Radiation source equivalent modeling method, apparatus, medium, and device

Abstract

This application provides a method, apparatus, medium, and device for equivalent modeling of a radiation source. The method includes: acquiring magnetic field amplitude data at multiple distances for each scanning surface of a target device at different orientations; independently constructing a local equivalent source model corresponding to each scanning surface based on the data; determining field data containing amplitude and phase at the boundary line between adjacent scanning surfaces based on the corresponding local equivalent source models; and merging multiple local equivalent source models based on the field data to obtain an overall equivalent radiation source model of the target device. In this way, by introducing spatial attenuation information of the radiation source's magnetic field through multi-distance magnetic field amplitude data, the insufficient depth direction constraint caused by phase loss is compensated. Merging local equivalent source models based on field data at the boundary of scanning surfaces reduces non-physical phase jumps caused by independent modeling, making the obtained equivalent radiation source model spatially continuous and physically self-consistent, effectively improving the stability and accuracy of the equivalent radiation source model.

Description

Technical Field

[0001] This application relates to the field of radiation source model construction technology, specifically to a radiation source equivalent modeling method, apparatus, medium, and equipment. Background Technology

[0002] With the increasing integration of electronic devices in intelligent connected vehicles, it is necessary to establish high-precision component-level equivalent radiation source models to effectively identify and resolve radio frequency interference issues in the early stages of research and development. These models should then be imported into a vehicle-level electromagnetic simulation environment for interference prediction and analysis. Such models must accurately reflect the spatial distribution characteristics of the radiation sources and meet the requirements for model consistency and stability in engineering applications.

[0003] In related technologies, in order to reduce the complexity of testing, some engineering scenarios have tried to adopt a scheme that only uses a spectrum analyzer to obtain near-field amplitude data. However, such schemes usually only scan a single height plane, and the information obtained is limited in dimension. There is a problem of missing depth information when inverting and solving, which leads to insufficient stability of the solution. Finally, the equivalent radiation source model obtained has poor stability and accuracy in the spatial dimension. Summary of the Invention

[0004] Based on the defects and shortcomings of the existing technology, this application proposes a radiation source equivalent modeling method, apparatus, medium and equipment.

[0005] According to a first aspect of this application, a radiation source equivalent modeling method is provided, the method comprising: For each scanning surface of the target device in different orientations, acquire magnetic field amplitude data at multiple different distances; Multiple local equivalent source models are constructed based on the magnetic field amplitude data, and each local equivalent source model corresponds to one of the scanning surfaces. Based on the boundary line between adjacent scanning planes and the corresponding local equivalent source model, the field data at the boundary line is determined, and the field data includes amplitude data and phase data. Based on the field data, the multiple local equivalent source models are merged to obtain the equivalent radiation source model of the target device.

[0006] According to a second aspect of the embodiments of this application, a radiation source equivalent modeling apparatus is provided, the apparatus comprising: The acquisition module is used to acquire magnetic field amplitude data at multiple different distances for each scan surface of the target device in different orientations; The construction module is used to construct multiple local equivalent source models based on the magnetic field amplitude data, and each local equivalent source model corresponds to one of the scanning surfaces; The determination module is used to determine the field data at the boundary line based on the boundary line of adjacent scanning planes and the corresponding local equivalent source model, wherein the field data includes amplitude data and phase data; The merging module is used to merge the multiple local equivalent source models based on the field data to obtain the equivalent radiation source model of the target device.

[0007] According to a third aspect of the embodiments of this application, an electronic device is provided, including a memory and a processor; The memory is connected to the processor and is used to store programs; The processor is used to implement the radiation source equivalent modeling method as described in the first aspect by running the program in the memory.

[0008] According to a fourth aspect of the embodiments of this application, a storage medium is provided, on which a computer program is stored, and when the computer program is run by a processor, it implements the radiation source equivalent modeling method as described in the first aspect.

[0009] According to a fifth aspect of the embodiments of this application, a computer program product is provided, wherein the computer program instructions, when executed by a processor, cause the processor to implement the radiation source equivalent modeling method as described in the first aspect.

[0010] In the aforementioned radiation source equivalent modeling method, apparatus, medium, and equipment, magnetic field amplitude data at multiple distances can be acquired for each scanning surface of the target device in different orientations. Based on this data, a local equivalent source model corresponding to each scanning surface is independently constructed. Based on the local model, field data containing amplitude and phase at the boundary line between adjacent scanning surfaces is determined. Based on this field data, multiple local equivalent source models are merged into a single equivalent radiation source model. In this way, spatial attenuation information can be introduced through multi-distance amplitude sampling, compensating for insufficient depth direction constraints caused by phase loss and enhancing the stability of the solution. On this basis, by using the boundary field data to merge the local equivalent source models, non-physical phase jumps generated by independent modeling can be corrected, making the obtained equivalent source radiation model spatially continuous and physically self-consistent. It can also be directly used for vehicle-level electromagnetic compatibility simulation. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0012] Figure 1This is a flowchart illustrating a radiation source equivalent modeling method provided in an embodiment of this application.

[0013] Figure 2 This is a schematic diagram illustrating an equivalent modeling process for a radiation source, as provided in an embodiment of this application.

[0014] Figure 3 This is a schematic diagram illustrating the effect of radiation source reconstruction in an embodiment of this application.

[0015] Figure 4 This is a schematic diagram of the structure of a radiation source equivalent modeling device provided in an embodiment of this application.

[0016] Figure 5 This is a schematic diagram of the structure of an electronic device proposed in an embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] In the field of electromagnetic compatibility testing and simulation analysis, a commonly used technical solution for constructing equivalent radiation source models of electronic devices and importing them into a vehicle-level electromagnetic simulation environment for interference prediction is the phase coherence measurement method relying on a vector network analyzer. Specifically, this method controls a near-field probe to perform planar scanning above the device under test, simultaneously acquiring the amplitude and phase information of the near-field magnetic field at each scanning point. Based on the acquired complex near-field data, a source reconstruction algorithm is used to invert and obtain the equivalent magnetic flux or magnetic dipole distribution. The basic working principle of this method is that the phase information provides complete electromagnetic field vector constraints for the inversion algorithm, enabling the intensity and spatial location of the equivalent source to be uniquely determined, thereby establishing an equivalent model with high accuracy and close consistency with the measured data.

