System intelligent electromagnetic compatibility optimization design method and system thereof

Abstract

The invention relates to the technical field of electromagnetic compatibility optimization design methods, in particular to a system intelligent electromagnetic compatibility optimization design method and system, and the method comprises the following steps: building an electromagnetic simulation unit model of a component, and constructing a component model library; performing interference path analysis on the component model through a path optimization algorithm, identifying key nodes and interference source nodes, and constructing an electromagnetic interference path model; according to the electromagnetic interference path model, carrying out adaptive adjustment on electromagnetic radiation and immunity parameters of system components so as to optimize electromagnetic compatibility; carrying out dynamic prediction and parameter optimization adjustment on electromagnetic compatibility parameters of the system through a deep learning model; the electromagnetic compatibility of the system and the sub-modules thereof is integrally optimized by adopting a progressive feedback adjustment mechanism, and the technical contradiction in the traditional method is effectively solved by introducing innovative mechanisms such as path weight calculation, nonlinear adaptive adjustment and multi-level non-uniform optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of electromagnetic compatibility optimization design methods, and more specifically, to a system intelligent electromagnetic compatibility optimization design method and a system thereof. Background Art

[0002] As the complexity of electronic systems continues to increase, the importance of electromagnetic compatibility (EMC) issues in modern electronic devices has become increasingly significant. Especially in high-frequency circuits and dense component structures, various electromagnetic radiation and interference can easily cause the superposition of internal interference in the system, affecting the normal operation of the equipment. Therefore, how to effectively reduce the level of electromagnetic radiation and improve the system's immunity during the design stage has become a core technical problem in the electronics field.

[0003] Most of the existing electromagnetic compatibility optimization methods are based on static path analysis and fixed adjustment. They usually use preset models to allocate paths to interference sources and sensitive nodes, and use fixed parameter adjustment models to set the electromagnetic parameters of components. However, these traditional methods have the following significant defects. First, the path analysis method often lacks a recursive optimization algorithm, which leads to the system's inaccurate identification of interference paths and inability to dynamically respond to complex electromagnetic interference environments. For example, in high-frequency circuits where components are densely distributed, the weights of interference paths vary greatly. Static path analysis tends to ignore in-depth analysis of key nodes, resulting in a large number of redundant paths, which not only increases the amount of calculation, but also makes it difficult to improve the system's immunity.

[0004] Secondly, the existing methods mostly use a single fixed adjustment factor, which cannot achieve nonlinear adaptive adjustment and lacks the ability to sensitively adjust interference of different frequency bands and different intensities. For example, the traditional fixed factor adjustment cannot adapt to the frequently changing interference frequency in a complex environment, which often makes it difficult for electromagnetic radiation and anti-interference to reach the ideal optimization state. The adjustment effect of such methods on interference in different frequency bands is often limited, and the overall electromagnetic compatibility improvement effect of the system is not obvious. In addition, the traditional adjustment method ignores the synergistic relationship between the various parameters, and the efficiency and accuracy of the adjustment are low, and it is impossible to effectively suppress electromagnetic radiation while maintaining anti-interference.

[0005] In recent years, deep learning has been widely used in various optimization scenarios. However, most traditional electromagnetic compatibility optimization methods have failed to introduce deep learning models to achieve dynamic adaptive adjustment. Even though some methods use simple machine learning models for prediction, they fail to adjust component parameters in real time in a rapidly changing environment, making it difficult for electromagnetic compatibility to achieve dynamic response in a changing environment. At the same time, existing methods are insufficient in optimizing feedback adjustment, and usually use fixed-frequency feedback, which is difficult to adapt to the frequent adjustment needs of interference changes in complex environments. The lack of adaptive feedback adjustment methods limits the overall dynamic adaptability of the system, especially in a multi-level electromagnetic interference environment, where traditional feedback mechanisms make it difficult to achieve coordinated optimization of the sub-modules of the system.

[0006] The necessity of the present invention lies in solving the above-mentioned problems of the prior art in electromagnetic compatibility optimization, especially providing an intelligent adaptive optimization solution that can effectively identify and analyze electromagnetic interference paths, and optimize the electromagnetic characteristics of the system in real time through nonlinear adaptive adjustment and deep learning prediction, so that the system can achieve excellent compatibility in interference environments of different frequency bands and different intensities. Summary of the invention

[0007] In view of the deficiencies of the prior art, the present invention proposes an intelligent electromagnetic compatibility optimization design method and system to achieve dynamic adaptability to complex electromagnetic environments. Based on the path optimization recursive algorithm, the present invention can identify and dynamically adjust key interference paths and nodes, and through the optimization of path weights, accurately identify and effectively adjust path nodes in a multi-level interference environment. The recursive optimization path analysis method helps to screen out the main interference paths in a complex electromagnetic environment, while avoiding the waste of resources on redundant paths, greatly improving the system's anti-interference ability and reducing the impact of electromagnetic interference on key nodes.

[0008] The present invention provides a system intelligent electromagnetic compatibility optimization design method, comprising the following steps:

[0009] Obtaining the components of the system and their design parameters, composition structure and electromagnetic characteristics, establishing electromagnetic simulation unit models of the components and building a component model library;

[0010] Perform interference path analysis on the component model through a path optimization algorithm, identify key nodes and interference source nodes, and construct an electromagnetic interference path model;

[0011] Adaptively adjusting electromagnetic radiation and immunity parameters of system components according to the electromagnetic interference path model to optimize electromagnetic compatibility;

[0012] Dynamically predict and optimize the electromagnetic compatibility parameters of the system through a deep learning model;

[0013] A progressive feedback regulation mechanism is used to optimize the electromagnetic compatibility of the system and its submodules as a whole, so that the system meets the electromagnetic compatibility requirements.

[0014] Preferably, the path optimization algorithm adopts a path weight recursive calculation formula:

[0015]

[0016] Among them, ω(N i ,N j ) is node N i and N j The path weight between ij is the distance between two nodes, λ is the path attenuation factor, φ p is the interference source node weight, κ p is the sensitive node weight factor, δ ij is the electromagnetic interference difference value; the path weight is used to determine the key nodes in the electromagnetic interference path and generate the interference path model.

[0017] Preferably, the electromagnetic radiation and immunity parameter adaptive adjustment adopts a nonlinear adjustment formula:

[0018]

[0019] Among them, ΔE is the optimization amount, ɑ, β, γ are adjustment factors, f(x), g(x), h(x) are the electromagnetic characteristic functions of the components, and ψ(x) is the dynamic attenuation function, which is used to perform nonlinear adjustment on the electromagnetic radiation and anti-interference of the components.

[0020] Preferably, the system-level non-uniform electromagnetic compatibility optimization adopts a multi-level non-uniform optimization model, and the system-level optimal adjustment factor is calculated by the following formula:

[0021]

[0022] Among them, θ is the adjustment parameter set used to control the electromagnetic radiation and immunity of the system, representing the specific adjustment state of each component during the optimization process, θ opt is the optimal adjustment factor set of the system, which aims to obtain the best adjustment parameters for the overall electromagnetic compatibility through the optimization process. n is the number of components in the system, and each component has independent electromagnetic characteristic parameters. i is the electromagnetic radiation adjustment factor for each component, which is used to weight the electromagnetic radiation intensity of different components in the system, β i is the immunity adjustment factor, which is used to control the immunity adjustment amplitude of each component, σ i is the radiation parameter adjustment coefficient, which is used to determine the sensitivity of the electromagnetic radiation parameter changes during the adjustment process, τi is the anti-interference parameter adjustment coefficient, which is used to adjust the adjustment range of the anti-interference parameter. i is the current electromagnetic radiation level of the i-th component in the system, E 标准 is the standard electromagnetic radiation value required by the system, is the preset target radiation level, S i is the current immunity value of the i-th component, indicating its anti-interference ability in the electromagnetic environment, S 标准 is the standard anti-interference degree of the system, is the expected anti-interference target value of the system, η i is the electromagnetic radiation characteristic parameter of the i-th component, which is used to describe the expected state of the electromagnetic characteristics of the component, ξ i is the anti-interference characteristic parameter of the i-th component, which is used to describe the expected state of the component's anti-interference ability in an electromagnetic environment.

[0023] Preferably, the deep learning model realizes real-time optimization of electromagnetic compatibility parameters of components and systems through dynamic adaptive parameter prediction and adjustment, and the calculation formula is:

[0024]

[0025] Among them, Δλ is the parameter adjustment amount of the deep learning model, which represents the adjustment value of the electromagnetic compatibility parameters of the components in the current environment. σ is the learning rate adjustment factor of the deep learning model, which is used to control the response speed of the model to changes in electromagnetic compatibility. ML (P i ) is the prediction function output by the deep learning model, where P i represents the current electromagnetic characteristic input of the i-th component. The model generates the corresponding prediction value according to the input electromagnetic characteristics. ξ is the time-dependent adjustment factor used to adjust the adaptive prediction to balance the response degree of the adjustment result to the real-time data change. It is the time derivative of the optimal electromagnetic compatibility parameters of the jth component in the system, which is used to describe the rate of change of component parameters over time, and helps the deep learning model to achieve dynamic adaptive adjustment.

