Dipole source construction method based on searching minimum impedance path and computer device

Abstract

The invention relates to the field of dipole source construction, in particular to a dipole source construction method based on finding a minimum impedance path and a computer device. The method comprises the following steps: determining the composition of a dipole model according to PCB noise source radiation on the basis of a multistage expansion theory, and calculating a dipole magnetic field according to a mirror image principle and a superposition theorem; based on the relation between the distance between an observation point and a radiation source and the magnetic field intensity and a near-field scanning result, initial value assignment calculation is carried out on the uniformly-distributed dipole array; the impedance of the dipole array is calculated through the negative correlation between impedance and current and the positive correlation between dipole moment and current, and a dipole minimum impedance path is found based on an average distance norm and an improved minimum path algorithm; and amplitude optimization is carried out on the dipoles forming the minimum impedance path by using a differential evolution algorithm, and the construction accuracy of the dipole equivalent source is evaluated. The method is suitable for constructing the simplified equivalent radiation source model of the PCB.

Description

Technical Field

[0001] The present invention relates to the field of dipole source construction, and in particular to a dipole source construction method and a computer device based on finding a minimum impedance path. Background Art

[0002] Electromagnetic interference has become an increasingly important issue in modern electronic system design, as the number and types of electronic devices continue to increase, while the size of the devices themselves and the distance between them continue to decrease. Electromagnetic radiation obtained through measurement or full-wave simulation includes interference from all emission sources, but it is impossible to determine the radiation amount of a specific radiation source.

[0003] At present, the method of constructing an equivalent dipole source based on near-field scanning technology is applied to predict electromagnetic interference effects, and the source reconstruction method based on phaseless near-field scanning has attracted the attention of the industry. The global optimization algorithm can search for the optimal solution in the entire range, and because of its strong versatility, it is more widely used in source modeling problems. When the existing source reconstruction iterative method searches for the dipole position and amplitude at the same time, the multi-dimensional solution space makes the iterative process need to run for a long time, which is inefficient and the results may not be accurate. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a dipole source construction method and a computer device based on finding the minimum impedance path, which reduces the calculation dimension of the dipole model and improves the efficiency and accuracy.

[0005] The present invention adopts the following technical solutions to achieve the above-mentioned purpose. In a first aspect, the present invention provides a dipole source construction method based on finding a minimum impedance path, comprising:

[0006] Based on the multi-level expansion theory, the composition of the dipole model is determined according to the radiation of the PCB (Printed Circuit Board) noise source, and the dipole phase-free magnetic field is calculated according to the mirror principle and superposition theorem;

[0007] Based on the relationship between the distance between the observation point and the radiation source and the magnetic field strength, the calculation of the initial value of the uniformly distributed dipole array is performed as the near-field scanning result;

[0008] The impedance of the dipole array is calculated by the negative correlation between impedance and current, and the positive correlation between dipole moment and current magnitude, and the dipole minimum impedance path is found based on the average distance norm and the improved minimum path algorithm;

[0009] The differential evolution algorithm is used to optimize the amplitude of the dipoles that constitute the minimum impedance path, and the accuracy of the dipole equivalent source construction is evaluated.

[0010] Furthermore, the calculation of the dipole magnetic field according to the mirror principle and the superposition theorem specifically includes:

[0011] A single dipole magnetic field calculation model is established. Based on the mirror principle, the amplitude at height h is The system composed of the dipole and the ground plane is equivalent to a pair of dipoles with a height of ±h. The magnetic field of each pair of dipoles at the observation point is expressed as Where T is the transformation matrix, which can be written as

[0012] The magnetic field calculation model of the dipole array is established, and the magnetic field generated by the equivalent dipole source at the observation point is calculated as follows: Assume that the magnetic field generated by the nth dipole at the sampling point is H x,n and H y,n According to the superposition theorem, the magnetic field at the observation point is equal to the sum of the effects of all N dipoles on the observation point, that is,

[0013] Furthermore, the calculation of initializing the uniformly distributed dipole array based on the near-field scanning result specifically includes:

[0014] The PCB trace layer plane is divided, and the position (x, y), number M×N and spacing delta of the dipole are set. Each dipole is located directly below the observation point.

[0015] Use a near-field probe to perform near-field scanning on the PCB circuit board at a set height to obtain the magnetic field data H at that height. x and H y ;

[0016] According to the near-field scanning magnetic field data H x and H y , set the proportionality factor k, and calculate the x and y component size M of each dipole x 、M y .

[0017] Furthermore, the impedance of the dipole array is calculated by the negative correlation between the impedance and the current and the positive correlation between the dipole moment and the current. Specifically, the impedance includes:

[0018] Establish the relationship between the dipole size and the impedance between adjacent dipoles. Assuming that there is a current channel between all adjacent dipoles, the current size between the dipoles distributed in the x and y directions is respectively related to its amplitude M. x 、M y The impedance between adjacent dipoles is positively correlated with the inverse of the dipole amplitude;

[0019] Establish the equivalent concept of relative impedance of dipole array, define the distance norm between adjacent dipoles, and the distance norm of adjacent dipoles along the x direction is the M of the two y The mean of the reciprocal, the distance norm of adjacent dipoles along the y direction is M x The mean of the reciprocal, defining the total distance norm as the sum of all distance norms on the connection path between two given dipoles, and defining the average distance norm as the ratio of the total distance norm of the dipole connection path to the number of distance norms on the path;

[0020] The relationship between the distance norm and the relative impedance of the dipole array path is established. According to the definition of the distance norm, the smaller the distance norm, the greater the current between the dipoles. The average distance norm of the entire path is used to evaluate the relative impedance of the path.

[0021] Furthermore, based on the average distance norm and the improved minimum path algorithm, the dipole minimum impedance path is found specifically including:

[0022] A minimum impedance path calculation method is established, the positions of the input and output ports are set, the dipole closest to the excitation port is set as the starting point, and the dipole closest to the output port is set as the end point, the path with the minimum total distance norm is found, and this path is recorded as the initial minimum path with the average distance norm, the average distance norm of the path is calculated, and recorded as the initial minimum value of the average distance norm, and then the path with the minimum total distance norm is found again, and the average distance norm of this path is calculated. When the new minimum value is less than the recorded minimum value, the recorded minimum value is updated, and the cycle is repeated until convergence, and the dipole minimum impedance path can be obtained.

[0023] Furthermore, the use of the differential evolution algorithm to optimize the amplitude of the dipole constituting the minimum impedance path specifically includes:

[0024] Establish a differential evolution algorithm to prepare the optimization population, define the population number as NP, initialize all populations, set each population as a dipole array, and each dipole can be represented by a six-dimensional vector:

[0025] in, represents the d-th dipole of the t-th population, Re(·) and Im(·) represent the real and imaginary parts of the complex number, respectively. and The coordinates representing the dipole position, the population initial path is the optimization result of the minimum impedance path, the population initial amplitude is the dipole initial value, first set the first population as the dipole distribution after path optimization, then set the range of values ​​of each dipole size in several populations, and randomly select values ​​within this range;

[0026] Determine the optimization process of the differential evolution algorithm. After initializing the dipole model, set the objective function as the correlation coefficient and the convergence target as 1 according to the accuracy evaluation method of the dipole equivalent source construction. Perform mutation, crossover, and selection operations on NP populations in each iteration to make the objective function meet the requirements. After the iteration, all dipoles are proportionally enlarged or reduced to find the solution with the minimum mean square error as the final result.

[0027] Furthermore, the evaluation of the accuracy of the construction of the dipole equivalent source specifically includes:

[0028] Establish the comparison relationship between the reconstructed phase-free magnetic field and the measured phase-free magnetic field, M is the number of observation points, and is the actual magnetic field measured, A represents the mean, SD represents the variance, and MSE is defined as total is the total mean square error, and are the mean square errors of the magnetic fields in the x and y directions, respectively, where m = 1, 2, …, M; ρ total Defined as the total correlation coefficient, and are the correlation coefficients of the magnetic fields in the x and y directions respectively, then the comparison formula between the reconstructed phase-free magnetic field and the measured phase-free magnetic field is MSE total =MSE x +MSE y ,

[0029] In a second aspect, the present invention provides a computer device, including a memory, wherein the memory stores program instructions, and when the program instructions are executed, the above-mentioned dipole source construction method based on finding the minimum impedance path is executed.