[0019] However, when this method is applied to the engineering testing environment of actual automotive electronic devices, especially in sensitive frequency bands such as 433MHz (e.g., remote key) or 1.575GHz (e.g., GPS / BeiDou), and under limited site and hardware conditions, its performance is less than ideal. The reason for this unsatisfactory performance is that obtaining a high-quality model requires stable and consistent phase measurement data. However, cable movement, minor probe jitter, and environmental disturbances in the engineering field can cause drastic phase fluctuations, especially at high frequencies, significantly reducing the reliability of phase data and resulting in poor data repeatability and low modeling success rate.

[0020] Through in-depth analysis, the inventors discovered that the root cause of the aforementioned problems lies in the rigid dependence of related technical solutions on phase data. While stable phase measurements serve as the basis for maintaining model phase consistency, the signal link itself is inherently fragile and prone to errors. To circumvent the high risk of phase measurement by acquiring single-layer magnetic field amplitude data, only the field distribution along the tangential direction of the scanning plane can be constrained. Insufficient constraints on the equivalent source location and intensity in the depth direction perpendicular to the scanning plane lead to insufficient information dimensions. In this case, the inversion results are prone to numerous possible non-physical interpretations, exhibiting significant randomness and instability. Furthermore, when different sides of the tested component are modeled independently in blocks, the lack of a unified phase reference for each local model results in non-physical field value jumps or discontinuities at their spatial boundaries. The resulting overall model may contain numerous non-physical jumps, rendering it unusable for full-wave simulation.

[0021] To overcome the aforementioned contradictions, this invention proposes a different technical approach. The core concept of this invention lies in: by introducing magnetic field amplitude data from multiple planes at different distances, the additional constraint formed by the spatial attenuation of near-field amplitude along the normal direction can be utilized to eliminate dependence on measured phase information, enabling phase-free amplitude inversion to obtain equivalent source models for each scanning plane; further, based on the local equivalent source models, the field data at the boundary lines of adjacent scanning planes is determined, and the field data is combined to merge the various local equivalent source models to obtain the equivalent radiation source model of the target device. This solves the problems of insufficient depth in amplitude inversion solutions and phase mismatch during multi-plane model stitching without relying on any measured phase data. In other words, this application provides an equivalent modeling method based on multi-layer near-field amplitude scanning and adaptive alignment of boundary continuity, enabling the construction of physically self-consistent equivalent radiation source models in a low-cost, non-ideal testing environment. This achieves the technical effect of avoiding phase measurement risks while ensuring the model's spatial continuity and the simulability of the field distribution.

[0022] Based on the above concept, this specification provides a radiation source equivalent modeling method, which will be described exemplarily below with reference to the accompanying drawings.

[0023] The radiation source equivalent modeling method provided in the embodiments of this specification is mainly applied in the radiation source equivalent modeling system. Specifically, it is used to construct the overall equivalent source model of the target radiation source based on the electromagnetic field amplitude information of multiple distance surfaces without relying on phase measurement data.

[0024] The system comprises two parts: a near-field scanning acquisition subsystem and a computer processing unit. The near-field scanning acquisition subsystem is responsible for acquiring electromagnetic field amplitude measurement data of the target radiation source on multiple spatial planes. The computer processing unit receives this amplitude measurement data and executes an equivalent source modeling program to generate a global equivalent source model.

[0025] In the near-field scanning acquisition subsystem, a movable near-field magnetic field probe is mounted on a multi-axis mechanical scanning device. The probe operates in a frequency band covering the radiation frequencies of interest to the device under test. The probe's output is connected to the input of a spectrum analyzer via an RF cable. The spectrum analyzer is configured to measure the amplitude of the signal picked up by the probe at specified frequencies and bandwidths. Both the control interface of the mechanical scanning device and the data interface of the spectrum analyzer are connected to a control computer. This control computer sends commands to drive the probe to move point-by-point within a predefined scanning area, scanning step size, and scanning height, while simultaneously acquiring the amplitude values ​​read by the spectrum analyzer at each sampling point in real time, and storing the amplitude data in association with the corresponding spatial coordinates.

[0026] Specifically, the near-field scanning acquisition subsystem performs a scan at different distances in all directions around the target radiation source, obtaining electromagnetic field amplitude measurement data for each sampling point on each surface at different distances. Subsequently, corresponding to the scans at different distances, the mechanical scanning device positions the probe on the corresponding distance surface and performs a scan operation to obtain electromagnetic field amplitude measurement data for each sampling point on each surface at those different distances. The acquired data is the amplitude information of the tangential component of the magnetic field.

[0027] The computer processing device is the core computing unit of this system; it can be a general-purpose computer, workstation, or server. This device contains at least one processor and one memory. The memory loads an equivalent source modeling program that can run on the processor. When running, this program can access amplitude measurement data from different distance planes recorded by the near-field scanning acquisition subsystem.

[0028] Internally, the equivalent source modeling program is logically organized into several collaborative processing components. The equivalent source definition component defines each group of equivalent radiation sources and their spatial configuration in the digital space. The objective function construction component forms a joint optimization objective that measures the deviation between theoretical amplitude and measured data. The parameter optimization component iteratively adjusts the equivalent radiation source parameters to ensure the objective function meets preset conditions. The local equivalent source model acquisition component processes the measured data from different spatial regions of the target radiation source, generating multiple local equivalent source models, each with its own independent phase reference. The boundary field extraction component performs forward prediction calculations on the electromagnetic field information of adjacent local equivalent source models in the boundary region. The discontinuity quantification component establishes a cost function that measures the difference in field values ​​in the boundary region. The phase alignment component solves for and applies a global phase rotation based on the cost function. The model merging component combines the phase-referenced local models into a single global equivalent source model, thus obtaining the target equivalent radiation source model.

[0029] In the data processing flow, amplitude data from the spectrum analyzer first enters the memory via the data acquisition interface and is preprocessed. The preprocessed data is then passed to the equivalent source definition component and the objective function construction component. The parameter optimization component repeatedly calls the forward electromagnetic field calculation process using the value of the objective function, continuously updating the equivalent source parameters and determining the local equivalent source model for each independent region. The computer processing device receives this amplitude measurement data and executes the equivalent source modeling program to generate the overall equivalent source model, which can be directly used for subsequent electromagnetic compatibility simulation analysis.

[0030] Please see Figure 1 In one exemplary embodiment, a radiation source equivalent modeling method is provided, applicable to any electronic device. For example... Figure 1 As shown, the radiation source equivalent modeling method includes steps S101-S104: S101: For each scanning surface of the target device in different orientations, acquire magnetic field amplitude data at multiple different distances.

[0031] The target equipment is either an in-vehicle electronic device that contains a radiation source or an in-vehicle electronic device that can serve as a radiation source.