[0026] Preferably, the progressive feedback regulation mechanism is used to coordinately optimize the electromagnetic compatibility of the system and submodules, and the feedback regulation coefficient matrix is ​​used to achieve the overall regulation of the system. The feedback calculation formula is:

[0027]

[0028] Among them, ψ is the feedback adjustment coefficient set, which is used to adjust the electromagnetic compatibility parameters of each component of the system during the progressive feedback process. final is the final feedback adjustment coefficient set, which is the overall electromagnetic compatibility adjustment scheme obtained after feedback optimization, o is the total number of system submodules, δkk is the diagonal element of the feedback adjustment coefficient matrix, representing the self-feedback adjustment strength of each component or submodule in the system, which is used to control the feedback strength of individual parameters. k is the target interference level of the kth submodule, is the expected interference suppression value set, a and b are the upper and lower limits of the integral, indicating the starting and ending points of the time period in the feedback regulation process, δ ij is the non-diagonal element in the feedback adjustment coefficient matrix, which represents the mutual adjustment coefficient between different components and is used to coordinate the electromagnetic compatibility between components or sub-modules in the system. ω is the feedback adjustment frequency, which is used to adjust the response rate to electromagnetic changes in the feedback process. The higher the frequency value, the more frequent the feedback adjustment. t is the time variable, which is used to integrally calculate the influence of the adjustment frequency on the feedback process. M ij is the target interference suppression level of the feedback adjustment coefficient, indicating that the electromagnetic interference between different components or sub-modules affects the control target.

[0029] An intelligent electromagnetic compatibility optimization design system for executing the method comprises:

[0030] A data processing module is used for data collection, preprocessing, feature extraction and deep learning model training of the electromagnetic characteristics of components, including a data collection unit, a preprocessing unit, a feature extraction unit and a model training unit;

[0031] A simulation module, used to establish an electromagnetic simulation model of the electrical and electronic equipment, perform parameter simulation analysis and extract simulation result data, including a modeling submodule, a parameter simulation submodule and a simulation result submodule;

[0032] An optimization module, used to determine the simulation result parameters, predict the parameters using a deep learning model, and adjust the parameters using a progressive optimization algorithm, including an electromagnetic interference optimization unit, an electromagnetic radiation optimization unit, and an electromagnetic immunity optimization unit;

[0033] The visualization module is used to display the visualization of data processing results and simulation optimization results, including data visualization submodule and result visualization submodule.

[0034] Preferably, the model training unit in the data processing module trains the deep learning model through the component model library to evaluate the electromagnetic compatibility performance of components for calling by the optimization module.

[0035] Preferably, the electromagnetic interference optimization unit in the optimization module dynamically adjusts the key nodes in the electromagnetic interference path according to the simulation results and the path weight recursive algorithm, and optimizes the configuration of the interference source and sensitive nodes.

[0036] Preferably, the electromagnetic radiation optimization unit in the optimization module uses the nonlinear adaptive adjustment formula to optimize the electromagnetic radiation parameters of the components, so that the electromagnetic radiation of the components meets the electromagnetic compatibility standard of the system.

[0037] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0038] The present invention adopts the system engineering concept of combining "bottom-up" and "top-down", and innovatively divides the optimization process into four levels: device level, circuit level, system level and propagation path analysis. In terms of algorithm design, by introducing innovative mechanisms such as path weight calculation, nonlinear adaptive adjustment, and multi-level non-uniform optimization, the technical contradictions in traditional methods are effectively solved. For example, the path weight calculation method cleverly balances the contradiction between calculation accuracy and efficiency, while the nonlinear adaptive adjustment mechanism well solves the conflict between system stability and dynamic performance.

[0039] The method of the present invention has achieved remarkable technical effects in practical applications. Through the application of device-level deep learning models, accurate evaluation and prediction of electromagnetic characteristics are achieved, and the optimization efficiency is improved by more than 50%. In circuit-level optimization, automated optimization based on electromagnetic simulation engines shortens the design cycle by about 40%. The system-level multi-level collaborative optimization mechanism improves the overall electromagnetic compatibility index by more than 30%. Especially in the analysis of electromagnetic interference propagation paths, the innovative iterative reduction strategy enables the accuracy of critical path identification to reach more than 95%. These technical effects are achieved thanks to the synergy between the various innovative points of the present invention. For example, the accurate prediction of the deep learning model provides reliable data support for nonlinear adaptive regulation, while the dynamic feedback mechanism ensures the convergence and stability of the multi-level optimization process.

[0040] In general, the intelligent electromagnetic compatibility optimization design method and system provided by the present invention not only solves various technical problems existing in the prior art, but also achieves a qualitative leap in optimization effect through algorithm innovation and mechanism collaboration, providing a more scientific and efficient technical solution for the electromagnetic compatibility design of electronic equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is the overall logical structure of the system of the present invention.

[0042] Figure 2 This is the core algorithm logic block diagram of the present invention. DETAILED DESCRIPTION

[0043] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects thereof are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0044] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0045] The following is a clear and complete description of the technical solution of the present invention in conjunction with specific implementation methods. Figure 1-2 The method proposed in the present invention includes an intelligent electromagnetic compatibility optimization design method and a system thereof, which aims to improve the electromagnetic compatibility of the system and reduce electromagnetic interference.