[0030] The beneficial effects of the present invention are:

[0031] The present invention can determine the PCB emission structure and realize the rapid construction of a simplified equivalent radiation source model thereof; not only can the construction of the dipole equivalent model of the PCB emission structure be completed on the basis of single-layer phase-free near-field scanning, but the constructed equivalent radiation source model can also reflect a more complete emission structure path; by decoupling the calculation of the dipole amplitude and position, the calculation dimension of the dipole model is reduced, so that when the method uses more dipoles to construct the equivalent model, the algorithm takes less time to achieve the same accuracy than the traditional method of using a global optimization algorithm, and the equivalent source accuracy is higher within the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A flow chart of a dipole source construction method based on finding a minimum impedance path provided by an embodiment of the present invention;

[0033] Figure 2 A flow chart of finding a path with a smaller average distance norm provided by an embodiment of the present invention;

[0034] Figure 3 A flow chart of a truncation operation provided by an embodiment of the present invention;

[0035] Figure 4 A flow chart of a dipole source construction method provided by an embodiment of the present invention;

[0036] Figure 5 A schematic diagram of a W-shaped microstrip line and a scanning plane provided by an embodiment of the present invention;

[0037] Figure 6 A distribution diagram of a dipole array after initialization of a W-type microstrip line provided by an embodiment of the present invention;

[0038] Figure 7 A distribution diagram of a dipole array after the W-shaped microstrip line path is optimized according to an embodiment of the present invention;

[0039] Figure 8 The final distribution diagram of the dipole array provided by the embodiment of the present invention;

[0040] Fig. 9 A schematic diagram of distance norm calculation of a dipole array provided in an embodiment of the present invention;

[0041] Fig.10 A schematic diagram of distance norm calculation of another dipole array provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0043] The present invention provides a dipole source construction method based on finding the minimum impedance path, such as Figure 1 As shown, including:

[0044] Based on the multi-level expansion theory, the composition of the dipole model is determined according to the radiation of the PCB noise source, and the dipole phase-free magnetic field is calculated according to the mirror principle and superposition theorem;

[0045] Based on the relationship between the distance between the observation point and the radiation source and the magnetic field strength, the calculation of the initial value of the uniformly distributed dipole array is performed as the near-field scanning result;

[0046] The impedance of the dipole array is calculated by the negative correlation between impedance and current, and the positive correlation between dipole moment and current magnitude, and the dipole minimum impedance path is found based on the average distance norm and the improved minimum path algorithm;

[0047] The differential evolution algorithm is used to optimize the amplitude of the dipoles that constitute the minimum impedance path, and the accuracy of the dipole equivalent source construction is evaluated.

[0048] Specifically, the calculation of the dipole phaseless magnetic field according to the mirror principle and superposition theorem includes:

[0049] A single dipole magnetic field calculation model is established. Based on the mirror principle, the amplitude at height h is The system composed of the dipole and the ground plane is equivalent to a pair of dipoles with a height of ±h. The magnetic field of each pair of dipoles at the observation point is expressed as Where T is the transformation matrix, which can be written as

[0050] The magnetic field calculation model of the dipole array is established, and the magnetic field generated by the equivalent dipole source at the observation point is calculated as follows: Assume that the magnetic field generated by the nth dipole at the sampling point is H x,n and H y,n According to the superposition theorem, the magnetic field at the observation point is equal to the sum of the effects of all N dipoles on the observation point, that is,

[0051] Specifically, the calculation of initializing the uniformly distributed dipole array based on the near-field scanning result specifically includes:

[0052] The PCB trace layer plane is divided, and the position (x, y), number M×N and spacing delta of the dipole are set. Each dipole is located directly below the observation point.

[0053] Use a near-field probe to perform near-field scanning on the PCB circuit board at a set height to obtain the magnetic field data H at that height. x and H y ;

[0054] According to the near-field scanning magnetic field data H x and H y , set the proportionality factor k, and calculate the x and y component size M of each dipole x 、M y .

[0055] Specifically, the impedance of the dipole array is calculated based on the negative correlation between impedance and current and the positive correlation between dipole moment and current magnitude, including:

[0056] Establish the relationship between the dipole size and the impedance between adjacent dipoles. Assuming that there is a current channel between all adjacent dipoles, the current size between the dipoles distributed in the x and y directions is respectively related to its amplitude M. x 、M yThe impedance between adjacent dipoles is positively correlated with the inverse of the dipole amplitude;

[0057] Establish the equivalent concept of relative impedance of dipole array, define the distance norm as the distance norm of adjacent dipoles along the x direction as M y The mean of the reciprocal, the distance norm of adjacent dipoles along the y direction is M x The mean of the reciprocal, defining the total distance norm as the sum of all distance norms on the connection path between two given dipoles, and defining the average distance norm as the ratio of the total distance norm of the dipole connection path to the number of distance norms on the path;

[0058] The relationship between the distance norm and the relative impedance of the dipole array path is established. According to the definition of the distance norm, the smaller the distance norm, the greater the current between the dipoles. The average distance norm of the entire path is used to evaluate the relative impedance of the path.

[0059] Specifically, based on the average distance norm and the improved minimum path algorithm, finding the dipole minimum impedance path specifically includes:

[0060] Establish a minimum impedance path calculation method, set the positions of the input and output ports, set the dipole closest to the excitation port as the starting point, and the dipole closest to the output port as the end point, find the path with the minimum total distance norm, and record this path as the initial minimum path with the average distance norm, calculate the average distance norm of the path, and record it as the initial minimum value of the average distance norm, then find the path with the minimum total distance norm again, calculate the average distance norm of this path, and when the new minimum value is less than the recorded minimum value, update the recorded minimum value, and repeat the cycle until convergence, and you can get the dipole minimum impedance path. The specific calculation flow of this process is as follows: Figure 2 , Figure 3 As shown, the truncation operation refers to that a certain section of the original path is not a path.

[0061] Specifically, using the differential evolution algorithm to optimize the amplitude of the dipole constituting the minimum impedance path specifically includes:

[0062] Establish a differential evolution algorithm to prepare the optimization population, define the population number as NP, initialize all populations, set each population as a dipole array, and each dipole can be represented by a six-dimensional vector:

[0063] in, represents the d-th dipole of the t-th population, Re(·) and Im(·) represent the real and imaginary parts of the complex number, respectively. and The coordinates representing the dipole position, the population initial path is the optimization result of the minimum impedance path, the population initial amplitude is the dipole initial value, first set the first population as the dipole distribution after path optimization, then set the range of values ​​of each dipole size in several populations, and randomly select values ​​within this range;

[0064] Determine the optimization process of the differential evolution algorithm. After initializing the dipole model, set the objective function as the correlation coefficient and the convergence target as 1 according to the accuracy evaluation method of the dipole equivalent source construction. Perform mutation, crossover, and selection operations on NP populations in each iteration to make the objective function meet the requirements. After the iteration, all dipoles are proportionally enlarged or reduced to find the solution with the minimum mean square error as the final result.

[0065] Specifically, the evaluation of the accuracy of the construction of the dipole equivalent source includes:

[0066] Establish the comparison relationship between the reconstructed phase-free magnetic field and the measured phase-free magnetic field, M is the number of observation points, and is the actual magnetic field measured, A represents the mean, SD represents the variance, and MSE is defined as total is the total mean square error, and are the mean square errors of the magnetic fields in the x and y directions, respectively, where m = 1, 2, …, M; ρ total Defined as the total correlation coefficient, and are the correlation coefficients of the magnetic fields in the x and y directions respectively, then the comparison formula between the reconstructed phase-free magnetic field and the measured phase-free magnetic field is MSE total =MSE x +MSE y ,

[0067] The present invention will be further described below in conjunction with specific embodiments.

[0068] like Figure 4 As shown, an embodiment of the present invention relates to a dipole source construction method based on finding a minimum impedance path, which is used to construct a simplified equivalent radiation source model of a PCB, and specifically includes the following steps:

[0069] Step 1: Build the PCB working circuit. The PCB model is as follows: Figure 5 As shown, set the PCB operating frequency, adjust the input signal amplitude of the signal source, and ensure that the PCB radiation emission intensity is at an appropriate level;

[0070] Step 2: Determine the near-field scanning frequency according to the PCB operating frequency, define the appropriate near-field scanning area, scanning step length and scanning height, use the near-field probe to perform near-field scanning on the PCB, and obtain the magnetic field distribution H in a certain area above the PCB. x , H y ;

[0071] Step 3: Divide the PCB trace layer plane according to the near-field scanning step length, set the position (x, y), number M×N and spacing delta of the dipoles, and each dipole is located directly below the observation point;

[0072] Step 4: Scan magnetic field data H through near field x , H y , arbitrarily set the proportional coefficient k, and then M = kH ref Calculate the x and y components of each dipole M x 、M y , the dipole array formed is as follows Figure 6 As shown;

[0073] Step 5: According to the position of the input and output ports, the dipole array in step 4 is searched for a path with a smaller average distance norm, and only the dipoles on the path are retained to obtain the optimized dipole distribution diagram as shown in Figure 7 As shown;

[0074] Step 6: Determine the number of populations N optimized by the differential evolution algorithm. First, set the first population as the dipole distribution after path optimization, then set the range of values ​​for each dipole size in the next N-1 populations, and randomly select values ​​within this range. Since we know the position distribution of the dipoles, set the last two dimensions of each dipole to fixed values ​​and do not participate in iterative optimization.

[0075] Step 7: After the model is initialized, the objective function is set to the correlation coefficient according to the correlation coefficient formula in the dipole equivalent source accuracy evaluation method, and the convergence target is 1. After the iteration, all dipoles are proportionally enlarged or reduced to find the solution with the minimum mean square error as the final result. Figure 8 shown.

[0076] Specifically, the distance norm described in step 5 is calculated based on the initial amplitude of the dipole array. For a simple 4-dipole array, Fig. 9 , Fig.10 As shown, the four red dots are four dipoles numbered 1-4.

[0077] Among them, the distance norm between 1 and 2 is

[0078] 1 / (2M x1 )+1 / (2M x2 )

[0079] If Fig. 9 The yellow path is defined as the connecting path between 1 and 2, then the total distance norm and average distance norm of the path are both

[0080] 1 / (2M x1 )+1 / (2M x2 )

[0081] Similarly, if Fig.10 The yellow path is defined as the connecting path between 1 and 2, then the total distance norm of the path is

[0082] 1 / (2M x1 )+1 / (2M x3 )+1 / (2M y3 )+1 / (2M y4 )+1 / (2M x4 )+1 / (2M x2 )

[0083] The average distance norm is:

[0084] [1 / (2M x1 )+1 / (2M x3 )+1 / (2M y3 )+1 / (2M y4 )+1 / (2M x4 )+1 / (2M x2 )] / 3.

[0085] According to the definition of the distance norm, the smaller the distance norm, the larger the current between the dipoles. The distance norm has a property similar to impedance, so the present invention uses the average distance norm of the entire path to evaluate the relative impedance of the path. The premise of the present invention is that the positions of the input and output ports are known, and the dipole closest to the excitation port is set as the starting point, and the dipole closest to the output port is set as the end point. Due to the non-directionality of the distance norm definition, the interchange of the starting point and the end point does not affect the result. Since the current always flows along a path with low impedance, this patent believes that among all the paths connecting the starting point and the end point, the path with a smaller average distance norm is more likely to be the actual path of the current. Therefore, the problem can be transformed into: finding a path with the smallest possible average distance norm between two points.

[0086] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A dipole source construction method based on finding the minimum impedance path, characterized in that: include: Based on the multi-level expansion theory, the composition of the dipole model is determined according to the radiation of the PCB noise source, and the dipole phase-free magnetic field is calculated according to the mirror principle and superposition theorem; Based on the relationship between the distance between the observation point and the radiation source and the magnetic field strength, the calculation of the initial value of the uniformly distributed dipole array is performed as the near-field scanning result; The impedance of the dipole array is calculated by the negative correlation between impedance and current, and the positive correlation between dipole moment and current magnitude, and the dipole minimum impedance path is found based on the average distance norm and the improved minimum path algorithm; The differential evolution algorithm is used to optimize the amplitude of the dipoles that constitute the minimum impedance path, and the accuracy of the dipole equivalent source construction is evaluated.

2. The dipole source construction method based on finding the minimum impedance path according to claim 1, characterized in that: The calculation of dipole phaseless magnetic field based on the mirror principle and superposition theorem specifically includes: A single dipole magnetic field calculation model is established. Based on the mirror principle, the amplitude at height h is The system composed of the dipole and the ground plane is equivalent to a pair of dipoles with a height of ±h. The magnetic field of each pair of dipoles at the observation point is expressed as Where T is the transformation matrix, The magnetic field calculation model of the dipole array is established, and the magnetic field generated by the equivalent dipole source at the observation point is calculated as follows: Assume that the magnetic field generated by the nth dipole at the sampling point is H x,n and H y,n According to the superposition theorem, the magnetic field at the observation point is equal to the sum of the effects of all N dipoles on the observation point, that is, 3. The dipole source construction method based on finding the minimum impedance path according to claim 1, characterized in that: The calculation of initialization for a uniformly distributed dipole array based on the near-field scanning results specifically includes: The PCB trace layer plane is divided, and the position (x, y), number M×N, and spacing delta of the dipoles are set. Each dipole is located directly below the observation point. Use a near-field probe to perform near-field scanning on the PCB circuit board at a set height to obtain the magnetic field data H at that height. x and H y ; According to the near-field scanning magnetic field data H x and H y , set the proportionality factor k, and calculate the x and y component size M of each dipole x 、M y .

4. The dipole source construction method based on finding the minimum impedance path according to claim 1, characterized in that: The impedance of the dipole array is calculated by the negative correlation between impedance and current and the positive correlation between dipole moment and current size, including: Establish the relationship between the dipole size and the impedance between adjacent dipoles. Assuming that there is a current channel between all adjacent dipoles, the current size between the dipoles distributed in the x and y directions can be obtained as x 、M y The impedance between adjacent dipoles is positively correlated with the inverse of the dipole amplitude; Establish the equivalent concept of relative impedance of dipole array, define the distance norm between adjacent dipoles, and the distance norm of adjacent dipoles along the x direction is the M of the two y The mean of the reciprocal, the distance norm of adjacent dipoles along the y direction is M x The mean of the reciprocal, defining the total distance norm as the sum of all distance norms on the connection path between two given dipoles, and defining the average distance norm as the ratio of the total distance norm of the dipole connection path to the number of distance norms on the path; The relationship between the distance norm and the relative impedance of the dipole array path is established. According to the definition of the distance norm, the smaller the distance norm, the greater the current between the dipoles. The average distance norm of the entire path is used to evaluate the relative impedance of the path.

5. The dipole source construction method based on finding the minimum impedance path according to claim 4, characterized in that: Based on the average distance norm and the improved minimum path algorithm, the dipole minimum impedance path is found by: A minimum impedance path calculation method is established, the positions of the input and output ports are set, the dipole closest to the excitation port is set as the starting point, and the dipole closest to the output port is set as the end point, the path with the minimum total distance norm is found, and this path is recorded as the initial minimum path with the average distance norm, the average distance norm of the path is calculated, and recorded as the initial minimum value of the average distance norm, and then the path with the minimum total distance norm is found again, and the average distance norm of this path is calculated. When the new minimum value is less than the recorded minimum value, the recorded minimum value is updated, and the cycle is repeated until convergence, and the dipole minimum impedance path can be obtained.

6. The dipole source construction method based on finding the minimum impedance path according to claim 1, characterized in that: The use of differential evolution algorithm to optimize the amplitude of the dipoles constituting the minimum impedance path specifically includes: Establish a preliminary optimization population for the differential evolution algorithm, define the population number as NP, initialize all populations, set each population as a dipole array, and each dipole is represented by a six-dimensional vector: in, represents the d-th dipole of the t-th population, Re(·) and Im(·) represent the real and imaginary parts of the complex number, respectively. and The coordinates representing the dipole position, the population initial path is the optimization result of the minimum impedance path, the population initial amplitude is the dipole initial value, first set the first population as the dipole distribution after path optimization, then set the range of values ​​of each dipole size in several populations, and randomly select values ​​within this range; Determine the optimization process of the differential evolution algorithm. After initializing the dipole model, set the objective function as the correlation coefficient and the convergence target as 1 according to the accuracy evaluation method of the dipole equivalent source construction. Perform mutation, crossover, and selection operations on NP populations in each iteration to make the objective function meet the requirements. After the iteration, all dipoles are proportionally enlarged or reduced to find the solution with the minimum mean square error as the final result.

7. The dipole source construction method based on finding the minimum impedance path according to claim 1, characterized in that: Evaluating the accuracy of the dipole equivalent source construction specifically includes: Establish the comparison relationship between the reconstructed phase-free magnetic field and the measured phase-free magnetic field, M is the number of observation points, and is the actual magnetic field measured, A represents the mean, SD represents the variance, and MSE is defined as total is the total mean square error, and are the mean square errors of the magnetic fields in the x and y directions, respectively, where m = 1, 2, …, M; ρ total Defined as the total correlation coefficient, and are the correlation coefficients of the magnetic fields in the x and y directions respectively, then the comparison formula between the reconstructed phase-free magnetic field and the measured phase-free magnetic field is MSE total =MSE x +MSE y , 8. A computer device, comprising a memory, wherein the memory stores program instructions, characterized in that: When the program instructions are executed, the dipole source construction method based on finding the minimum impedance path as described in any one of claims 1 to 7 is executed.

Images