[0032] The near-field scanning acquisition subsystem uses a movable near-field magnetic field probe to scan the target device at different distances on its scanning surface (or scanning area) in different directions, thereby obtaining magnetic field amplitude data at different distances on the scanning surface in each direction.

[0033] The computer sends commands to drive the probe to move within a predefined scanning plane, and the spectrum analyzer collects the amplitude values ​​read at each sampling point within the scanning plane in real time. The predefined scanning plane refers to the scanning planes at different locations of the target device.

[0034] For example, different orientations of the target device refer to the orientation of the top surface and / or at least one side surface of the target device.

[0035] Specifically, for each orientation, the mechanical scanning device sequentially positions the probe at multiple different distances on the scanning surface, such as a first distance scanning surface and a second distance scanning surface, and performs scanning to obtain magnetic field amplitude measurement data for each sampling point at different distances on each scanning surface. These distance scanning surfaces are located in the near-field region of the target device and have a certain distance difference along the direction perpendicular to the scanning surface, i.e., the normal direction of the scanning surface, in order to capture the attenuation characteristics of the electromagnetic field during its propagation in space.

[0036] For example, taking the top surface as the scanning surface, the first distance scanning surface is a plane that is a first distance from the top surface and parallel to it, and the second distance scanning surface is a plane that is a second distance from the top surface and parallel to it. Similarly, taking a side surface as the scanning surface, the first distance scanning surface is a plane that is a first distance from the side surface and parallel to it, and the second distance scanning surface is a plane that is a second distance from the side surface and parallel to it. In other words, the first and second distance scanning surfaces are planes that are parallel to the scanning surface in the normal direction of their respective scanning surfaces and are separated from the scanning surface by a certain distance.

[0037] For example, the first distance scanning surface is the near-field detail layer h1, which is relatively close to the scanning surface, and the second distance scanning surface is the spatial attenuation layer h2, which is relatively far from the scanning surface.

[0038] More specifically, based on the physical dimensions and structural characteristics of the target device, the scanning parameters of the spectrum analyzer and the near-field magnetic field probe are configured so that the probe can be controlled by a mechanical scanning device to perform block scanning at different positions on the surface of the target device.

[0039] Specifically, magnetic field amplitude data refers to the amplitude data of the tangential component of the magnetic field. The tangential component of the magnetic field refers to the tangential direction component of the magnetic field vector on the current scanning plane, that is, the magnetic field component parallel to the scanning plane, rather than the normal component perpendicular to the scanning plane.

[0040] For example, during probe scanning, for a horizontal top surface scan, the Hx and Hy components are acquired (i.e., Figure 3 (The x-axis and y-axis are shown). For vertical side scans, two tangential components parallel to the vertical side are acquired.

[0041] By acquiring amplitude data of the tangential component of the magnetic field without including phase data, the phase measurement fluctuation problem caused by factors such as probe jitter, cable bending or environmental disturbance in high-frequency testing can be effectively avoided, greatly improving the engineering robustness of data acquisition and the success rate of modeling.

[0042] Specifically, after the scan is completed, the original amplitude data is preprocessed (such as coordinate alignment) to form a standard dataset for subsequent modeling.

[0043] S102: Multiple local equivalent source models were constructed based on magnetic field amplitude data.

[0044] Among them, the local equivalent source model corresponds one-to-one with the scanning surface.

[0045] For each scanning surface, magnetic field amplitude data at different distances corresponding to the scanning surface are selected from the acquired magnetic field amplitude data, and a local equivalent source model corresponding to the scanning surface is constructed based on the magnetic field amplitude data corresponding to the scanning surface.

[0046] Specifically, based on the magnetic field amplitude data at different distances corresponding to the scanning surface, an inversion constraint containing the spatial attenuation gradient is constructed, and the equivalent source parameters, i.e. the parameters in the local equivalent source model corresponding to the scanning surface, are determined based on the constraint, thereby obtaining the local equivalent source model of the scanning surface.

[0047] More specifically, an equivalent source parameter determination model is used to determine the local equivalent source model corresponding to each scanning surface. This model can be trained using a neural network and is used to determine the equivalent source parameters corresponding to the scanning surface based on the magnetic field amplitude data. For example, for any scanning surface, the magnetic field amplitude data at different distances corresponding to the scanning surface are input into the equivalent source parameter determination model to obtain the parameters in the local equivalent source model of that scanning surface output by the model. Then, the local equivalent source model corresponding to that scanning surface is determined based on these parameters.

[0048] S103: Determine the field data at the boundary line based on the boundary line of adjacent scanning planes and the corresponding local equivalent source model.

[0049] The field data includes amplitude data and phase data, which are calculated from the local equivalent source model.

[0050] Specifically, for the boundary line between two adjacent scanning planes, the field data at the boundary line is obtained by calculating using the corresponding local equivalent source model.

[0051] That is, for multiple sampling points on the boundary line of adjacent scanning surfaces, the corresponding complex magnetic field prediction values ​​are obtained by forward calculation using the corresponding local equivalent source model. These complex magnetic field prediction values ​​include amplitude and phase values, or amplitude and phase prediction values.

[0052] More specifically, after the local equivalent source models of all spatial regions are generated, the boundary field extraction component within the computer processing device begins its work. This component first automatically identifies and determines the boundary lines between adjacent scanning surfaces and the regions where these boundaries are located, based on the known geometrical relationships of each scanning surface in space. It then defines a set of discrete sampling points on each scanning surface within its corresponding boundary region. Subsequently, for each sampling point on the boundary region, the boundary field extraction component calls the corresponding local equivalent source models from two adjacent local equivalent source models to perform forward electromagnetic field calculations, obtaining the amplitude and phase prediction values ​​for each sampling point on each boundary region.

[0053] In other words, using the equivalent source parameters of local equivalent source model A, the amplitude and phase prediction values ​​are calculated at sampling point P1 in the boundary region corresponding to local equivalent source model A. Similarly, using the equivalent source parameters of local equivalent source model B, the amplitude and phase prediction values ​​are calculated at sampling point P2 in the boundary region corresponding to local equivalent source model B. Sampling points P1 and P2 are adjacent on either side of the boundary line, or coincide on the boundary line.

[0054] The aforementioned field data refers to electromagnetic field data at sampling points in the boundary region, predicted by forward calculation from local equivalent source models. It is typically in complex form and includes both amplitude and phase information. This calculation method eliminates reliance on measured data, directly revealing field value differences at the same physical location caused by inconsistent phase references among local models. This provides direct and accurate input data for subsequent quantification of discontinuities.

