

[0157]

[0158] In the formula, is the x-component of the magnetic dipole moment of the ith magnetic dipole, is the y component of the magnetic dipole moment of the ith magnetic dipole, is the z component of the magnetic dipole moment of the i-th magnetic dipole;

[0159] Among them, the parameter r i,j As shown below:

[0160]

[0161] x i ,y i 、z i is the coordinate of the ith magnetic dipole, x j ,y j 、z j are the coordinates of the jth near-field measurement point.

[0162] Embodiment 2:

[0163] A method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data comprises the following steps:

[0164] 1) Aiming at the need of modeling the low-frequency radiation characteristics of a certain strong electromagnetic device, a low-frequency electromagnetic field test system is first used to obtain near-field test data on one or all of the six surfaces of the strong electromagnetic device through testing and other methods, see Figure 2 and Figure 6 shown.

[0165] 2) On the plane of the magnetic dipole array, the relationship matrix between the magnetic dipole array and the near-field test data is obtained from the relationship between a single magnetic dipole and its radiation field.

[0166] Specifically, the equivalent magnetic dipoles are placed on a two-dimensional plane, and the plane is parallel to a certain surface of the strong electromagnetic field device. The equivalent magnetic dipoles are evenly distributed in the two-dimensional plane in the form of an array. Each equivalent dipole consists of components M in the x, y and z directions. x 、M y and M z composition.

[0167] Assume that a magnetic dipole array is composed of N magnetic dipoles. Here, the magnetic dipoles in the magnetic dipole array are numbered, and different numbers correspond to different magnetic dipoles. Similarly, the points of the near-field magnetic field scan are numbered, and different numbers correspond to different magnetic field points. Among them, the coordinates of the magnetic dipole are (x n ,y n , z n ), where n = 1, 2, ..., N. The number of points scanned in the near field magnetic field is M, and the coordinates are (x m ,y m , z m ), where m = 1, 2, ..., M. The near-field magnetic field expression generated by the equivalent magnetic dipole array can be written in the form of the following matrix:.

[0168] Ax=b

[0169] in,

[0170]

[0171]

[0172] in, represents the x-component of the first magnetic dipole, H 1x Represents the x-component of the magnetic field for the first near-field test data point. represents the conversion coefficient between the magnetic dipole moment in the magnetic dipole array and the near-field test data. The superscript represents the moment component of the corresponding magnetic dipole, and the subscript represents the magnetic field component of the corresponding near-field test data point, that is, It represents the conversion coefficient between the magnetic dipole moment of the x-component of the magnetic dipole numbered 2 and the x-component of the point radiation magnetic field numbered 1.

[0173]

[0174] in, is the free space wave number, λ is the wavelength, |r-r'|=r, r' is the coordinate vector of magnetic dipole numbered 2, r is the coordinate vector of near-field measurement point 1, y2 and z2 are the y-axis and z-axis coordinates of magnetic dipole numbered 2 respectively.

[0175] Similarly, the expressions of the remaining conversion coefficients in the coefficient matrix A are as follows:

[0176]

[0177]

[0178]

[0179]

[0180]

[0181]

[0182]

[0183]

[0184] 3) By solving the linear equation system (1), the dipole moment of the equivalent dipole array to be determined can be obtained. Since the condition number of the coefficient matrix A of the linear equation system is large and it is an ill-conditioned matrix, the error between the direct solution result and the true value is large. A stable solution with small error can be obtained by using the least squares method and regularization.

[0185] To solve the ill-conditioned matrix, the Tikhonov regularization method is often used. The Tikhonov regularization method combined with the least squares method is as follows:

[0186] min||Ax-b|| 2 +λ 2 ||x|| 2 ,x∈R n

[0187] Among them, λ is the regularization parameter.

[0188] It can be seen that in formula (14), A and b are known quantities, x is the quantity to be solved, and the regularization parameter λ is an unknown quantity. Therefore, when solving the matrix equation, the regularization parameter needs to be solved first.