[0046] The intelligent electromagnetic compatibility optimization design method of the present invention includes multiple key steps. First, the method establishes an electromagnetic simulation unit model of the components by acquiring the components of the system and their design parameters, composition structure and electromagnetic characteristics. For example, in a complex system containing multiple components, the electrical and electromagnetic characteristics of basic elements such as transistors, capacitors, resistors, etc., such as voltage tolerance value, electromagnetic immunity and radiation value, can be obtained, and the parameters of these components can be stored in the component model library in the data processing module 1. Preferably, in one embodiment of the present invention, the initial parameters of each component in the system are modeled within a frequency range of 11000MHz to ensure the comprehensiveness of the data. These simulation unit models can be established in a parameterized manner, such as using numerical models of electrical properties and structural parameters to convert the specific electromagnetic behavior of the components into adjustable parameters suitable for simulation, so as to facilitate subsequent electromagnetic compatibility optimization. Preferably, in one embodiment of the present invention, the established simulation unit model can be stored in the data processing module 1, which normalizes the electrical characteristics of different components to unify the data format for subsequent use.

[0047] Next, the present invention uses a path optimization recursive algorithm to perform interference path analysis on the component model, identify key nodes and interference source nodes, and construct an electromagnetic interference path model. For example, the path optimization recursive algorithm calculates the path weight ω of each component and recursively optimizes according to the path weight. The path weight ω(N i ,N j ) is calculated as:

[0048]

[0049] Among them, ω(N i,N j ) is node N i and N j The path weight between ij is the distance between two nodes, λ is the path attenuation factor, φ p is the interference source node weight, κ p is the sensitive node weight factor, δ ij is the electromagnetic interference difference value; the path weight is used to determine the key nodes in the electromagnetic interference path and generate the interference path model. Through this path weight calculation, the path that contributes most to the system interference can be identified, which helps to focus on it in the subsequent optimization steps.

[0050] In a practical application scenario, if λ=0.01, φ p =0.7,κ p =1.5, which can more effectively assign lower weights to components far away from interference sources, thereby highlighting paths with greater interference. The construction of this model is conducive to the subsequent optimization of key paths and reducing the occupation of computing resources by invalid paths. By controlling the λ parameter to adjust the path weights, the system can focus more on the path nodes with greater interference impact. This algorithm can effectively reduce the weight calculation burden of low-interference nodes, optimize the resource utilization of the system, and show stronger anti-interference performance in environments with wider frequency bands.

[0051] In the process of adaptively adjusting the electromagnetic radiation and immunity parameters, the electromagnetic radiation and immunity parameters of the components are adjusted according to the electromagnetic interference path model to optimize electromagnetic compatibility. Preferably, in one embodiment of the present invention, the nonlinear adjustment formula is:

[0052]

[0053] Among them, ΔE is the optimization amount, α, β, γ are adjustment factors, f(x), g(x), h(x) are the electromagnetic characteristic functions of the components, and ψ(x) is the dynamic attenuation function, which is used to nonlinearly adjust the electromagnetic radiation and immunity of the components. The adjusted parameters will be used to feed back to the electromagnetic compatibility analysis of the system as a whole, and can achieve interference suppression effect without changing the design of components.

[0054] In one embodiment, the adjustment factors α=0.1, β=0.3, and γ=0.6 are set to correspond to the structure, material, and sensitivity characteristics of the components, respectively, so as to achieve priority adjustment of the main interference sources, improve the system's adaptability to interference sources in different frequency bands, maintain the adaptability to different radiation sources in the case of multi-band interference, and reduce the overall electromagnetic radiation of the system. This optimization method enables the system to show good stability in both high-interference and low-interference environments by preferably setting the α, β, and γ parameters.

[0055] In the system-level non-uniform electromagnetic compatibility optimization, the present invention uses a multi-level non-uniform optimization model to perform hierarchical adjustment on the system electromagnetic radiation and immunity, and the calculation formula is as follows:

[0056]

[0057] Among them, θ is the adjustment parameter set used to control the electromagnetic radiation and immunity of the system, representing the specific adjustment state of each component during the optimization process, θ opt is the optimal adjustment factor set of the system, which aims to obtain the best adjustment parameters for the overall electromagnetic compatibility through the optimization process. n is the number of components in the system, and each component has independent electromagnetic characteristic parameters. i is the electromagnetic radiation adjustment factor for each component, which is used to weight the electromagnetic radiation intensity of different components in the system, β i is the immunity adjustment factor, which is used to control the immunity adjustment amplitude of each component, σ i is the radiation parameter adjustment coefficient, which is used to determine the sensitivity of the electromagnetic radiation parameter changes during the adjustment process, τ i is the anti-interference parameter adjustment coefficient, which is used to adjust the adjustment range of the anti-interference parameter. i is the current electromagnetic radiation level of the i-th component in the system, E 标准 is the standard electromagnetic radiation value required by the system, is the preset target radiation level, S i is the current immunity value of the i-th component, indicating its anti-interference ability in the electromagnetic environment, S 标准 is the standard anti-interference degree of the system, is the expected anti-interference target value of the system, η i is the electromagnetic radiation characteristic parameter of the i-th component, which is used to describe the expected state of the electromagnetic characteristics of the component, ξ i is the anti-interference characteristic parameter of the ith component, which is used to describe the expected state of the anti-interference ability of the component in the electromagnetic environment. This makes the system more compatible in different electromagnetic environments, especially in the scenario of unstable frequency, showing excellent electromagnetic anti-interference.

[0058] Preferably, in one embodiment of the present invention, a multi-level optimization model adjusts the overall electromagnetic radiation and immunity parameters of the system layer by layer based on non-uniform distribution to achieve a global optimal effect.

[0059] Preferably, in one embodiment of the present invention, n is set i and i The value range is 4060 to ensure a wide range of electromagnetic compatibility adjustment to meet the electromagnetic compatibility requirements of different systems.

[0060] In addition, the deep learning adaptive model is used to optimize the electromagnetic compatibility parameters of components and systems in real time, and dynamic prediction and parameter adjustment are achieved through the following formula:

[0061]

[0062] Among them, Δλ is the parameter adjustment amount of the deep learning model, which represents the adjustment value of the electromagnetic compatibility parameters of the components in the current environment. σ is the learning rate adjustment factor of the deep learning model, which is used to control the response speed of the model to changes in electromagnetic compatibility. ML (P i ) is the prediction function output by the deep learning model, where P i represents the current electromagnetic characteristic input of the i-th component. The model generates the corresponding prediction value according to the input electromagnetic characteristics. ξ is the time-dependent adjustment factor used to adjust the adaptive prediction to balance the response degree of the adjustment result to the real-time data change. It is the time derivative of the optimal electromagnetic compatibility parameters of the jth component in the system, which is used to describe the rate of change of component parameters over time, and helps the deep learning model to achieve dynamic adaptive adjustment.

[0063] Set σ=0.05 and ξ=0.1 to adapt to the electromagnetic variation characteristics of most components. For example, when the electromagnetic parameters of a component in the system suddenly change, the deep learning model can adjust the corresponding radiation value and immunity parameters in time, effectively improving the dynamic response speed of the system. By implementing adaptive parameter adjustment through the deep learning model, the present invention can quickly respond to changes in electromagnetic parameters within the system and effectively adjust the immunity and radiation values ​​in a short time.

[0064] Preferably, in one embodiment of the present invention, the deep learning model can continuously receive electromagnetic characteristic data at the system level and the device level to achieve adaptive parameter adjustment.

[0065] In the progressive feedback regulation mechanism, the present invention optimizes the electromagnetic compatibility of the system and its submodules as a whole through a progressive feedback regulation algorithm to meet the compatibility requirements of the overall system. The progressive feedback regulation mechanism is based on the following feedback regulation formula:

[0066]

[0067] Among them, ψ is the feedback adjustment coefficient set, which is used to adjust the electromagnetic compatibility parameters of each component of the system during the progressive feedback process. final is the final feedback adjustment coefficient set, which is the overall electromagnetic compatibility adjustment scheme obtained after feedback optimization, o is the total number of system submodules, δ kkis the diagonal element of the feedback adjustment coefficient matrix, representing the self-feedback adjustment strength of each component or submodule in the system, which is used to control the feedback strength of individual parameters. k is the target interference level of the kth submodule, is the expected interference suppression value set, a and b are the upper and lower limits of the integral, indicating the starting and ending points of the time period in the feedback regulation process, δ ij is the non-diagonal element in the feedback adjustment coefficient matrix, which represents the mutual adjustment coefficient between different components and is used to coordinate the electromagnetic compatibility between components or sub-modules in the system. ω is the feedback adjustment frequency, which is used to adjust the response rate to electromagnetic changes in the feedback process. The higher the frequency value, the more frequent the feedback adjustment. t is the time variable, which is used to integrally calculate the influence of the adjustment frequency on the feedback process. M ij is the target interference suppression level of the feedback adjustment coefficient, indicating that the electromagnetic interference between different components or sub-modules affects the control target.

[0068] Preferably, ω=0.2, δ ij =0.8 to balance the feedback frequency of the submodule and make the electromagnetic compatibility adjustment effect more stable, thereby achieving compatibility optimization of the entire system and maintaining a high level of electromagnetic compatibility of the system, which is especially applicable in complex electromagnetic environments.

[0069] Preferably, in one embodiment of the present invention, the feedback adjustment mechanism will perform feedback adjustment on key parameters at a certain frequency when optimizing the electromagnetic radiation and anti-interference performance of the system, so as to ensure that the system maintains stable electromagnetic compatibility when the external environment changes.

[0070] Corresponding to the above method, the present invention also discloses an intelligent electromagnetic compatibility optimization design system, including multiple functional modules. First, the data processing module 1 is used to collect, process and store the electrical and electromagnetic characteristic parameters of each component, and ensure the consistency of the input data and the model training effect by preprocessing and extracting the features of the original data. Preferably, in one embodiment of the present invention, the data processing module 1 may include normalization and noise reduction steps to reduce data redundancy and improve system processing efficiency.

[0071] Secondly, the simulation module 2 is used to create an electromagnetic simulation model of electrical and electronic equipment, support parameter simulation analysis and extract simulation result data. The simulation module 2 may specifically include a modeling submodule, a parameter simulation submodule and a simulation result submodule to ensure compatibility testing and evaluation of the system in different scenarios. Preferably, in one embodiment of the present invention, the simulation result data can be directly input into the optimization module 3 to achieve rapid parameter adjustment.

[0072] Then, the optimization module 3 is responsible for predicting and adjusting the electromagnetic interference path, radiation and immunity parameters using the deep learning model. The optimization module 3 includes the electromagnetic interference parameter optimization unit, the electromagnetic radiation optimization unit and the electromagnetic immunity optimization unit. Specifically, these units will adjust the interference source and sensitive nodes according to the analysis results of the electromagnetic interference path to ensure that the system electromagnetic compatibility reaches the optimal state.

[0073] Finally, the system also includes a visualization module 4 for graphically displaying the optimization design and simulation results. The data visualization submodule is used to display the process results of data processing and simulation analysis, and the result visualization submodule displays the final effect of system optimization. Preferably, in one embodiment of the present invention, the visualization module 4 can also realize data monitoring during the optimization process, so that the user can understand the progress of electromagnetic compatibility optimization in real time.

[0074] Preferably, in one embodiment of the present invention, the model training unit of the data processing module 1 can train the deep learning model through the training data of the component model library to evaluate the electromagnetic compatibility parameters of the components. The real-time updated model can make the optimization module more intelligent and flexible in dealing with interference in different electrical environments.

[0075] Secondly, the electromagnetic interference parameter optimization unit in the optimization module 3 dynamically adjusts the key nodes according to the simulation results and the path optimization recursive algorithm. Preferably, in one embodiment of the present invention, the optimized electromagnetic interference path will be used to further optimize the electromagnetic radiation and immunity parameters to meet the compatibility standards of the entire system.

[0076] In addition, the electromagnetic radiation optimization unit uses a nonlinear adaptive adjustment formula to optimize the electromagnetic radiation parameters of the components. Preferably, in an embodiment of the present invention, the optimized electromagnetic radiation level will be limited to a specific range so that the system meets the electromagnetic compatibility requirements in different environments.

[0077] In order to verify the superiority of the intelligent electromagnetic compatibility optimization design method of the present invention, the present invention provides a specific embodiment and compares the performance through comparative examples to demonstrate its significant improvement effect in electromagnetic compatibility. The test indicators include electromagnetic radiation level, anti-interference, path weight distribution uniformity and deep learning prediction response time. These indicators are set around the innovative points of the present invention and are intended to reflect its stability and adaptability in different environments. The specific configurations and test results of Example 1 and Comparative Example 1 are described in detail below.

[0078] In Example 1, the electromagnetic compatibility optimization system is implemented based on the method of the present invention, and specifically adopts the following configuration:

[0079] Electromagnetic interference path analysis: The path weight of each key node and interference source node is calculated through the path optimization recursive algorithm to better identify and adjust the paths with greater electromagnetic interference impact. The path weight recursive coefficient λ is set to 0.05 to ensure accuracy.

[0080] Nonlinear adaptive adjustment: The electromagnetic radiation and immunity parameters are dynamically adjusted using nonlinear adjustment formulas. The adjustment factors α, β, and γ are set to 0.1, 0.3, and 0.6, respectively, to balance the various electromagnetic characteristics.

[0081] Deep learning model: Use the deep learning model to predict real-time data, with the learning rate adjustment factor σ being 0.04 and ξ being 0.1 to adapt to the frequently changing electromagnetic environment.

[0082] Feedback regulation: The progressive feedback regulation mechanism is used to optimize the overall electromagnetic compatibility of the system. The feedback frequency adjustment factor ω is set to 0.2 to balance the response speed and stability of the system. In this embodiment, the test object includes an electronic system in a complex electromagnetic environment. The system includes a variety of components that are susceptible to electromagnetic interference. The frequency range of the tested electromagnetic environment is 1-1000MHz.

[0083] In this embodiment, the test object includes an electronic system in a complex electromagnetic environment. The system includes a variety of components susceptible to electromagnetic interference. The frequency range of the tested electromagnetic environment is 1-1000 MHz.

[0084] In Comparative Example 1, the test used a conventional electromagnetic compatibility optimization design method, which did not adopt a path optimization recursive algorithm and nonlinear adaptive adjustment, and did not integrate a deep learning model for adaptive prediction. The feedback adjustment mechanism also had a fixed frequency and could not be adaptively adjusted with the environment.

[0085] Electromagnetic interference path analysis: Uniform path weight distribution is adopted, path optimization recursion is not performed, key nodes cannot be identified, and path interference distribution is relatively even.

[0086] Nonlinear regulation: A single regulation factor α=0.2 is used to perform fixed regulation on electromagnetic radiation, without dynamic adaptive characteristics.

[0087] Deep learning model: The deep learning prediction model is not used, and the system parameters need to be manually adjusted when the electromagnetic characteristics change.

[0088] Feedback adjustment: fixed feedback frequency, frequency factor ω=0.1, limited adjustment range.

[0089] The main indicators of the test include the following aspects:

[0090] 1. Electromagnetic radiation level: The electromagnetic radiation of the measuring system in different frequency bands is evaluated according to the IEC6100043 standard. The lower the value, the better the electromagnetic radiation control of the system.

[0091] 2. Immunity: According to the IEC6100046 standard, the system's immunity to electromagnetic interference is measured in the 1-1000MHz frequency band. The higher the immunity, the stronger the system's ability to resist interference.

[0092] 3. Path weight distribution uniformity: Test the uniformity of interference distribution of each path in the system, measured by mean square error (MSE). The smaller the mean square error, the more accurate the system is in identifying interference sources.

[0093] 4. Deep learning prediction response time: Measures the response time of the deep learning model when the electromagnetic characteristics change, in milliseconds (ms). The shorter the response time, the stronger the system's adaptability to environmental changes.

[0094] The test results are as follows

[0095]

[0096]

[0097] It can be seen from the above test results that Example 1 of the present invention is superior to Comparative Example 1 in all key indicators. In terms of electromagnetic radiation level, the electromagnetic radiation of Example 1 is only 45dBμV / m, which is significantly lower than 62dBμV / m of Comparative Example 1, indicating that the present invention more effectively controls the electromagnetic radiation of the system through path optimization and nonlinear regulation, and reduces the leakage of electromagnetic interference. In the anti-interference test, Example 1 reached 80dBμA, which is much higher than 65dBμA of Comparative Example 1, which means that under the same electromagnetic environment, Example 1 can better resist external electromagnetic interference, showing superior performance in complex electromagnetic environments.

[0098] In terms of path weight distribution uniformity, the mean square error of Example 1 is 0.15, while the mean square error of Comparative Example 1 is 0.45. The lower mean square error indicates that Example 1 is more accurate in identifying interference source paths, effectively avoiding interference calculations on irrelevant paths, and significantly optimizing system resources. Finally, in terms of deep learning prediction response time, the response time of Example 1 is 5ms, showing that the system can adjust quickly in a rapidly changing electromagnetic environment, while Comparative Example 1 is not configured with a deep learning model and cannot respond in time when parameters change drastically.

[0099] In summary, the optimization method of the electromagnetic compatibility of the present invention not only significantly reduces electromagnetic radiation and improves anti-interference, but also performs well in resource optimization and response speed, and is suitable for electronic systems in complex electromagnetic environments.

[0100] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Intelligent electromagnetic compatibility optimization design method, characterized in that: The following steps are involved: Obtaining the components of the system and their design parameters, composition structure and electromagnetic characteristics, establishing electromagnetic simulation unit models of the components and building a component model library; Perform interference path analysis on the component model through a path optimization algorithm, identify key nodes and interference source nodes, and construct an electromagnetic interference path model; Adaptively adjusting electromagnetic radiation and immunity parameters of system components according to the electromagnetic interference path model to optimize electromagnetic compatibility; Dynamically predict and optimize the electromagnetic compatibility parameters of the system through a deep learning model; A progressive feedback regulation mechanism is used to optimize the electromagnetic compatibility of the system and its submodules as a whole, so that the system meets the electromagnetic compatibility requirements.

2. The intelligent electromagnetic compatibility optimization design method according to claim 1 is characterized in that: The path optimization algorithm adopts the path weight recursive calculation formula: Among them, ω(N i ,N j ) is node N i and N j The path weight between ij is the distance between two nodes, λ is the path attenuation factor, φ p is the interference source node weight, κ p is the sensitive node weight factor, δ ij is the electromagnetic interference difference value; the path weight is used to determine the key nodes in the electromagnetic interference path and generate the interference path model.

3. The intelligent electromagnetic compatibility optimization design method according to claim 1 is characterized in that: The electromagnetic radiation and immunity parameters are adaptively adjusted using a nonlinear adjustment formula: Among them, ΔE is the optimization amount, α, β, γ are adjustment factors, f(x), g(x), h(x) are the electromagnetic characteristic functions of the components, and ψ(x) is the dynamic attenuation function, which is used to perform nonlinear adjustment on the electromagnetic radiation and anti-interference of the components.

4. The intelligent electromagnetic compatibility optimization design method according to claim 1 is characterized in that: The system-level non-uniform electromagnetic compatibility optimization adopts a multi-level non-uniform optimization model and calculates the system-level optimal adjustment factor through the following formula: Among them, θ is the adjustment parameter set used to control the electromagnetic radiation and immunity of the system, representing the specific adjustment state of each component during the optimization process, θ opt is the optimal adjustment factor set of the system, which aims to obtain the best adjustment parameters for the overall electromagnetic compatibility through the optimization process. n is the number of components in the system, and each component has independent electromagnetic characteristic parameters. i is the electromagnetic radiation adjustment factor for each component, which is used to weight the electromagnetic radiation intensity of different components in the system, β i is the immunity adjustment factor, which is used to control the immunity adjustment amplitude of each component, σ i is the radiation parameter adjustment coefficient, which is used to determine the sensitivity of the electromagnetic radiation parameter changes during the adjustment process, τ i is the anti-interference parameter adjustment coefficient, which is used to adjust the adjustment range of the anti-interference parameter. i is the current electromagnetic radiation level of the i-th component in the system, E 标准 is the standard electromagnetic radiation value required by the system, is the preset target radiation level, S i is the current immunity value of the i-th component, indicating its anti-interference ability in the electromagnetic environment, S 标准 is the standard anti-interference degree of the system, is the expected anti-interference target value of the system, η i is the electromagnetic radiation characteristic parameter of the i-th component, which is used to describe the expected state of the electromagnetic characteristics of the component, ξ i is the anti-interference characteristic parameter of the i-th component, which is used to describe the expected state of the component's anti-interference ability in an electromagnetic environment.