[0055] S104: Based on the field data, multiple local equivalent source models are merged to obtain the equivalent radiation source model of the target device.

[0056] Specifically, the phase alignment component performs overall phase correction on multiple local equivalent source models based on field data. In this way, these local equivalent source models can accurately reproduce the measured amplitude in their respective regions, and they have a unified phase reference with each other. The field distribution in the boundary region is also continuous.

[0057] Then, the corrected local equivalent source models are merged using the model merging component to obtain the equivalent radiation source model of the target device.

[0058] More specifically, the model merging component combines these local models in a unified coordinate system to form a complete, overall equivalent source model covering the entire space of the target radiation source. The final model obtained by the model merging component contains both amplitude and phase information and can be output in a standard engineering data format.

[0059] Subsequently, the overall equivalent source model can be directly imported into mainstream electromagnetic simulation software such as CST or HFSS as an excitation source, replacing complex actual devices under test in vehicle-level electromagnetic compatibility simulation analysis, and realizing rapid and high-precision assessment of the radiation interference risk of vehicle electronic equipment.

[0060] In this embodiment, spatial attenuation information can be introduced by multi-distance amplitude sampling to compensate for insufficient depth direction constraints caused by phase loss and enhance the stability of the solution. On this basis, by using the boundary field data to merge the local equivalent source model, the non-physical phase jump generated by independent modeling can be corrected, so that the obtained equivalent source radiation model is spatially continuous and physically self-consistent, and can also be directly used for vehicle-level electromagnetic compatibility simulation.

[0061] To further optimize the ability of the above embodiments to characterize the complex current distribution of the radiation source, in some embodiments, the equivalent radiation source is specifically defined as a magnetic dipole array characterized by a complex magnetic dipole moment. That is, the local equivalent source model is composed of a magnetic dipole array, wherein the parameters of each magnetic dipole in the magnetic dipole array include its spatial coordinates and a complex magnetic dipole moment, and the direction of the complex magnetic dipole moment is characterized by three orthogonal components.

[0062] Specifically, a magnetic dipole array is a collection of several basic magnetic dipole units. The radiation behavior of each magnetic dipole is described by a complex magnetic dipole moment vector. This complex magnetic dipole moment not only defines the radiation intensity of the magnetic dipole at that location, but its complex form can also contain phase information.

[0063] Since the near-field data collected is the amplitude of the tangential component of the magnetic field, the goal of the inversion is to make the theoretical magnetic field amplitude generated by the equivalent magnetic dipole array approximate the measured magnetic field amplitude. The magnetic dipole is the inversion unit that is directly related to the measured physical quantity and has a relatively clear physical mechanism. Compared with the use of electric dipoles, the magnetic dipole is more effective here.

[0064] In this embodiment, a local equivalent source model is constructed using a magnetic dipole array, which can flexibly and accurately approximate various complex radiation current distributions inside the device under test with limited degrees of freedom. Using complex magnetic dipole moments, even without measured phase data input, the boundary phase prediction values ​​are solved for each local equivalent source model. Although these solved phase prediction values ​​have inconsistent references when independently inverting in different regions, they can provide basic data and calibration targets for subsequent adaptive rotation and alignment of the phase reference, ultimately enabling the merged overall model to possess a physically self-consistent and complete phase distribution. Thus, by adopting the above scheme, this embodiment can construct a vector equivalent radiation source model that can both reproduce measured amplitude and is spatially phase-consistent, thereby synergistically enhancing the overall technical effect of this invention in constructing a complete radiation model using only amplitude data.

[0065] In some embodiments, multiple local equivalent source models are constructed based on the magnetic field amplitude data, including: for each scanning surface, the following operations are performed: constructing an objective function, optimizing the objective function by adjusting the candidate equivalent source parameters until a preset condition is met, and determining the parameters of the local equivalent source model. The objective function measures the joint deviation between the theoretical magnetic field amplitude generated based on a set of candidate equivalent source parameters corresponding to the scanning surface and the magnetic field amplitude data corresponding to the scanning surface.

[0066] Specifically, we first define a set of candidate equivalent source parameters. Here, equivalent (radiation) source parameters refer to the set of basic radiating units that can approximately replace real radiation sources in electromagnetic field numerical calculations; candidates are those to be selected / adjusted.

[0067] More specifically, the definition operation is performed in digital space by an equivalent source definition component within a computer processing device. Based on the spatial geometry and potential distribution range of the radiation source corresponding to the target device, the equivalent source definition component creates and initializes a mathematical model of a set of candidate equivalent radiation source parameters in computer memory. These candidate equivalent radiation source parameters, such as their spatial location and complex quantities characterizing radiation intensity, constitute the set of variables to be optimized in subsequent steps.

[0068] Specifically, the objective function is constructed using an objective function construction component within the computer processing device. This objective function construction component receives measured amplitude data from multiple distance scanning planes corresponding to any scanning plane of the near-field scanning acquisition subsystem, and invokes the forward electromagnetic field calculation process for candidate equivalent radiation source parameters to obtain their theoretical magnetic field amplitude at the same spatial location.

[0069] The objective function can be understood as a quantitative scoring standard for evaluating the merits of a set of candidate equivalent radiation source parameters. The core of this scoring standard lies in the degree to which the theoretical field generated by the same set of equivalent radiation source parameters simultaneously approximates the corresponding measured amplitude distribution at various distance planes, or in other words, the error relative to the measured magnetic field amplitude. The smaller the error, the higher the score; the better the set of candidate equivalent radiation source parameters can reproduce the measured amplitude distribution, and the better the inversion result.

[0070] Accordingly, the preset condition could be, for example, maximizing the objective function value.

[0071] Because there is a defined geometric spacing between each range scanning surface, the corresponding magnetic field amplitude data packets contain information on amplitude attenuation and spatial diffusion as the electromagnetic field propagates along the normal direction of the scanning surface. Combining the fitting deviations from multiple range scanning surfaces to form the objective function is equivalent to introducing a spatial attenuation gradient constraint for the inversion of the equivalent radiation source parameters. In this way, non-physical solutions that achieve good amplitude fitting on a single range scanning surface but do not conform to the actual spatial propagation attenuation laws, but have significant errors with other range scanning surfaces, will be exposed and suppressed in the joint objective function. This effectively converges the inversion solution space and alleviates the common problems of solution instability and lack of depth information in single-layer amplitude data inversion.

[0072] Specifically, the weighted sum of the relative root mean square errors of the magnetic field amplitude data of multiple distance scanning surfaces (such as the first distance surface and the second distance surface mentioned above) corresponding to the scanning surface is constructed as the objective function.

[0073] More specifically, the relative root mean square error between the theoretical magnetic field amplitude generated by the equivalent radiation source at each distance scanning surface and the corresponding measurement data is calculated using the objective function construction component. These error values ​​are then weighted and summed according to preset weighting factors to form the final joint fitness function. The weights can be flexibly adjusted based on the signal-to-noise ratio or importance of the data at different distance scanning surfaces.

[0074] For example, the relative root mean square error is calculated by dividing the sum of the squares of the differences between the theoretical and measured values ​​by the sum of the squares of the measured values, and then taking the square root. This indicator can objectively reflect the relative magnitude of the fitting deviation and is not affected by the absolute strength of the signal. Finally, the relative root mean square errors of each distance scanning surface are linearly weighted and summed according to preset weights to obtain the final objective function value. For example, the objective function value J is calculated according to J=w1×NRMSE1+w2×NRMSE2, where NRMSE1 and w1 are the magnetic field amplitude data of the first distance scanning surface and their corresponding preset weights, and NRMSE2 and w2 are the magnetic field amplitude data of the second distance scanning surface and their corresponding preset weights.

[0075] For example, if the first distance scan plane is closer to the target device surface, it has richer near-field detail information and can usually be assigned a higher weight w1 to force the algorithm to prioritize the accurate reproduction of hotspot locations and amplitudes in the near-field detail layer. The error terms of other distance scan planes serve as a strong regularization constraint, ensuring that the spatial attenuation trend of the model's radiation field as it propagates outwards conforms to physical laws. Some candidate solutions that cannot simultaneously satisfy the amplitude distributions of multiple layers, especially those that cause the radiation source to be incorrectly located at non-physical depth locations, will be eliminated by the optimization algorithm due to the excessive total cost of the weighted sum, thus preserving the optimal solution.

[0076] Specifically, the objective function is optimized by adjusting the parameters of the equivalent radiation source through a parameter optimization component within a computer processing device until a preset condition is met, thereby determining the optimization operation of the local equivalent source model corresponding to the scanning surface.

[0077] The parameter optimization component repeatedly calls the objective function construction component to evaluate the quality of the current equivalent radiation source parameters, and iteratively adjusts the parameters based on the evaluation results, driving the objective function value to gradually increase.

[0078] Since the objective function is constructed based on multi-level amplitude errors, its shape is highly nonlinear and has many local minima. The specific means to achieve the above parameter optimization function may include using a global optimization algorithm.

[0079] More specifically, the parameters of the equivalent radiation source can be adjusted using a differential evolution algorithm. That is, when multiple local equivalent source models are constructed based on magnetic field amplitude data, the parameters of the local equivalent source models are optimized using a differential evolution algorithm.

[0080] Among them, the differential evolution algorithm is a population-based heuristic global search strategy that continuously evolves the candidate solution population in the complex solution space through operations such as mutation, crossover, and selection. This algorithm has the inherent characteristic of escaping local optima and can effectively overcome the shortcomings of traditional gradient-based optimization algorithms, such as sensitivity to initial values ​​and easy getting trapped in local optima. This allows the parameter optimization component to stably converge to an optimal solution that can simultaneously satisfy the amplitude distribution constraints of multiple distance scan surfaces, even when the initial conditions of the equivalent radiation source are unknown.

[0081] In this context, "unknown initial conditions" refers to the fact that, in engineering practice, it is not necessary to obtain the precise physical location, phase distribution, and dipole moment parameters of the actual radiation source beforehand as the starting point for iteration. The optimization process itself is still subject to the physical constraints of the acquired scanning surface position, amplitude data at multiple distances, the equivalent source search region, and the forward electromagnetic field model. The algorithm automatically completes the search within the solution space jointly defined by these constraints.

[0082] Furthermore, the aforementioned differential evolution algorithm is not limited to the standard form; various variants can also be employed, such as adaptive differential evolution algorithms (jDE, SaDE), which can automatically adjust the mutation factor and crossover probability according to the optimization process, further improving the optimization efficiency at different iteration stages. By adopting this global optimization algorithm, the robustness and solution accuracy of amplitude inversion modeling under complex engineering conditions can be significantly improved.

[0083] The optimization process requires numerous forward electromagnetic field calculations to evaluate the theoretical field distribution of each candidate equivalent radiation source on different distance surfaces. Specifically, forward electromagnetic field calculations from the equivalent radiation source to the scanning point are performed based on the analytical form of the free-space Green's function. For the equivalent radiation source characterized by a magnetic dipole array, the magnetic field generated at the field point can be directly calculated from the analytical expression of the magnetic dipole moment vector and the free-space Green's function. This analytical calculation method avoids the time-consuming numerical grid partitioning and iterative solution process for the entire solution space, and can quickly provide accurate theoretical field values ​​in each individual population evaluation, thus ensuring the feasibility and efficiency of the global optimization algorithm in engineering applications.

[0084] By performing the above process on each scanning surface, multiple local equivalent source models are obtained. Each local equivalent source model corresponds one-to-one with a scanning surface and has its own independent phase reference. When modeling the target radiation source as a whole, due to the geometric structure or test accessibility limitations of the device under test (i.e., the target device), it is usually necessary to perform near-field scanning and independent modeling for different spatial regions (e.g., the top surface and multiple side surfaces). Each local equivalent source model consists of a set of equivalent radiation sources (such as a magnetic dipole array), and its parameters are independently optimized and determined in their respective inversion processes. Therefore, the relative phase between the equivalent source elements within each local model is self-consistent and can independently reproduce the measured amplitude distribution of its corresponding region.

[0085] However, the phase references between different local models are independent of each other, and no unified phase reference relationship is established between them. If they are simply combined directly, non-physical field value jumps or discontinuities may occur in the spatial boundary region.

[0086] For example, based on the spatial geometric coordinates of each scanned surface, the common boundary of adjacent surfaces is automatically identified. For instance, an edge of the top surface (+z surface) coincides with or is adjacent to the top edge of a side surface (e.g., +y surface) in space. Along this common boundary line or within a narrow overlapping region, the boundary field extraction component defines a series of discrete sampling points. For each sampling point P, the boundary field extraction component first calls the local equivalent magnetic dipole model corresponding to the top surface, applies the analytical form of the free-space Green's function, and forward calculates the complex magnetic field vector H_pred_top generated by the model at point P. Then, it independently calls the local equivalent magnetic dipole model corresponding to the side surface, similarly calculating the complex magnetic field vector H_pred_side generated at point P. These two complex magnetic field predictions are in one-to-one correspondence, forming paired data.

[0087] These paired complex magnetic field predictions directly reveal discontinuities at the boundary. Because point P is the same or very close physical location, if the two local equivalent source models are under a unified phase reference, the calculated H_pred_top and H_pred_side are approximately equal. Any significant differences, especially systematic deviations in the phase angle, stem directly from the different internal phase references formed by the two models during independent inversion. Calculating complex magnetic field predictions in this way eliminates the limitations of measured data, providing a direct and clean input signal for the discontinuity quantification component, and accurately quantifying abrupt changes in non-physical field values ​​caused by inconsistent phase references.

[0088] To eliminate phase differences, in some embodiments, multiple local equivalent source models are merged based on field data, including: constructing a cost function, adjusting the overall phase reference of at least one local equivalent source model with the goal of minimizing the cost function, so that the phases of multiple local equivalent source models are aligned; merging the aligned local equivalent source models to form an equivalent radiation source model of the target device.

[0089] The cost function is used to quantify the degree of discontinuity in field data from adjacent local equivalent source models at the boundary line.

[0090] Specifically, a cost function is constructed using a discontinuity quantization component. This component receives the complex magnetic field prediction value output by the boundary field extraction component and constructs a mathematical scalar function, i.e., the cost function, based on this.

[0091] According to the physical laws of electromagnetic fields, in source-free free space, the electromagnetic field should be continuous and smoothly changing in space. Therefore, although adjacent local equivalent source models are modeled independently, they should describe the electromagnetic field of the same physical space in the boundary region. Thus, if there are significant differences, especially phase jumps, between the two complex magnetic field predictions from the two local models at the sampling points in the boundary region, it indicates a non-physical discontinuity caused by the inconsistency of the phase reference of the local models.

[0092] Specifically, a cost function is constructed based on the differences in complex magnetic field predictions across all sampling points in the boundary region. That is, the cost function is constructed based on the difference in complex magnetic field predictions between adjacent local equivalent source models at sampling points on the boundary line. Then, when adjusting the overall phase reference of at least one local equivalent source model, one local equivalent source model is selected as a reference, and a complex rotation factor is calculated and applied to the remaining local equivalent source models.

[0093] For example, the degree of discontinuity can be quantified by calculating the square of the magnitude of the difference between the predicted values ​​of the pair of complex magnetic fields, or by calculating some measure of their phase difference, using a discontinuity quantification component, and then summing or averaging the results over all sampling points in the boundary region.

[0094] Generally, the larger the value of the cost function, the more significant the discontinuity in the field distribution in the boundary region. This cost function can transform the fuzzy physical continuity requirement into a clear and calculable mathematical optimization objective, providing a definite optimization direction for subsequent automatic phase alignment.

[0095] For example, for the boundary region between the top and side surfaces, all K (K is a positive integer) sampling points defined on this boundary region are traversed. For the pairwise complex magnetic field prediction value of the nth (n is a positive integer, n≤K) sampling point, the square of the modulus of the difference between the pairwise complex magnetic field prediction values ​​is calculated using the discontinuity quantization component, and used as a measure of local discontinuity at the nth sampling point. Then, the local discontinuity measures of the K sampling points are accumulated to form a single scalar cost value, i.e., the cost function value. This cost value can accurately quantify the total amount of field value discontinuities, such as gaps or misalignments, caused by different phase references of the local models across the entire boundary region. The larger the cost function value, the less physically compliant the field distribution at the splice is, and the more severe the break or abrupt change.

[0096] Specifically, the phase correction amount is solved by using a phase alignment component with the goal of minimizing the cost function. Based on this phase correction amount, the global phase reference of the corresponding local equivalent source model is adjusted by applying the correction amount, so as to eliminate or minimize the non-physical discontinuities quantified by the cost function.

[0097] Specifically, a local equivalent source model is first selected as the phase reference. Then, the phase correction values ​​of the remaining local equivalent source models are used as variables to be optimized, and optimization operations are performed on these variables. For any set of phase correction values, these values ​​are applied as complex rotation factors to their corresponding local equivalent source models. Next, the discontinuity quantization component is used to evaluate the new cost function values ​​of adjacent local models in the boundary region after applying phase rotation. Finally, the set of phase correction values ​​that minimizes this cost function value is selected as the target phase correction values.

[0098] For example, a top-scan model or the model with the best data quality can be selected as the phase reference and its overall phase offset can be set to 0.

[0099] Through the above, the phase alignment component transforms the phase alignment problem among multiple models into a low-dimensional optimization problem. The component first selects one of several locally equivalent source models as the phase reference, typically the top-scan model, fixing its overall phase offset (φ_ref) to 0. For each of the remaining subordinate local models, its phase alignment is abstracted as a variable to be optimized: the phase rotation angle (φ_i). This (φ_i) corresponds to a complex rotation factor e^(jφ_i). The phase alignment component is primarily used to find an optimal set of φ_i within a relatively small search space (e.g., (φ_i ∈ [-π,π])).

[0100] During the search process, for each set of candidate φ_i, the phase alignment component applies these phase rotation factors to the complex magnetic dipole moments of all equivalent magnetic dipoles of the corresponding dependent model, thereby rotating its overall phase reference. Subsequently, the discontinuity quantization component (260) is triggered to recalculate the cost function values ​​of the boundary region. Through a simple optimization algorithm (e.g., parameter sweep or simplex method), the phase alignment component can quickly find a set of φ_i that minimizes the cost function.

[0101] The advantage of this overall rotation correction strategy lies in its ability to strictly limit the correction operation to rigid rotation of local models, ensuring that the internal relative phase relationships and amplitude distributions of each local model obtained in the independent inversion stage remain completely unchanged, thus perfectly preserving the fitting accuracy of multi-distance amplitude joint inversion. By adjusting only one or a few phase rotation angles, the global phase incompatibility problem caused by block modeling can be solved with extremely high efficiency, ultimately achieving a physically seamless stitching of multiple local equivalent source models across the entire spatial boundary, resulting in an equivalent radiation source model of the target device with good continuity.

[0102] It should be noted that the phase correction strategy described above is mathematically equivalent to finding an optimal common phase bias for each local model. While maintaining the fitting accuracy of the amplitude distribution and the self-consistency of the internal relative phase of each local model, it can effectively eliminate the pseudo-boundary effect introduced by block modeling. As a result, the complex field values ​​of each local equivalent source model at the boundary line tend to be consistent, the field data (including amplitude and phase) at the boundary line can be determined, and the physical continuity of the entire spatial field can be restored.