[0189] a) Use the generalized cross-validation method to solve the Tikhonov regularization parameter.

[0190] The two commonly used methods for solving regularization parameters are generalized cross-validation and L-curve. The idea of ​​generalized cross-validation is that from a statistical point of view, a good regularization parameter can predict the missing results. Using the generalized cross-validation method, the data is divided into two parts, one for finding an approximate solution and the other for verifying the approximate solution. For example, if we delete the i-th element b in b, i , we get vector b (i) , correspondingly delete the i-th row of matrix A, and get matrix A (i) , calculate the Tychonov regularization solution as follows:

[0191]

[0192] Then the deleted element b i You can pass Estimate. Select m near-field data to verify. A suitable regularization parameter λ can minimize the expected error of all elements.

[0193]

[0194] Right now:

[0195]

[0196] Where t ii is the matrix A(A T A+λ 2 I) -1 A T The diagonal elements of t are replaced by the average value of the diagonal elements. ii , then:

[0197]

[0198] Right now:

[0199]

[0200] The singular value decomposition method processes the matrix A to obtain:

[0201]

[0202] Therefore, the optimal regularization parameter obtained by the generalized cross-validation method of Tikhonov regularization satisfies the condition:

[0203]

[0204] b) Tikhonov regularization combined with least squares method.

[0205] After solving the regularization parameter, substitute it into equation (14) and solve the corresponding x in equation (14). The expression of x is:

[0206] x=((A) T A+λ 2 I) -1 (A) T b

[0207] Equation (22) is the stable solution of the matrix equation Ax=b.

[0208] 4) Using the relationship between the magnetic dipole and its radiation field and vector summation, the low-frequency radiation field generated by the equivalent magnetic dipole array on the extrapolated plane is solved.

[0209] 4.1) Spherical coordinate system expression of electromagnetic field radiated by magnetic dipole

[0210] The magnetic dipole is placed along the z-axis. Assume that the current on the line element changes sinusoidally with time i(t) = I m cosωt=Re[Ie jωt ], whose current amplitude is I m .

[0211] Then the components of the electromagnetic field radiated by the magnetic dipole in the spherical coordinate system can be obtained as:

[0212]

[0213] in, is the free space wave number, λ is the wavelength, is the free space wave impedance, and r is the magnitude of the position vector of the field point.

[0214] To simplify the calculation formula, make the following substitutions:

[0215]

[0216] The electromagnetic field generated by the magnetic dipole in the z direction can be expressed as:

[0217]

[0218] in, is the magnetic dipole moment of the magnetic dipole in the z direction.

[0219] 4.2) The expression of the magnetic field generated by the magnetic dipole is converted from the spherical coordinate system to the rectangular coordinate system so that the dipole array vector can be summed to solve the low-frequency magnetic field generated by the equivalent dipole array.

[0220] (a) Transforming equation (25) from the spherical coordinate system to the rectangular coordinate system, the magnetic field generated by the magnetic dipole in the z direction can be expressed as

[0221]

[0222] Using the symmetry of the components of the rectangular coordinate system, the magnetic dipole moment M can be derived x and M y The corresponding relationship with the magnetic field. Then the magnetic dipole M x 、M y and M z The magnetic field generated by the combination can be written as:

[0223]

[0224] (b) The magnetic field distribution generated by the equivalent model of the radiation source is the vector sum of the magnetic fields generated by each magnetic dipole.

[0225]

[0226] in, is the x-component of the magnetic dipole moment of the ith magnetic dipole, is the y component of the magnetic dipole moment of the ith magnetic dipole, is the z component of the magnetic dipole moment of the ith magnetic dipole

[0227]

[0228] Embodiment 3:

[0229] An experiment on the equivalent modeling method of low-frequency radiation of strong electromagnetic equipment based on test data includes the following contents:

[0230] 1) A three-phase synchronous generator with a fundamental frequency of 60Hz is established in Maxwell3D software. The generator model is established to simulate the near-field information of the generator, and the two-dimensional planar radiation source is reconstructed by the equivalent dipole method using the near-field information. The size of the near-field measurement plane is 400mm*400mm, and the plane is located above the motor and 300mm away from the motor; using the near-field information, the magnetic field distribution at other positions can be obtained by reconstructing the radiation source according to the equivalent dipole method. In this example, the magnetic field information at 300mm is used to obtain the equivalent magnetic dipole array model, and then the equivalent model is used to obtain the magnetic field distribution of the extrapolated plane 450mm away, and then compared with the simulation results of Maxwell3D. Figure 2 There are three planes in total, from top to bottom, they are the extrapolation plane, the measurement plane, and the plane where the equivalent dipole array is located. The definition of relative error is as follows:

[0231]

[0232] Among them, H calc is the magnetic field obtained by the equivalent dipole method, H ref is the simulated magnetic field.

[0233] 2) Using the same motor model, set the frequency to 150Hz, and use the near-field information to reconstruct the three-dimensional radiation source through the equivalent dipole method. The near-field measurement plane size is 180mm*180mm, and a total of 6 planes are located around the motor, 180mm away from the motor; use the near-field information to reconstruct the equivalent model of the three-dimensional radiation source according to the equivalent dipole method. The magnetic field information of the 6 measurement surfaces is used to obtain the three-dimensional equivalent magnetic dipole array model, and then the equivalent model is used to obtain the magnetic field distribution of 6 planes 300mm away from the motor and with a plane size of 300mm*300mm, and then compared with the simulation results of Maxwell3D. The definition of relative error is as follows:

[0234] .

Claims

1. A method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data, characterized in that: The following steps are involved: 1) Scan the near-field magnetic field generated by strong electromagnetic equipment and establish the near-field magnetic field equation; 2) Establish a regularized objective function based on the near-field magnetic field equation; 3) Solve the dipole moment matrix of the magnetic dipole using the regularized objective function; 4) Solve the low-frequency radiation field generated by the magnetic dipole array; The regularized objective function is as follows: min||Ax-b|| 2 +λ 2 ||x|| 2 ,x∈R n (1) Among them, λ is the regularization parameter; Matrix A, matrix b and matrix x are as follows: In the formula, represents the x-component of the magnetic dipole moment of the nth magnetic dipole, H nx represents the x-component of the magnetic field at the nth near-field test data point; Represents the conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the x-component of the magnetic field at the n'-th near-field magnetic field scanning point; n'=1,2,...,M; Among them, the conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the x-component of the radiated magnetic field at the n'th point is Conversion coefficient between the magnetic dipole moment of the y-component of the n-th magnetic dipole and the x-component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the z component of the nth magnetic dipole and the x component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the y-component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the y component of the nth magnetic dipole and the y component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the z component of the nth magnetic dipole and the y component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the x-component of the n-th magnetic dipole and the z-component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the y component of the nth magnetic dipole and the z component of the radiated magnetic field at the n'th point Conversion coefficient between the magnetic dipole moment of the z component of the nth magnetic dipole and the z component of the radiated magnetic field at the n'th point They are as follows: in, is the free space wave number, λ' is the wavelength, |r-r'| = r, r' is the coordinate vector of the magnetic dipole numbered n, and r is the coordinate vector of the near-field measurement point n'; r is the modulus of the position vector of the field point; ω is the angular frequency; ε0 is the dielectric constant of air; μ0 is the magnetic permeability of air; The steps to solve the low-frequency radiation field generated by the magnetic dipole array on the extrapolated plane include: 4.1) Based on the magnetic dipole, establish the spherical coordinate system expression of the electromagnetic field radiated by the magnetic dipole; 4.2) Convert the spherical coordinate system expression of the electromagnetic field radiated by the magnetic dipole into a rectangular coordinate system expression, and perform vector summation of the dipole array to obtain the low-frequency magnetic field generated by the equivalent dipole array; The steps to establish the spherical coordinate system expression of the electromagnetic field radiated by the magnetic dipole include: 4.1.1) Let the current on the magnetic dipole line element i(t) = I m cosωt, the current amplitude is I m ; t is time; ω is angular frequency; Establish the components of the electromagnetic field radiated by the magnetic dipole in the spherical coordinate system, namely: In the formula, is the free space wave number, λ' is the wavelength, is the free space wave impedance, r is the modulus of the position vector of the field point; l is the length of the magnetic dipole line element; θ is one of the coordinates of the field point in the spherical coordinate system. In the spherical coordinate system, the coordinates of any point in space are is the electric field component; is the magnetic field component; 4.1.2) Introducing intermediate parameters f(r) and K e , intermediate parameter K h , intermediate parameter g1(r), intermediate parameter g2(r), intermediate parameter g3(r), simplify formula (14) and get: In the formula, is the magnetic dipole moment of the magnetic dipole in the z direction; Among them, the intermediate parameters f(r) and K e , intermediate parameter K h , intermediate parameters g1(r), intermediate parameters g2(r), and intermediate parameters g3(r) are as follows: The steps of converting the spherical coordinate system expression of the electromagnetic field radiated by the magnetic dipole into a rectangular coordinate system expression and performing the dipole array vector summation include: 4.2.1) Transform equation (16) into a rectangular coordinate system, that is: 4.2.2) Using the symmetry of the components of the rectangular coordinate system, the magnetic dipole moment M is derived x and M y The corresponding relationship with the magnetic field is calculated from the magnetic dipole M x 、M y and M z The magnetic field generated by the combination of 4.2.3) The magnetic field generated by each magnetic dipole is vector-summed to obtain the low-frequency radiation field generated by the magnetic dipole array, that is: In the formula, is the x-component of the magnetic dipole moment of the ith magnetic dipole, is the y component of the magnetic dipole moment of the ith magnetic dipole, is the z component of the magnetic dipole moment of the i-th magnetic dipole; Among them, the parameter r i,j As shown below: x i ,y i 、z i is the coordinate of the ith magnetic dipole, x j ,y j 、z j are the coordinates of the jth near-field measurement point.