5. The intelligent electromagnetic compatibility optimization design method according to claim 1 is characterized in that: The deep learning model achieves real-time optimization of electromagnetic compatibility parameters of components and systems through dynamic adaptive parameter prediction and adjustment. The calculation formula is: Among them, Δλ is the parameter adjustment amount of the deep learning model, which represents the adjustment value of the electromagnetic compatibility parameters of the components in the current environment. σ is the learning rate adjustment factor of the deep learning model, which is used to control the response speed of the model to changes in electromagnetic compatibility. ML (P i ) is the prediction function output by the deep learning model, where P i represents the current electromagnetic characteristic input of the i-th component. The model generates the corresponding prediction value according to the input electromagnetic characteristics. ξ is the time-dependent adjustment factor used to adjust the adaptive prediction to balance the response degree of the adjustment result to the real-time data change. It is the time derivative of the optimal electromagnetic compatibility parameters of the jth component in the system, which is used to describe the rate of change of component parameters over time, and helps the deep learning model to achieve dynamic adaptive adjustment.

6. The intelligent electromagnetic compatibility optimization design method according to claim 1 is characterized in that: The progressive feedback adjustment mechanism is used to coordinately optimize the electromagnetic compatibility of the system and submodules, and the feedback adjustment coefficient matrix is ​​used to achieve the overall adjustment of the system. The feedback calculation formula is: Among them, ψ is the feedback adjustment coefficient set, which is used to adjust the electromagnetic compatibility parameters of each component of the system during the progressive feedback process. final is the final feedback adjustment coefficient set, which is the overall electromagnetic compatibility adjustment scheme obtained after feedback optimization, o is the total number of system submodules, δ kk is the diagonal element of the feedback adjustment coefficient matrix, representing the self-feedback adjustment strength of each component or submodule in the system, which is used to control the feedback strength of individual parameters. k is the target interference level of the kth submodule, is the expected interference suppression value set, a and b are the upper and lower limits of the integral, indicating the starting and ending points of the time period in the feedback regulation process, δ ij is the non-diagonal element in the feedback adjustment coefficient matrix, which represents the mutual adjustment coefficient between different components and is used to coordinate the electromagnetic compatibility between components or sub-modules in the system. ω is the feedback adjustment frequency, which is used to adjust the response rate to electromagnetic changes in the feedback process. The higher the frequency value, the more frequent the feedback adjustment. t is the time variable, which is used to integrally calculate the influence of the adjustment frequency on the feedback process. M ij is the target interference suppression level of the feedback adjustment coefficient, indicating that the electromagnetic interference between different components or sub-modules affects the control target.

7. An intelligent electromagnetic compatibility optimization design system for executing the method according to any one of claims 16, characterized in that: include: A data processing module is used for data collection, preprocessing, feature extraction and deep learning model training of the electromagnetic characteristics of components, including a data collection unit, a preprocessing unit, a feature extraction unit and a model training unit; A simulation module, used to establish an electromagnetic simulation model of the electrical and electronic equipment, perform parameter simulation analysis and extract simulation result data, including a modeling submodule, a parameter simulation submodule and a simulation result submodule; An optimization module, used to determine the simulation result parameters, predict the parameters using a deep learning model, and adjust the parameters using a progressive optimization algorithm, including an electromagnetic interference optimization unit, an electromagnetic radiation optimization unit, and an electromagnetic immunity optimization unit; The visualization module is used to display the visualization of data processing results and simulation optimization results, including data visualization submodule and result visualization submodule.

8. The intelligent electromagnetic compatibility optimization design system according to claim 7 is characterized in that: The model training unit in the data processing module trains the deep learning model through the component model library to evaluate the electromagnetic compatibility performance of components for calling by the optimization module.

9. The intelligent electromagnetic compatibility optimization design system according to claim 7, characterized in that: The electromagnetic interference optimization unit in the optimization module dynamically adjusts the key nodes in the electromagnetic interference path according to the simulation results and the path weight recursive algorithm, and optimizes the configuration of the interference source and sensitive nodes.

10. The intelligent electromagnetic compatibility optimization design system according to claim 7, characterized in that: The electromagnetic radiation optimization unit in the optimization module uses the nonlinear adaptive adjustment formula to optimize the electromagnetic radiation parameters of the components, so that the electromagnetic radiation of the components meets the electromagnetic compatibility standard of the system.

Images