[0103] In order to provide the above-mentioned global optimization algorithm with a feasible region that conforms to physical laws and can effectively accelerate the search, in some embodiments, the equivalent magnetic dipole is configured in a three-dimensional spatial region located below the target device.

[0104] An equivalent source definition component defines a well-defined, bounded three-dimensional search volume within the computer, stipulating that the spatial position of the equivalent magnetic dipole can only change within this volume during the search process. The location and size of this search volume are determined based on the actual physical dimensions of the target device, the geometry of the scanning surface, and prior knowledge of the potential radiation source, and it is typically located inside the scanning surface (i.e., below the device under test).

[0105] For example, if the target device is a flat vehicle electronic control unit and the scanning is performed directly above it, the three-dimensional search volume can be defined as a cuboid region that is close to the lower surface of the target device, with a length and width similar to the scanning area and a depth of about a few centimeters.

[0106] This constraint has a dual advantage: it reflects the basic judgment of the physical location of the radiation source, that is, the equivalent source should be located within the space of the actual circuit traces and devices, eliminating a large number of non-physical solutions in the space, which greatly improves the physical reliability of the inversion results. It can also provide a finite and clearly defined search space for the differential evolution algorithm, narrowing the search range and avoiding the algorithm from blindly searching in an infinite space, thereby significantly improving the convergence speed of the algorithm and the success rate of finding the physical optimal solution.

[0107] It should also be noted that although the search scope is limited to "the three-dimensional spatial region located below the device under test," the relative orientation of this region can be adjusted according to the specific orientation of the scanning surface. For example, for a side scan, this region is located inside the target device. Essentially, it restricts the equivalent source from appearing in the pseudo-physical space outside the scanning surface.

[0108] For example, the equivalent modeling process for radiation sources can be as follows: Figure 2 As shown.

[0109] First, acquire and process the magnetic field amplitude data at multiple heights (i.e., the distances mentioned above) of each scanning surface, and then perform the independent reconstruction stage of each (scanning) surface, that is, perform the operation of generating a local equivalent source model for each scanning surface.

[0110] When generating a local equivalent source model, first select a single scan surface data, construct the corresponding three-dimensional search volume and dipole array, construct an objective function based on multi-height amplitude data, execute the differential evolution algorithm to optimize the parameters, and generate the local equivalent source model of the surface.

[0111] Subsequently, the reconstructed predicted phase data (i.e. the predicted values ​​mentioned above) at the intersection of adjacent scanning planes are extracted, the discontinuity difference in phase distribution in the intersection region is calculated, and phase reference adaptive rotation and alignment are performed to perform phase alignment and model merging of each local equivalent source model, and finally output the continuous overall equivalent radiation source model.

[0112] For example, based on the solution provided in the above embodiments, the highly sensitive 1.575GHz frequency (corresponding to the GPS / BeiDou satellite navigation band) in vehicle electronic devices is selected as the experimental frequency for verification. The experiment uses a near-field scanning system to acquire near-field magnetic field amplitude data of the device under test (DUT) surface and compares it with the predicted data of the equivalent source model generated based on the algorithm of this embodiment.

[0113] At the 1.575 GHz frequency point, the measured and reconstructed results are compared at different height levels (including the near-field layer (low layer, abbreviated as low in the diagram) and the high layer (high layer, with a spacing of, for example, 20 mm between the near-field layer and the high layer) and at different polarization directions (Hx, Hy, Hz) (i.e., the x, y, z axes in the diagram). Figure 3 As shown. Each set of comparison charts contains three sub-charts, the specific meanings of which are as follows: Left figure (measured amplitude): Thermal map of magnetic field amplitude distribution actually collected by a spectrum analyzer and a near-field probe (unit: dB A / m). This figure reflects the true distribution of radiant energy.

[0114] The middle figure (reconstructed amplitude): the predicted magnetic field amplitude distribution calculated by the equivalent magnetic dipole array generated using the method described in this invention (multi-height joint inversion + differential evolution algorithm).

[0115] Right figure (Predicted Phase - Calibrated): The recovered magnetic field phase distribution (unit: Degree, deg) after boundary consistency correction. This figure shows the spatial phase information reconstructed by physical algorithms under conditions without phase measurement.

[0116] Comparing the left (measured) and middle (reconstructed) figures, it is evident that the equivalent radiation source model obtained based on the above embodiments maintains a very high degree of consistency with the measured results in terms of spatial energy distribution. Quantitative error analysis based on the original measurement point coordinates shows that the maximum error between the reconstructed amplitude and the measured amplitude is controlled within a low range. For example, in the main radiation components, the amplitude difference is typically less than 3 dB, demonstrating that the multi-distance amplitude joint sampling strategy can effectively solve the problem of solution instability in single-layer inversion. The right figure (predicted phase) visually illustrates the effect of "boundary continuity correction" in the above embodiments. In the boundary regions of different scanning planes, no obvious and drastic non-physical phase jumps or breaks occur, indicating that the physical self-consistency of the radiation source in three-dimensional space can be successfully restored through this embodiment.

[0117] Exemplary device like Figure 4 As shown in the figure, this application embodiment also provides a radiation source equivalent modeling device, including an acquisition module 401, a construction module 402, a determination module 403, and a merging module 404.

[0118] in, The acquisition module 401 is used to acquire magnetic field amplitude data at multiple different distances for each scanning surface of the target device in different orientations; The construction module 402 is used to construct multiple local equivalent source models based on the magnetic field amplitude data, and the local equivalent source models correspond one-to-one with the scanning surface; The determination module 403 is used to determine the field data at the boundary line based on the boundary line of adjacent scanning planes and the corresponding local equivalent source model. The field data includes amplitude data and phase data. The merging module 404 is used to merge the multiple local equivalent source models based on the field data to obtain the equivalent radiation source model of the target device.

[0119] The radiation source equivalent modeling apparatus provided in this embodiment belongs to the same application concept as the radiation source equivalent modeling method provided in the above embodiments of this application. It can execute the radiation source equivalent modeling method provided in any of the above embodiments of this application and has the corresponding functional modules and beneficial effects of executing the method. Technical details not described in detail in this embodiment can be found in the specific processing content of the radiation source equivalent modeling method provided in the above embodiments of this application, and will not be repeated here.

[0120] Exemplary electronic devices Another embodiment of this application also provides an electronic device, see [link to relevant documentation] Figure 5 As shown, the electronic device includes a memory 500 and a processor 510.

[0121] The memory 500 is connected to the processor 510 and is used to store programs; The processor 510 is used to implement the radiation source equivalent modeling method disclosed in any of the above embodiments by running the program stored in the memory 500.

[0122] Specifically, the electronic device may also include: a bus, a communication interface 520, an input device 530, and an output device 540.

[0123] The processor 510, memory 500, communication interface 520, input device 530, and output device 540 are interconnected via a bus. Among them: A bus can include a pathway for transmitting information between various components of a computer system.

[0124] The processor 510 can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present application. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0125] The processor 510 may include a main processor, as well as a baseband chip, modem, etc.

[0126] The memory 500 stores a program for executing the technical solution of this application, and may also store an operating system and other critical business functions. Specifically, the program may include program code, which includes computer operation instructions. More specifically, the memory 500 may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, etc.

[0127] Input device 530 may include a device for receiving data and information input by a user, such as a keyboard, mouse, camera, scanner, light pen, voice input device, touch screen, pedometer, or gravity sensor.

[0128] Output device 540 may include devices that allow information to be output to a user, such as a display screen, printer, speaker, etc.

[0129] The communication interface 520 may include a device that uses any transceiver to communicate with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0130] The processor 510 executes the program stored in the memory 500 and calls other devices, which can be used to implement each step of any radiation source equivalent modeling method provided in the above embodiments of this application.

[0131] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0132] This application also proposes a chip, which includes a processor and a data interface. The processor reads and runs a program stored in a memory through the data interface to execute the radiation source equivalent modeling method described in any of the above embodiments. For the specific processing procedure and its beneficial effects, please refer to the embodiments of the radiation source equivalent modeling method described above.

[0133] In addition to the methods and apparatus described above, embodiments of this application provide a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the radiation source equivalent modeling method according to various embodiments of this application as described in the "Exemplary Methods" section of this specification.

[0134] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0135] Furthermore, embodiments of this application also propose a storage medium storing a computer program that is executed by a processor in the steps of the radiation source equivalent modeling method according to various embodiments of this application described in the "Exemplary Methods" section above.

[0136] The basic principles of the present invention have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in the present invention are merely examples and not limitations, and should not be considered as essential features of each embodiment of the present invention. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the present invention to the necessity of employing the aforementioned specific details.

[0137] The block diagrams of devices, apparatuses, and systems involved in this invention are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0138] It should also be noted that in the apparatus and method of the present invention, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of the present invention.

[0139] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the invention. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the invention. Therefore, the invention is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0140] It should be understood that the qualifying terms "first", "second", "third", "fourth", "fifth" and "sixth" used in the description of the embodiments of the present invention are only used to more clearly illustrate the technical solutions and are not intended to limit the scope of protection of the present invention.

[0141] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the invention to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A method for equivalent modeling of a radiation source, characterized in that, The method includes: For each scanning surface of the target device in different orientations, acquire magnetic field amplitude data at multiple different distances; Multiple local equivalent source models are constructed based on the magnetic field amplitude data, and each local equivalent source model corresponds to one of the scanning surfaces. Based on the boundary line between adjacent scanning planes and the corresponding local equivalent source model, the field data at the boundary line is determined, and the field data includes amplitude data and phase data. Based on the field data, the multiple local equivalent source models are merged to obtain the equivalent radiation source model of the target device.

2. The radiation source equivalent modeling method according to claim 1, characterized in that, The local equivalent source model is composed of a magnetic dipole array; The parameters of each magnetic dipole in the magnetic dipole array include its spatial coordinates and complex magnetic dipole moment, the direction of which is characterized by three orthogonal components.

3. The radiation source equivalent modeling method according to claim 1, characterized in that, The process involves constructing multiple local equivalent source models based on the magnetic field amplitude data, including: For each scanned surface, perform the following operations: Construct an objective function to measure the joint deviation between the theoretical magnetic field amplitude generated based on a set of candidate equivalent source parameters corresponding to the scanning surface and the magnetic field amplitude data corresponding to the scanning surface. The objective function is optimized by adjusting the parameters of the candidate equivalent source until a preset condition is met, thereby determining the parameters of the local equivalent source model.

4. The radiation source equivalent modeling method according to claim 1, characterized in that, The process of constructing multiple local equivalent source models based on the magnetic field amplitude data includes: optimizing the parameters of the local equivalent source models using a differential evolution algorithm.

5. The radiation source equivalent modeling method according to claim 1, characterized in that, The determination of field data at the boundary line based on the boundary line of adjacent scan planes and the corresponding local equivalent source model includes: For multiple sampling points on the boundary line of adjacent scanning planes, the predicted complex magnetic field values ​​are obtained by forward calculation using the corresponding local equivalent source model. The predicted complex magnetic field values ​​include amplitude and phase values.

6. The radiation source equivalent modeling method according to claim 1, characterized in that, The merging of the multiple local equivalent source models based on the field data includes: A cost function is constructed to quantify the degree of discontinuity of the field data from adjacent local equivalent source models at the boundary line; With the goal of minimizing the cost function, the global phase reference of at least one of the local equivalent source models is adjusted so that the phases of the multiple local equivalent source models are aligned. The aligned local equivalent source models are merged to form the equivalent radiation source model of the target device.

7. The radiation source equivalent modeling method according to claim 6, characterized in that, The construction of the cost function includes: constructing the cost function based on the difference in the complex magnetic field prediction values ​​of adjacent local equivalent source models at the sampling points on the boundary line; The adjustment of the overall phase reference of at least one of the local equivalent source models includes: selecting a local equivalent source model as a reference, calculating and applying a complex rotation factor to the remaining local equivalent source models.

8. A radiation source equivalent modeling device, characterized in that, The device includes: The acquisition module is used to acquire magnetic field amplitude data at multiple different distances for each scan surface of the target device in different orientations; The construction module is used to construct multiple local equivalent source models based on the magnetic field amplitude data, and each local equivalent source model corresponds to one of the scanning surfaces; The determination module is used to determine the field data at the boundary line based on the boundary line of adjacent scanning planes and the corresponding local equivalent source model, wherein the field data includes amplitude data and phase data; The merging module is used to merge the multiple local equivalent source models based on the field data to obtain the equivalent radiation source model of the target device.

9. A non-transient computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, Including memory and processor; The memory is connected to the processor and is used to store programs; The processor is used to implement the radiation source equivalent modeling method as described in any one of claims 1 to 7 by running the program in the memory.

Images