2. The method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data according to claim 1 is characterized in that: The magnetic dipole array includes N magnetic dipoles, and the magnetic dipoles are placed along the z-axis; wherein the coordinate of the nth magnetic dipole is (x n ,y n , z n ); n=1,2,...,N.

3. The method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data according to claim 1 is characterized in that: The number of scanning points for scanning the near-field magnetic field generated by the strong electromagnetic device is M, where the coordinates of the mth near-field magnetic field scanning point are (x m ,y m , z m ), m=1,2,...,M.

4. The method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data according to claim 1 is characterized in that: The near-field magnetic field equation is shown below: Ax=b (21).

5. The method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data according to claim 1 is characterized in that: The steps of solving the magnetic dipole coordinates using the regularized objective function include: 1) Calculate the regularization parameter λ; 2) Substitute the regularization parameter λ into the regularization objective function (1) to obtain the dipole moment matrix x of the magnetic dipole, that is: x= ( (A) T A+λ 2 I ) - 1 (A) T b(22) Where I is a unit vector.

6. The method for equivalent modeling of low-frequency radiation of strong electromagnetic equipment based on test data according to claim 5 is characterized in that: The steps to calculate the regularization parameter λ include: 1) Update the regularized objective function (1) and obtain: Where b i A represents the new matrix after deleting the i-th element from matrix b; (i) Represents the new matrix after deleting the i-th element from matrix A; Represents the new matrix after deleting the i-th element from the matrix x; 2) Introduce the diagonal element t ii , update the regularized objective function (23) and obtain: Where, t ii is the matrix A(A T A+λ 2 I) -1 A T The diagonal elements of ; 3) Use the average value of the diagonal elements instead of t ii , update the regularized objective function (24) and obtain: The regularized objective function (25) is simplified as follows: 4) Use the singular value decomposition method to process the matrix A and obtain: In the formula, σ i is the characteristic root; 5) Calculate the regularization parameter λ, that is: