Test methods and systems for electromagnetic radiation immunity of components

Abstract

This invention provides a method and system for testing the electromagnetic radiation immunity of components, relating to the field of electromagnetic immunity testing technology. The invention sets a range for the electromagnetic frequency and amplitude of the interference signal, generating multiple particles characterizing the interference signal within this range. Through iterative particle swarm analysis, it rapidly collects highly sensitive sample points that cause functional abnormalities in components during electromagnetic interference testing. By clustering these highly sensitive samples, it locates the electromagnetic interference sensitive range of the target component. Compared to full-band scanning and fixed-point scanning methods for electromagnetic interference testing, this invention enables automated and quantitative extraction of the electromagnetic sensitivity range, reducing subjective bias from manual location and improving analytical consistency and repeatability. Furthermore, by focusing on highly sensitive frequency bands and field strength ranges, it optimizes the test path, reduces redundant scanning points, and significantly improves the efficiency of electromagnetic interference testing.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic immunity testing technology, specifically to a method and system for testing the electromagnetic radiation immunity of components. Background Technology

[0002] For testing the electromagnetic radiation immunity of electronic components, the conventional method is to place the components in a standard test environment that can shield them from external interference and eliminate internal reflections, and then test the input and output of the target components using electromagnetic signals of different frequency bands. However, in the expensive electromagnetic "black box" test environment, scanning the target components sequentially with electromagnetic signals of all frequencies would make the testing time for a single component extremely slow and inefficient; while using electromagnetic signals of a fixed few frequency bands for interference testing makes it difficult to comprehensively identify the sensitive response areas of the target components in complex electromagnetic environments.

[0003] In existing technologies, some methods attempt to improve testing efficiency and electromagnetic signal coverage by using preset interference modes or segmented scanning. However, the combination of preset parameters lacks adaptability and intelligent guidance, resulting in high redundancy in the testing process and limited ability to capture nonlinear sensitive features. Furthermore, most methods only use pass or fail as the evaluation result for a specific interference frequency, lacking cluster analysis and boundary quantification of the sensitive frequency bands of components, and thus failing to provide detailed data support for subsequent protection design.

[0004] To address the aforementioned shortcomings, there is a lack of an efficient testing method in this field that can automatically explore the parameter space and adaptively focus on sensitive areas to improve testing efficiency and accuracy and meet the stringent electromagnetic immunity assessment requirements of electronic components. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for testing the electromagnetic radiation immunity of components, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for testing the electromagnetic radiation immunity of a component, comprising the following steps:

[0008] S1: Set the range of electromagnetic frequency and electromagnetic amplitude, generate multiple particles within the range, each particle corresponds to a position feature composed of frequency and amplitude, set the frequency step size and amplitude step size as the initial velocity feature of all particles, initialize the optimal interference degree of each particle to 0, and the optimal position is the initial position feature.

[0009] S2: Apply electromagnetic interference to the target component based on the position characteristics of each particle and collect its feedback signal. Compare and analyze the preset standard response with the feedback signal, and obtain the current interference degree of each particle according to the interference degree calculation rule. For particles whose current interference degree is greater than the preset interference threshold, extract their position characteristics and classify them into the high-sensitivity sample set.

[0010] S3: For each particle, if the current disturbance degree is greater than the optimal disturbance degree, update its optimal disturbance degree to the current disturbance degree, update its optimal position to the current position feature, select the optimal disturbance degree with the largest value among all particles as the global disturbance degree, and its corresponding position feature as the global optimal position.

[0011] S4: For each particle, calculate the individual deviation between its current position feature and its optimal position, as well as the global deviation from the global optimal position. Weigh and fuse the obtained deviation with the current velocity feature. After boundary correction of the weighted result, we obtain a new velocity feature. We add the current position feature and the new velocity feature to obtain a new position feature. We maintain the new position feature within the range of values ​​through boundary correction and construct a new particle swarm.

[0012] S5: Repeat S2 to S4. When the global interference degree meets the preset convergence condition and reaches the preset round limit, the iteration ends. Perform cluster analysis on all positional features in the high-sensitivity sample set. The value boundaries of electromagnetic frequency and electromagnetic amplitude in each cluster are the electromagnetic interference sensitive range of the target component.

[0013] Furthermore, the logic for generating multiple particles within the set value range is as follows: the range of electromagnetic frequency values ​​is equally divided into multiple frequency intervals to obtain each frequency division point; the range of electromagnetic amplitude values ​​is equally divided into multiple amplitude intervals to obtain each amplitude division point; a two-dimensional orthogonal grid is constructed using all frequency division points and amplitude division points, where the frequency-amplitude coordinates corresponding to each node are determined as the frequency-amplitude parameters of a particle and used as the initial position feature of the particle.

[0014] Furthermore, the method of applying electromagnetic interference to the target component is as follows: place the target component in an environment without electromagnetic interference, extract the position characteristics of the current particles, set the carrier frequency of the interference signal according to its frequency parameters, set the electromagnetic field strength of the interference signal according to its amplitude parameters, and apply the set electromagnetic interference signal to the target component.

[0015] Furthermore, the preset standard response is the output signal of the target component in normal working condition under interference-free environment;

[0016] The specific method for calculating the current interference degree of the particle is as follows: calculate the normalized mean square error between the feedback signal and the standard response data, and use the result as the distortion degree of the target component output feedback signal under the current interference particle; set multi-polarization judgment rules, and label the failure level of the obtained feedback signal according to the distortion degree, interruption status and logic error of the feedback signal; and perform weighted fusion of the distortion degree and failure level of the feedback signal to obtain the interference degree of the current particle.

[0017] Furthermore, the calculation method for the individual deviation and the global deviation is as follows: based on the current interference degree of each particle, the optimal interference degree and optimal position of all particles in the particle swarm are updated, and the global interference degree and global optimal position of the current test round are determined.

[0018] For each particle, calculate the difference between its optimal position and current position features in terms of frequency and amplitude components. The results are individual frequency difference and individual amplitude difference, respectively. The vector formed by these two differences is the individual deviation of the particle.

[0019] For each particle, calculate the difference between the optimal position in the current test round and the particle's current position features in terms of frequency and amplitude components. The results are the global frequency difference and the global amplitude difference, respectively, and the vector formed by these two differences is the global deviation of the particle.

[0020] Furthermore, the new velocity characteristics are determined jointly based on the particle's current velocity characteristics, individual deviations, and global deviations. The specific calculation process includes:

[0021] Set the inertia weight coefficient, individual adjustment coefficient, and global adjustment coefficient respectively;

[0022] Multiply the current velocity feature by the inertia weighting coefficient to obtain the frequency component and amplitude component of the current velocity feature, respectively.

[0023] Multiplying the individual deviation by the individual adjustment coefficient yields the individual frequency adjustment component and the individual amplitude adjustment component, respectively.

[0024] Multiplying the global deviation by the global adjustment coefficient yields the global frequency adjustment component and the global amplitude adjustment component, respectively.

[0025] The frequency component, individual frequency adjustment component, and global frequency adjustment component of the current velocity feature are added together, and the result is taken as the frequency change; the amplitude component, individual amplitude adjustment component, and global amplitude adjustment component of the current velocity feature are added together, and the result is taken as the amplitude change.

[0026] A new velocity characteristic is obtained by performing boundary correction on the vector formed by the obtained frequency change and amplitude change.

[0027] Furthermore, the logic for obtaining the new velocity features and the new position features is as follows:

[0028] The calculated frequency change and amplitude change are corrected to integer multiples of the frequency step size and the amplitude step size, respectively. The absolute value of the difference between the corrected frequency change and its original value is less than the frequency step size, and the absolute value of the difference between the corrected amplitude change and its original value is less than the amplitude step size. The corrected frequency change and amplitude change constitute a new velocity characteristic.

[0029] The current position feature is added to the corrected new velocity feature. If the resulting frequency value exceeds the range of electromagnetic frequency of the interference signal, it is corrected to the boundary value of the electromagnetic frequency range. If the resulting amplitude value exceeds the range of electromagnetic amplitude of the interference signal, it is corrected to the boundary value of the electromagnetic amplitude range. This yields the new position feature after boundary correction.

[0030] Furthermore, the logic for determining the electromagnetic interference sensitive range of the target component using cluster analysis is as follows:

[0031] Treat each location feature in the high-sensitivity sample set as an independent cluster; calculate the inter-cluster distance between any two clusters by mapping the samples contained in any two clusters pairwise, traversing all sample combinations, calculating the Euclidean distance between each pair of samples, and selecting the maximum value of the obtained Euclidean distance as the inter-cluster distance between the two clusters; construct the dissimilarity matrix using all the calculated inter-cluster distances.

[0032] Iterative processing of the dissimilarity matrix: Locate the two clusters with the smallest inter-cluster distance among all current cluster pairs and merge them into a new cluster; Update the dissimilarity matrix, recalculate the distance between the new cluster and other clusters, and repeat the above merging process until the minimum inter-cluster distance in the cluster pair is lower than the preset convergence threshold and the number of clusters obtained meets the value range of the preset number of clusters, then stop merging and obtain the partitioning result;

[0033] For each cluster, calculate the geometric center of the positional features of all samples within the cluster. Using the obtained geometric center as a reference, calculate the standard deviation of the sample points within the cluster in terms of frequency and amplitude parameters. Weight the standard deviation of the frequency parameter and the standard deviation of the amplitude parameter to obtain the boundary of the range of electromagnetic frequency and electromagnetic amplitude values ​​corresponding to the cluster.

[0034] By integrating the ranges of all cluster electromagnetic frequencies and amplitudes, the electromagnetic interference sensitive range of the target component is formed.

[0035] The present invention also provides a testing system for the electromagnetic radiation immunity of components, the testing system for the electromagnetic radiation immunity of components being used to perform the above-described testing method for the electromagnetic radiation immunity of components, including:

[0036] Initialization module: Used to set the range of electromagnetic frequency and electromagnetic amplitude, generate multiple particles within the range, each particle corresponds to a position feature composed of frequency and amplitude, set the frequency step size and amplitude step size as the initial velocity feature of all particles, initialize the optimal disturbance degree of each particle to 0, and the optimal position is the initial position feature.

[0037] Signal interference module: It is used to apply electromagnetic interference to target components based on the position characteristics of each particle and collect its feedback signal. It compares and analyzes the preset standard response with the feedback signal, and obtains the current interference degree of each particle according to the interference degree calculation rules. For particles whose current interference degree is greater than the preset interference threshold, its position characteristics are extracted and it is included in the high-sensitivity sample set.

[0038] Parameter update module: For each particle, if the current disturbance degree is greater than the optimal disturbance degree, update the optimal disturbance degree to the current disturbance degree and the optimal position to the current position feature. Select the optimal disturbance degree with the largest value among all particles as the global disturbance degree and its corresponding position feature as the global optimal position.

[0039] Particle Update Module: For each particle, calculate the individual deviation between its current position feature and its optimal position, as well as the global deviation from the global optimal position. The obtained deviation is weighted and fused with the current velocity feature. After boundary correction of the weighted result, a new velocity feature is obtained. The current position feature is added to the new velocity feature to obtain a new position feature. The boundary correction keeps the new position feature within the range of values ​​and constructs a new particle swarm.

[0040] Clustering analysis module: used from the repetitive signal interference module to the particle update module. When the global interference degree meets the preset convergence condition and reaches the preset round limit, the iteration ends. Clustering analysis is performed on all positional features in the high-sensitivity sample set. The value boundaries of electromagnetic frequency and electromagnetic amplitude in each cluster are the electromagnetic interference sensitive range of the target component.

[0041] Compared with the prior art, the beneficial effects of the present invention are:

[0042] This invention pre-sets the range of electromagnetic frequency and amplitude values ​​for the interference signal of the target component, and achieves adaptive exploration of the parameter space based on the particle swarm initialization and iterative update mechanism. It can quickly and directionally approach the sensitive response area of ​​the component, significantly improving the testing efficiency and coverage of sensitive frequency points compared with traditional uniform or fixed-point scanning methods.

[0043] This invention introduces an interference assessment and high-sensitivity sample acquisition mechanism to dynamically update the individual and global optimal positions during iteration. This enables the testing process to focus on the actual sensitive electromagnetic frequencies of the components, avoiding redundant testing in invalid electromagnetic frequency bands. At the same time, boundary correction ensures stable convergence of the search process, enhancing the reliability of the testing method under complex electromagnetic interference.

[0044] This invention quantifies the electromagnetic interference sensitivity range of components by performing cluster analysis on highly sensitive samples collected in all iterations and outputting the frequency and amplitude value boundaries of each sensitive cluster. This makes the electromagnetic interference test of the same type of component repeatable and greatly improves the efficiency of electromagnetic interference testing. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the overall method flow of the present invention;

[0046] Figure 2 The initial particle distribution diagram provided in the embodiments of the present invention;

[0047] Figure 3 This is a schematic diagram illustrating the global interference convergence trend provided in an embodiment of the present invention;

[0048] Figure 4 This is a distribution diagram of interference characteristics of highly sensitive samples provided in an embodiment of the present invention;

[0049] Figure 5 This is a structural block diagram of the overall system of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0051] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0052] Example:

[0053] Please see Figures 1 to 4 The present invention provides a technical solution:

[0054] A method for testing the electromagnetic radiation immunity of a component, comprising the following steps:

[0055] S1: Set the range of electromagnetic frequency and electromagnetic amplitude, generate multiple particles within the range, each particle corresponds to a position feature composed of frequency and amplitude, set the frequency step size and amplitude step size as the initial velocity feature of all particles, initialize the optimal interference degree of each particle to 0, and the optimal position is the initial position feature.

[0056] In electromagnetic interference (EMI) testing, the sensitive response signals of electronic components are not uniformly distributed across an infinitely wide frequency and amplitude range, but rather concentrated within specific sensitive frequency bands and amplitude intervals. Furthermore, different electronic components exhibit significant differences in their sensitive response ranges. Therefore, by pre-setting the range of electromagnetic frequency and amplitude values, the test parameter space can be reasonably constrained and targeted, avoiding redundant testing under invalid or non-physically meaningful conditions, thereby improving the specificity and efficiency of interference testing.

[0057] Electromagnetic frequency and electromagnetic amplitude constitute a two-dimensional parameter space, and their respective value ranges define the boundaries of this parameter space, enabling efficient iterative updates of the particle swarm within a finite region in subsequent steps. The range of electromagnetic frequency values ​​is directly related to the coverage and effectiveness of interference testing; an excessively wide range increases redundant testing, while an excessively narrow range may miss sensitive frequency bands. The range of electromagnetic amplitude values ​​determines the strength of the interference signal; a weak interference signal may fail to trigger a sensitive response in the target component, while a strong interference signal may cause permanent damage to the tested component.

[0058] The electromagnetic frequency and amplitude values ​​covered in the electromagnetic immunity testing standards for electronic products are very extensive. In actual electromagnetic interference testing, the specific value ranges are usually set based on the operating characteristics of the target component, historical operating data, and relevant industry standards.

[0059] The core objective of electromagnetic interference testing is to identify the electromagnetically sensitive frequency bands and amplitudes of the tested components. To this end, this embodiment uses any combination of frequency-amplitude parameters as a particle within a predefined two-dimensional parameter space. Initial position and velocity characteristics are set for each particle, and the electromagnetically sensitive points of the target component are found through iterative updates of the particle swarm.

[0060] Furthermore, the logic for generating multiple particles within the set value range is as follows: the range of electromagnetic frequency values ​​is equally divided into multiple frequency intervals to obtain each frequency division point; the range of electromagnetic amplitude values ​​is equally divided into multiple amplitude intervals to obtain each amplitude division point; a two-dimensional orthogonal grid is constructed using all frequency division points and amplitude division points, where the frequency-amplitude coordinates corresponding to each node are determined as the frequency-amplitude parameters of a particle and used as the initial position feature of the particle.

[0061] Please see Figure 2 Taking the electromagnetic frequency range of 100-900MHz and the electromagnetic amplitude range of 5-35V / m as an example, this embodiment sets the frequency step size to 100MHz and the amplitude step size to 5V / m, thereby constructing a two-dimensional orthogonal grid and generating a total of 63 particles. The values ​​of the horizontal and vertical coordinates of each particle in the two-dimensional grid are respectively used as the electromagnetic frequency and electromagnetic amplitude of the signal it represents, and the coordinates of each particle are used as its initial position features.

[0062] This embodiment initializes all particle position feature parameters by equally dividing the intervals and constructing an orthogonal grid. The aim is to achieve a uniform distribution of the initial particle swarm in the search space, avoiding "local clustering" or "hollow regions" that may be caused by random initialization. The orthogonal structure ensures the decoupling of the frequency and amplitude variables, making it easier for subsequent algorithms to independently adjust the update behavior of each dimension. At the same time, this structure naturally supports systematic traversal and repeatability, meeting the requirements of standardized testing.

[0063] However, using particles to find sensitive samples in a two-dimensional parameter space, scanning electromagnetic frequencies and amplitudes one by one within a certain range in the initial stage would not only increase testing time but also place a huge computational burden on the system. Therefore, in addition to setting the initial position characteristics of each particle, it is also necessary to set an initial velocity characteristic for each particle to determine the parameter changes of each particle in the next iteration.

[0064] In this embodiment, frequency and amplitude parameters are set with separate frequency and amplitude step sizes. In the first round of interference testing, the vector formed by the set frequency and amplitude step sizes is used as the initial velocity characteristic of each particle. The frequency and amplitude step sizes essentially quantify the smallest unit of change in the two-dimensional parameters. Their specific values ​​determine the precision of finding electromagnetically sensitive points in the two-dimensional parameter space, are closely related to the operating frequency accuracy of the target component, and are constrained by the accuracy of the testing equipment, the resolution of the signal generator, and other relevant testing requirements in actual testing projects.

[0065] Setting the frequency step size and amplitude step size as the initial velocity characteristics of the particles constrains subsequent iterations so that all updates to particle position characteristics must be based on these two step sizes. This avoids parameter precision that cannot be implemented on the device, and reduces invalid small-amplitude updates while ensuring search accuracy, thus accelerating iteration convergence. Furthermore, the set frequency and amplitude step sizes are crucial for boundary correction in subsequent steps, ensuring that the updated calculation results match the response precision of the target component. This guarantees that all particles fall on the discrete parameter grid points that the device can execute.

[0066] In this embodiment, during the electromagnetic interference testing of target components using particles, the main criterion for selecting electromagnetically sensitive points is the particle interference level. Interference level is an important indicator for quantifying the interference effect of electromagnetic interference signals on target components; a higher value indicates a stronger interference effect and a more sensitive response from the component under the influence of the current electromagnetic frequency and amplitude.

[0067] To compare the interference performance of particles during the iteration process, this embodiment sets two parameters: optimal interference degree and optimal position, to record the historical optimal solution of each particle. Before the first round of testing begins, no interference data has been generated, so the value of the optimal interference degree of all particles is initialized to 0; and at this time, all particles have only one position data, which is the initial position feature marked by dividing the grid, so it is set as the optimal position of each particle.

[0068] This completes the initialization settings for all particle position characteristics, velocity characteristics, optimal interference degree, and optimal position, forming the initial particle swarm.

[0069] S2: Apply electromagnetic interference to the target components based on the positional characteristics of each particle and collect its feedback signal. Compare and analyze the preset standard response with the feedback signal, and obtain the current interference degree of each particle according to the interference degree calculation rule. For particles whose current interference degree is greater than the preset interference threshold, extract their positional characteristics and classify them into the high-sensitivity sample set.

[0070] Based on the constructed particle swarm, the system needs to transform the abstract algorithm into physically executable test instructions to drive subsequent testing processes. The core of this transformation process lies in mapping the particle position characteristics to control parameters of the electromagnetic signal generating device. In specific implementations, it is necessary to define the logical architecture and signal interaction mode for generating physical interference signals based on position characteristics, i.e., how to construct an electromagnetic interference field based on the particle position characteristics.

[0071] Furthermore, the method of applying electromagnetic interference to the target component is as follows: place the target component in an environment without electromagnetic interference, extract the position characteristics of the current particles, set the carrier frequency of the interference signal according to its frequency parameters, set the electromagnetic field strength of the interference signal according to its amplitude parameters, and apply the set electromagnetic interference signal to the target component.

[0072] The act of extracting particle position features is essentially a process of converting mathematical variables into executable physical control instructions. The frequency parameter is used to set the carrier frequency of the electromagnetic interference signal, and the amplitude parameter is used to set the electromagnetic field strength of the signal. Together, they constitute a radio frequency interference signal, which is applied to the target component.

[0073] After applying electromagnetic interference signals to the target components, it is also necessary to obtain the current interference level of each particle according to the preset standard response and interference degree calculation rules.

[0074] Furthermore, the preset standard response is the output signal of the target component in normal working condition under interference-free environment.

[0075] The preset standard response of a target component is prior data calibrated under standard operating conditions after passing factory inspection. It represents the output signal pattern generated by the target component under normal operation in a "pure" electromagnetic environment free from any external electromagnetic interference. In an electromagnetic interference testing system, calculating the interference degree of particles essentially means calculating the deviation between the feedback signal of the target component and the preset standard response under the current particle interference. Without this standard response, the system cannot determine whether the abnormal feedback of the component is caused by applied electromagnetic interference, a quality problem of the component itself, or environmental noise. Therefore, the preset standard response can eliminate background interference and ensure the accuracy of the test results. In addition, by collecting preset standard response data, the system can understand the performance boundary of the target component under ideal conditions. Once the feedback signal deviates, the amount of deviation directly reflects the degree of impact of interference on the component's function.

[0076] Furthermore, the specific method for calculating the current interference degree of the particle is as follows: calculate the normalized mean square error between the feedback signal and the standard response data, and use the result as the distortion degree of the target component output feedback signal under the current interference particle; set a multi-polarization judgment rule, and label the failure level of the obtained feedback signal according to the distortion degree, interruption status and logic error of the feedback signal; and perform weighted fusion of the distortion degree and failure level of the feedback signal to obtain the interference degree of the current particle.

[0077] In electromagnetic interference testing, abnormal responses of components manifest in multiple ways, including signal waveform distortion, communication or functional interruption, and output logic errors. Relying on any single indicator alone cannot comprehensively evaluate the degree of interference experienced by the components. Therefore, this embodiment employs a comprehensive evaluation standard combining distortion degree and failure level to calculate the interference level of particles.

[0078] Distortion, in particular, focuses on the quality degradation of analog signals or continuous outputs, reflecting the continuous impact of interference on signal fidelity. In this embodiment, the collected feedback signal is compared with the standard response data of the target component, and the normalized mean square error (NMSE) between the two is used as the distortion of the feedback signal.

[0079] The amplitude of the standard response signal can vary significantly depending on the components or testing conditions. NMSE normalizes the distortion by dividing the mean square error by the variance of the standard response signal, making the distortion a dimensionless relative ratio. This means that even with different signal baselines, the calculated distortion can be directly compared, avoiding misjudgments where a large absolute error under strong signals may have a relatively small impact, or where a small absolute error under weak signals may indicate severe distortion. Furthermore, the square term in the NMSE calculation significantly amplifies the contribution of larger deviations to the results. In electromagnetic interference assessment, occasional spikes, oscillations, or severe distortions are often key indicators of component sensitivity or potential faults. The square characteristic ensures that these severe distortion phenomena can be sensitively captured and reflected in the distortion value.

[0080] Within an observation window with N sampling points, for the acquired feedback signal sequence: and the corresponding standard response data sequence obtained: The formula for calculating the normalized mean square error is:

[0081]

[0082] This represents the signal value at the i-th sampling point in the standard response data sequence. This represents the signal value at the i-th sampling point in the feedback signal sequence. The resulting NMSE is the distortion degree of the feedback signal; the larger the NMSE value, the more severe the distortion.

[0083] In addition to distortion, this embodiment also introduces the interference degree of particles that jointly determine the failure level.

[0084] Failure levels focus on discrete errors in digital logic or functional states, reflecting the degree to which electromagnetic interference damages the functional integrity of target components. This embodiment, by setting multi-polarization judgment rules, assigns a failure level to the obtained feedback signal based on its distortion, interruption status, and logic errors.

[0085] The impact mechanisms of electromagnetic interference on electronic components are multidimensional and heterogeneous, and a single-dimensional assessment cannot fully capture their complex failure characteristics. For example, a communication module may experience severe waveform distortion under certain interference but still function with minimal effort, while under interference in another frequency band, the waveform distortion is slight but the communication link is interrupted. Multipolar judgment rules mean evaluating the feedback signals of components from multiple independent dimensions, each focusing on a specific failure mode. This enables a refined and structured diagnosis of the component's functional state, providing accurate and multidimensional input for subsequent fusion into a comprehensive interference level assessment, ensuring that the test evaluation results are both comprehensive and have clear engineering guidance significance.

[0086] This embodiment constructs a multi-polar judgment rule around three core dimensions, and divides the failure level of the feedback signal into 4 levels.

[0087] First, there's the distortion of the feedback signal. This dimension divides the continuous distortion value calculated based on the feedback signal into discrete level intervals to determine the severity of quantized waveform distortion. Its significance lies in providing a basic and comparable continuous-discrete mapping for interference effects, allowing gradual degradation of signal quality to be incorporated into the level system, reflecting the impact of interference on component performance and accuracy. The specific method for dividing the distortion intervals is as follows:

[0088] By setting thresholds for mild and severe distortion, the distortion of the feedback signal is divided into three ranges: slight distortion, moderate distortion, and severe distortion.

[0089] The severe distortion threshold should be set to reflect the critical point at which a component's function is severely damaged or about to fail. This embodiment sets the severe distortion threshold... The NMSE threshold is set as the value obtained through destructive testing, i.e., the NMSE value at which the control system misjudges due to sensor output signal error. In actual electromagnetic interference testing, the severe distortion threshold is usually set between 0.1 and 0.9, with the specific value depending on the safety requirements of the component's operating performance.

[0090] The mild distortion threshold is set based on the component's inherent noise level and tolerance range, and its calculation formula is as follows:

[0091]

[0092] in, This is the threshold for slight distortion. These are prior parameters calibrated based on the factory performance specifications of the target component. They characterize the allowable fluctuation range of the output signal of the component under interference-free conditions and are set according to the type and accuracy level of the target component. This represents the average noise energy calculated from multiple standard response data acquisitions in an interference-free environment. The formula for its calculation is:

[0093]

[0094] Where N represents the number of sampling points contained in the standard response data signal sequence. Let i be the signal value at the i-th sampling point. This represents the average value of the standard response data signal sequence. The slight distortion threshold is typically set to... to The specific values ​​vary depending on the safety requirements of the component's performance.

[0095] like The distortion state is judged as slight distortion; if and The distortion state is determined to be moderate distortion; if The distortion state is determined to be severe distortion.

[0096] Besides distortion, interruption is another important criterion. This dimension monitors whether the continuity of system functions or data flow is disrupted. Its judgment logic is to monitor the breakpoints in the function execution flow, aiming to capture interference effects that prevent components from completing their core timing functions on time and in sequence. Even if waveform distortion is small, delays, packet loss, or silence can directly lead to system-level failures. This dimension assesses the disruption of component functionality caused by interference.

[0097] Ideally, the feedback signal is a continuous time-series data. If an interruption occurs, a time interval will be generated for receiving the feedback signal. Determining whether an interruption occurs requires a comprehensive analysis of both the number of interrupt intervals and the duration of each interrupt interval. Therefore, this embodiment focuses on the number of interrupt intervals. Duration of interruption interval Quantity thresholds were set separately. With duration threshold Only when both the number and duration of interruption intervals are below the corresponding preset thresholds, i.e. and Only when the interruption occurs is it considered a minor interruption; feedback signals from other states are considered serious interruptions.

[0098] Finally, the judgment dimension of the multipolar judgment rule also includes logical errors. The judgment logic of this dimension is to verify whether the functional decision is correct or not, and output a binary judgment result M. If the value of M is 1, it means that there is interference in the feedback signal that can cause the component to produce incorrect output, incorrect state or illegal communication; if the value of M is 0, it means that there are no logical errors in the feedback signal.

[0099] Based on three judgment dimensions—distortion, interruption status, and logical error—the logic for determining the failure level in this embodiment is as follows:

[0100] First, assess the logic errors and interruptions in the feedback signal. If a logic error is found in the feedback signal or a serious interruption occurs, the failure level is determined to be 3.

[0101] For feedback signals without logical errors and with only minor interruptions, assess their distortion level. If the feedback signal is severely distorted, the failure level is determined to be 2; if it is moderately distorted, the failure level is determined to be 1; if it is slightly distorted, the failure level is determined to be 0.

[0102] The mathematical expression for the above rules is:

[0103]

[0104] Table 1 describes the distortion degree, interruption status, and the three dimensions of distortion degree. For feedback signals of electronic components, the determination of logical errors has the highest priority, followed by interruption status. Only when there are no logical errors or serious interruptions is the feedback signal valuable for evaluation. Based on this, the distortion degree of the feedback signal can be strongly correlated with its failure level; that is, the higher the distortion degree, the higher the failure level.

[0105] Table 1: Failure Level Assessment Table

[0106]

[0107] After calculating the distortion degree (NMSE) and failure level (L) of the feedback signal, the interference degree of the current particle on the target component is obtained through weighted fusion, using the following formula:

[0108]

[0109] in, These are weighting coefficients, pre-calibrated based on the working characteristics of the components and engineering experience, and can be adjusted to meet different testing standards or application scenarios. Values ​​typically range from 0.1 to 0.9, used to adjust the relative weights of signal distortion and functional failure in comprehensive interference assessment. This is particularly relevant for components requiring high signal fidelity (such as analog sensors). A larger value of around 0.6 to 0.9 can be used; for safety-critical components with high requirements for functional continuity (such as controllers). A smaller value of approximately 0.2 to 0.4 is acceptable. Those skilled in the art can calibrate the value according to the specific testing objectives and component characteristics.

[0110] The particle interference level I obtained through the above steps is a comprehensive sensitivity index. The larger the value of I, the more severe the overall interference to the function of the target component under the combination of electromagnetic frequency and amplitude represented by the current particle.

[0111] During interference testing, not all particle test results are of equal value. Only those test particles that have shown significant interference effects are considered high-sensitivity samples that contribute to finding the electromagnetically sensitive areas of the target component. Therefore, this embodiment constructs a high-sensitivity sample set to collect the parameters of these particles with high interference levels after each round of interference testing. The determination method involves setting an interference threshold. For all particles in the current test round whose interference level is greater than the interference threshold, i.e. The particles, whose positional characteristics are determined by combining electromagnetic frequency and electromagnetic amplitude parameters, are recorded in a high-sensitivity sample set, thus completing the collection of high-interference samples.

[0112] The interference threshold is used to identify test particles with significant interference effects, and its value typically ranges from 0.1 to 0.8. The specific value can be set according to the stringency requirements of the test. A lower interference threshold results in more comprehensive sample collection, but may include some particles with insignificant interference; a higher interference threshold results in more typical samples, but may miss some edge-sensitive points. In this embodiment, Configured to filter out failure levels The minimum interference value of all particles is used to ensure that all abnormal samples are included in the high-sensitivity sample set.

[0113] S3: For each particle, if the current disturbance degree is greater than the optimal disturbance degree, update its optimal disturbance degree to the current disturbance degree and its optimal position to the current position feature. Select the optimal disturbance degree with the largest value among all particles as the global disturbance degree and its corresponding position feature as the global optimal position.

[0114] This step serves as a crucial bridge between individual and global information updates in the interference testing system. After calculating the interference degree for the current round, the historical optimal state of each particle is updated, and the global optimal position in the current test round is determined. For each particle, the system compares its current interference degree with the historically recorded optimal interference degree. If the current interference degree is greater, it indicates that the electromagnetic frequency and amplitude of the current position feature have a stronger interference effect on the target component during the particle's iteration process. Therefore, the optimal interference degree needs to be updated to the current value, and the current position feature is taken as the particle's optimal position. After the update, each particle can remember the most sensitive electromagnetic parameter combination it discovers during the iteration process, forming an individual memory. After all particles have updated their individual optimal values, the system selects the optimal interference degree with the largest value as the global interference degree, and its corresponding position feature as the global optimal position. The global optimal position represents the electromagnetic parameter point that has found the strongest interference to the target component during the entire particle swarm's iteration process so far.

[0115] After completing the individual particle update and the global update for the current test round, the individual optimal position and global optimal position output in this step will be directly used for updating the particle velocity and position features in subsequent steps.

[0116] S4: For each particle, calculate the individual deviation between its current position feature and its optimal position, as well as the global deviation from the global optimal position. Weight the obtained deviation with the current velocity feature and perform boundary correction on the weighted result to obtain a new velocity feature. Add the current position feature to the new velocity feature to obtain a new position feature. Maintain the new position feature within the range of values ​​through boundary correction to construct a new particle swarm.

[0117] This step calculates the individual and global deviations of all particles, reflecting the differences between each particle's current state and its own historical best state and the group's best state, respectively. The two deviations are then weighted and combined with the current velocity characteristic to determine the particle's step size for the next iteration. Individual deviations guide particles towards their historical best direction, reflecting the influence of individual experience; global deviations guide particles towards the group's best direction, reflecting the influence of group cooperation; and the current velocity characteristic reflects the particle's inertia from the previous moment. These three factors together determine the particle's motion trend in the next iteration.

[0118] Furthermore, the calculation method for the individual deviation and the global deviation is as follows: First, as described in S3, based on the current interference degree of each particle, the optimal interference degree and optimal position of all particles in the particle swarm are updated, and the global interference degree and global optimal position of the current test round are determined.

[0119] For each particle, calculate the difference between its optimal position and current position characteristics in terms of frequency and amplitude components. The results are the individual frequency difference and the individual amplitude difference, respectively. The vector formed by these two differences is the individual deviation of the particle.

[0120] The essence of individual bias is to quantify the parameter difference between the current position of a particle and its historical best position, and to express this difference in vector form. For any particle, its current position feature is denoted as... Its updated optimal position is denoted as Calculate the individual frequency difference separately. Individual amplitude difference The formula is:

[0121]

[0122]

[0123] The individual deviation of this particle is denoted as ;like If the frequency value increases in the next iteration, then the particle tends to increase its frequency value. If the particle's current position is far from its historical optimal position, then it tends to increase its amplitude value in the next iteration. The greater the individual deviation, the further the particle's current position is from its historical optimal position, and the greater its adjustment amplitude in the next iteration. If the particle's current position is its historical optimal position, then... and If the obtained individual deviation is a zero vector, the particle will move mainly based on its inertia, i.e. its current velocity characteristics, and its global deviation in the next iteration.

[0124] For each particle, calculate the difference between the optimal position in the current test round and the particle's current position features in terms of frequency and amplitude components. The results are the global frequency difference and the global amplitude difference, respectively, and the vector formed by these two differences is the global deviation of the particle.

[0125] The essence of global bias is to quantify the parameter difference between the current particle's position and the optimal position of all particles in the current test round, and express this difference in vector form as a guide for particles to move towards the group's optimal direction. Global bias guides all particles to gather towards the current optimal region, accelerating the convergence process, especially in the early iteration stage where it can quickly locate potential sensitive regions.

[0126] Let the global optimal position of the current test round be denoted as For any particle's current position characteristics Calculate the global frequency difference between each position and the global optimal position. and global amplitude difference :

[0127]

[0128]

[0129] The global deviation of this particle is denoted as . ;like If the particle moves in the direction of increasing frequency in the next iteration, then the particle tends to move in the direction of increasing frequency; if If the global deviation is greater, the particle tends to move in the direction of increasing amplitude in the next iteration. The greater the global deviation, the farther the particle's current position is from the optimal position of the group. Therefore, the particle will adjust its amplitude in the next iteration and the group guidance effect will be stronger.

[0130] Furthermore, the new velocity characteristic is determined jointly based on the particle's current velocity characteristic, individual deviation, and global deviation. The specific calculation process is as follows:

[0131] Three key control parameters are set respectively, namely the inertia weighting coefficient. Individual adjustment coefficient and global adjustment coefficient The parameters collectively constitute the adjustment mechanism for particle velocity updates. These three coefficients correspond to three behavioral components in the particle motion process: maintaining motion inertia, individual experience guidance, and group experience guidance. Appropriately setting these coefficients can balance the algorithm's exploration and development capabilities, directly affecting the efficiency and convergence performance of the iteration. The individual adjustment coefficient focuses on local fine-tuning updates, the global adjustment coefficient focuses on global directional guidance, and the inertia weight coefficient maintains the continuity and stability of the iteration. By adjusting the magnitudes of these three coefficients, the system can flexibly control the particle's movement strategy in the parameter space, adapting to the electromagnetic sensitivity characteristics of different components.

[0132] For inertia weight coefficient Its main function is to control the degree to which the particle's current velocity characteristics are retained in the next iteration, typically taking a value between 0.4 and 0.9. When When the value is large, particles tend to maintain their original direction of motion, enhancing their ability to update across the entire parameter space; when... When the particle size is small, it is easier for it to change direction, which is beneficial for fine-grained updates and convergence within a local area.

[0133] For individual adjustment coefficients Its main function is to control the tendency of particles to move toward their historical best position, and it is usually taken between 1.5 and 2.5. The larger the value, the more the particle values ​​individual experience and tends to search around already discovered favorable areas; The smaller the value, the less individual experience contributes to the iterative process, which may lead to a decrease in search efficiency.

[0134] For global adjustment coefficients Its main function is to control the tendency of particles to move toward the optimal position of the group, and it is usually taken between 1.5 and 2.5. The larger the value, the more likely the particles are to gather in the optimal region of the current population, thus accelerating convergence; The smaller the value, the less the group cooperates, which may lead to a slower convergence speed.

[0135] In practical electromagnetic interference testing, control coefficients for specific types of components are typically calibrated based on historical test data, and an adaptive adjustment mechanism is introduced to dynamically adjust the control coefficients according to particle swarm diversity or convergence status. Furthermore, in actual systems, the setting of control coefficients should be constrained by the step size accuracy of the signal generator and the test time to ensure that particle movement is neither too violent nor too sluggish, achieving efficient and stable iteration.

[0136] For any particle, the particle's current velocity characteristic, i.e. , with inertia weight coefficient Multiply them to obtain the frequency components of their current velocity characteristics. Amplitude component of the current velocity characteristic The formula is:

[0137]

[0138]

[0139] Individual deviation of the particle With individual adjustment coefficient Multiplying them together yields the individual frequency adjustment components. Individual amplitude adjustment component The formula is:

[0140]

[0141]

[0142] global deviation of the particle With global adjustment coefficient Multiplying them yields the global frequency adjustment component and the global amplitude adjustment component, respectively.

[0143]

[0144]

[0145] Based on the above calculation results, the frequency components of the current velocity characteristics will be... Individual frequency adjustment components and global frequency adjustment components The sum is used as the frequency change. ,Right now:

[0146]

[0147] The amplitude component of the current velocity characteristic Individual amplitude adjustment components and global amplitude adjustment components The sum is used as the magnitude of the change. ,Right now:

[0148]

[0149] From frequency change With amplitude change Constitutes new velocity characteristics ,Right now .

[0150] During particle swarm iteration, after each particle is updated, its new velocity and position characteristics may exceed the preset range of electromagnetic frequency and amplitude values, or may not meet the minimum step accuracy requirements of the test equipment. Therefore, boundary correction must be performed on the updated particles to ensure that all parameter combinations are within the physically executable range and meet the engineering constraints of the test system.

[0151] Boundary correction requires addressing two key aspects simultaneously: the discretization correction of velocity characteristics and the boundary constraints of position characteristics. Since the three control parameters involved in the calculation are typically set as decimals in practical engineering, the calculated frequency change... With amplitude change The calculated results may not match the frequency or amplitude step resolution, causing the physical device to be unable to recognize them. Therefore, the results need to be adjusted to a form that the signal generator can recognize. Furthermore, since the new position feature is obtained by adding the current position feature to the new velocity feature, if the current position feature is already at the boundary of the parameter space, the sum of the current and new velocity features will exceed the preset range, also causing the system to be unable to process it. Therefore, the updated position feature needs to be constrained within the preset range of electromagnetic frequency and electromagnetic amplitude values.

[0152] Furthermore, the logic for obtaining the new velocity features and the new position features is as follows:

[0153] The calculated frequency change and amplitude change are corrected to integer multiples of the frequency step size and the amplitude step size, respectively. The absolute value of the difference between the corrected frequency change and its original value is less than the frequency step size, and the absolute value of the difference between the corrected amplitude change and its original value is less than the amplitude step size. The corrected frequency change and amplitude change constitute a new velocity characteristic.

[0154] For the new velocity characteristics The frequency change was respectively With amplitude change Corrected to frequency step size With amplitude step size Integer multiples of the corrected frequency change Its original value The difference is less than the frequency step size And the magnitude of the change after correction Its original value The difference is less than the amplitude step size ,Right now:

[0155]

[0156]

[0157] in, and These are the adjustment factors for the frequency change and the amplitude change, respectively. and It can be a positive or negative number, indicating an adjustment in the direction of increase or decrease.

[0158] For the adjusted and It must meet the following conditions:

[0159]

[0160]

[0161] This constraint is to ensure that the corrected value is as close as possible to the original value, avoiding systematic errors introduced by discretization. The final corrected velocity characteristics are then obtained. .

[0162] The current position feature is added to the corrected new velocity feature. If the resulting frequency value exceeds the range of electromagnetic frequency of the interference signal, it is corrected to the boundary value of the electromagnetic frequency range. If the resulting amplitude value exceeds the range of electromagnetic amplitude of the interference signal, it is corrected to the boundary value of the electromagnetic amplitude range. This yields the new position feature after boundary correction.

[0163] For any particle, add the current position feature to the corrected new velocity feature. If the resulting frequency value exceeds the range of electromagnetic frequency of the interference signal, then correct it to the boundary value of the electromagnetic frequency range. If the resulting amplitude value exceeds the range of electromagnetic amplitude of the interference signal, then correct it to the boundary value of the electromagnetic amplitude range.

[0164] For the current position characteristics of any particle its frequency parameters With the adjustment Add them together to get a new frequency value. ; its amplitude parameter With the adjustment Add them together to get a new amplitude value. :

[0165]

[0166]

[0167] Within the preset electromagnetic frequency range With respect to the range of electromagnetic amplitude values Inside, regarding the results and Make corrections, and the correction rules are as follows:

[0168]

[0169]

[0170] Finally, the corrected new positional features are obtained. .

[0171] Through the above steps, the velocity and position characteristics of each particle in the particle swarm are updated, and a new particle swarm is finally generated. The next round of electromagnetic interference testing is then conducted based on the new particle swarm.

[0172] S5: Repeat S2 to S4. When the global interference degree meets the preset convergence condition and reaches the preset round limit, the iteration ends. Perform cluster analysis on all positional features in the high-sensitivity sample set. The value boundaries of electromagnetic frequency and electromagnetic amplitude in each cluster are the electromagnetic interference sensitive range of the target component.

[0173] This step is the iterative convergence and result output stage. The core task is to end the particle swarm update by setting dual stopping conditions, namely, the convergence of global interference degree and the reaching of the upper limit of iteration rounds, and to extract the sensitive range boundary of the target component in the actual electromagnetic interference environment from the collected high-sensitivity samples using cluster analysis methods.

[0174] Using global perturbation as one of the convergence conditions ensures that the system stops engaging in meaningless searches after finding sufficiently strong perturbation points. This embodiment sets the change in global perturbation as the convergence criterion; that is, the absolute value of the difference between the current global perturbation and the previous round's global perturbation is less than a preset convergence threshold. When the change in global perturbation falls below the preset convergence threshold, it indicates that the particle swarm can no longer find more sensitive perturbation points, and further iterations will yield even lower returns. Furthermore, setting an upper limit on the number of iterations prevents infinite loops caused by improper convergence threshold settings or the algorithm getting stuck in local optima, ensuring the testing process is completed within a controllable timeframe.

[0175] Please see Figure 3 In this embodiment, the upper limit of the number of iterations is set to 50, and the preset convergence threshold is 0.05. During the particle swarm iteration process, the global disturbance degree increases with the increase of the number of iterations. After 32 iterations, the global disturbance degree no longer increases significantly and tends to stabilize. Moreover, starting from the 32nd iteration, the absolute value of the difference between the global disturbance degree of each iteration and the global disturbance degree of the previous iteration is less than the convergence threshold of 0.05 set in this embodiment. Therefore, it can be determined that the iteration has converged at this time. After 50 iterations, the generation of new particle swarms is stopped.

[0176] The highly sensitive sample set collected during particle swarm optimization contains multiple sample points that trigger component malfunctions under specific combinations of frequency and amplitude. However, these sample points may be scattered, isolated, or even contain noise. Directly using these discrete sample points to describe electromagnetically sensitive regions may lead to blurred value boundaries and make it difficult to form continuous and reliable intervals.

[0177] Please see Figure 4 For particles collected from highly sensitive samples, a thermal distribution map is plotted based on their location characteristics and interference levels, revealing the electromagnetic interference sensitivity distribution of target components in the two-dimensional frequency-amplitude parameter space. In the map, colors transitioning from cool to warm visually represent the increasing interference level. Warm-colored concentrated areas correspond to specific frequency-amplitude parameter combinations where components are most prone to functional abnormalities or performance degradation. The thermal map reveals the spatial continuity and boundary transition characteristics of electromagnetic sensitivity intensity and can be used to determine the accuracy of subsequent clustering results.

[0178] This embodiment uses cluster analysis to group discrete high-sensitivity sample points according to their distribution in the parameter space, with each cluster corresponding to a potential sensitive region. By statistically analyzing the distribution characteristics of sample frequency and amplitude within each cluster, the boundary range of the electromagnetically sensitive region represented by each cluster can be calculated, thereby identifying and describing each independent electromagnetically sensitive region.

[0179] Furthermore, the logic for determining the electromagnetic interference sensitive range of the target component using cluster analysis is as follows:

[0180] Each location feature in the high-sensitivity sample set is treated as an independent cluster. The inter-cluster distance between any two clusters is calculated as follows: for any two clusters, the samples contained in them are mapped pairwise, all sample combinations are traversed, the Euclidean distance between each pair of samples is calculated, and the maximum value of the obtained Euclidean distance is selected as the inter-cluster distance between the two clusters. The dissimilarity matrix is ​​constructed using all the calculated inter-cluster distances.

[0181] For a high-sensitivity sample set H containing m samples, denoted as... Each sample represents a location feature vector, i.e. , The electromagnetic frequency representing the location characteristic. This represents the electromagnetic amplitude characteristic of that location. Treating each sample as an independent cluster, the set of all highly sensitive samples is represented as:

[0182]

[0183] in This indicates that before clustering begins, the number of clusters is the same as the number of sample points, and the k-th cluster... It contains only one sample, namely the k-th sample. .

[0184] For any two clusters and The inter-cluster distance Defined as:

[0185]

[0186] in, For the sample With sample The Euclidean distance between them is:

[0187]

[0188] Will and Pairwise mapping of all samples in the two clusters is performed. All sample pairs are iterated through, and the Euclidean distance between each pair is calculated. The maximum value is taken as the distance between the pairs. and The inter-cluster distance, i.e.:

[0189]

[0190] Construct a dissimilarity matrix W using the pairwise distances between all pairs of clusters in cluster set C. Then W is expressed as:

[0191]

[0192] The resulting dissimilarity matrix W is iteratively processed as follows: the two clusters with the smallest inter-cluster distance among all current cluster pairs are searched in a loop and merged into a new cluster; the dissimilarity matrix is ​​updated, the distance between the new cluster and other clusters is recalculated, and the above merging process is repeated until the minimum inter-cluster distance in the cluster pair is lower than the preset convergence threshold and the number of obtained clusters meets the preset range of clustering number, then the merging is stopped and the partitioning result is obtained.

[0193] Set an iteration counter z, and denote the dissimilarity matrix obtained in the z-th iteration as follows: In each iteration, from the current dissimilarity matrix... Find the smallest off-diagonal element in the given list, i.e.:

[0194]

[0195] Record the two clusters corresponding to the minimum inter-cluster distance. and And merge them into a new cluster, denoted as:

[0196]

[0197] Remove from cluster set C and , and add .

[0198] Repeat the above merging and updating process. When the minimum inter-cluster distance in the dissimilarity matrix is ​​less than the preset convergence threshold and the preset number of clusters is reached, stop the iteration and output the result of the current cluster division.

[0199] For each cluster, calculate the geometric center of the positional features of all samples within the cluster. Using the obtained geometric center as a reference, calculate the standard deviation of the sample points within the cluster in terms of frequency and amplitude parameters. Then, perform a weighted expansion of the standard deviation of the frequency parameter and the standard deviation of the amplitude parameter to obtain the boundary of the range of electromagnetic frequency and electromagnetic amplitude values ​​corresponding to the cluster.

[0200] For the k-th cluster For the u samples contained in the cluster, first calculate the set center of the positional features of all samples within the cluster:

[0201]

[0202]

[0203] Then cluster The geometric center is represented as ;

[0204] Calculate the standard deviations of the sample points within this cluster for both frequency and amplitude parameters:

[0205]

[0206]

[0207] By weighting and expanding the standard deviations of the frequency and amplitude parameters, the range boundaries of the electromagnetic frequency and amplitude values ​​corresponding to the cluster are obtained. For the cluster... The range of electromagnetic frequencies it represents is as follows:

[0208]

[0209]

[0210] cluster The range of values ​​for the electromagnetic amplitude represented is as follows:

[0211]

[0212]

[0213] in, As the boundary expansion factor, its value determines the range of values ​​that expands in both positive and negative directions centered on the standard deviation; and These are the weighted expansion coefficients for frequency and amplitude, respectively, which determine the influence weights of electromagnetic frequency and electromagnetic amplitude. , as well as All three are constants greater than 0, and their specific values ​​can be set and adjusted according to the working characteristics of different types of electronic components to meet the testing requirements of different working conditions.

[0214] Please see Figure 5 The present invention also provides a testing system for the electromagnetic radiation immunity of components, wherein the testing system is used to perform the above-described testing method for the electromagnetic radiation immunity of components, including:

[0215] Initialization module: Used to set the range of electromagnetic frequency and electromagnetic amplitude, generate multiple particles within the range, each particle corresponds to a position feature composed of frequency and amplitude, set the frequency step size and amplitude step size as the initial velocity feature of all particles, initialize the optimal disturbance degree of each particle to 0, and the optimal position is the initial position feature.

[0216] Signal interference module: It is used to apply electromagnetic interference to target components based on the position characteristics of each particle and collect its feedback signal. It compares and analyzes the preset standard response with the feedback signal, and obtains the current interference degree of each particle according to the interference degree calculation rules. For particles whose current interference degree is greater than the preset interference threshold, its position characteristics are extracted and it is included in the high-sensitivity sample set.

[0217] Parameter update module: For each particle, if the current disturbance degree is greater than the optimal disturbance degree, update the optimal disturbance degree to the current disturbance degree and the optimal position to the current position feature. Select the optimal disturbance degree with the largest value among all particles as the global disturbance degree and its corresponding position feature as the global optimal position.

[0218] Particle Update Module: For each particle, calculate the individual deviation between its current position feature and its optimal position, as well as the global deviation from the global optimal position. The obtained deviation is weighted and fused with the current velocity feature. After boundary correction of the weighted result, a new velocity feature is obtained. The current position feature is added to the new velocity feature to obtain a new position feature. The boundary correction keeps the new position feature within the range of values ​​and constructs a new particle swarm.

[0219] Clustering analysis module: used from the repetitive signal interference module to the particle update module. When the global interference degree meets the preset convergence condition and reaches the preset round limit, the iteration ends. Clustering analysis is performed on all positional features in the high-sensitivity sample set. The value boundaries of electromagnetic frequency and electromagnetic amplitude in each cluster are the electromagnetic interference sensitive range of the target component.

[0220] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0221] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0222] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0223] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for testing the electromagnetic radiation immunity of a component, characterized in that, The specific steps include: S1: Set the range of electromagnetic frequency and electromagnetic amplitude, generate multiple particles within the range, each particle corresponds to a position feature composed of frequency and amplitude, set the frequency step size and amplitude step size as the initial velocity feature of all particles, initialize the optimal interference degree of each particle to 0, and the optimal position is the initial position feature. S2: Apply electromagnetic interference to the target component based on the position characteristics of each particle and collect its feedback signal. Compare and analyze the preset standard response with the feedback signal, and obtain the current interference degree of each particle according to the interference degree calculation rule. For particles whose current interference degree is greater than the preset interference threshold, extract their position characteristics and classify them into the high-sensitivity sample set. S3: For each particle, if the current disturbance degree is greater than the optimal disturbance degree, update its optimal disturbance degree to the current disturbance degree, update its optimal position to the current position feature, select the optimal disturbance degree with the largest value among all particles as the global disturbance degree, and its corresponding position feature as the global optimal position. S4: For each particle, calculate the individual deviation between its current position feature and its optimal position, as well as the global deviation from the global optimal position. Weigh and fuse the obtained deviation with the current velocity feature. After boundary correction of the weighted result, we obtain a new velocity feature. We add the current position feature and the new velocity feature to obtain a new position feature. We maintain the new position feature within the range of values ​​through boundary correction and construct a new particle swarm. S5: Repeat S2 to S4. When the global interference degree meets the preset convergence condition and reaches the preset round limit, the iteration ends. Perform cluster analysis on all positional features in the high-sensitivity sample set. The value boundaries of electromagnetic frequency and electromagnetic amplitude in each cluster are the electromagnetic interference sensitive range of the target component.

2. The method for testing the electromagnetic radiation immunity of a component according to claim 1, characterized in that, The logic for generating multiple particles within the set value range is as follows: the range of electromagnetic frequency values ​​is equally divided into multiple frequency intervals to obtain each frequency division point; the range of electromagnetic amplitude values ​​is equally divided into multiple amplitude intervals to obtain each amplitude division point; a two-dimensional orthogonal grid is constructed using all frequency division points and amplitude division points, where the frequency-amplitude coordinates corresponding to each node are determined as the frequency-amplitude parameters of a particle and used as the initial position feature of the particle.

3. The method for testing the electromagnetic radiation immunity of a component according to claim 2, characterized in that, The method of applying electromagnetic interference to the target component is as follows: place the target component in an environment without electromagnetic interference, extract the position characteristics of the current particles, set the carrier frequency of the interference signal according to its frequency parameters, set the electromagnetic field strength of the interference signal according to its amplitude parameters, and apply the set electromagnetic interference signal to the target component.

4. The method for testing the electromagnetic radiation immunity of a component according to claim 3, characterized in that, The preset standard response is the output signal of the target component in normal working condition under interference-free environment; The specific method for calculating the current interference level of the particles is as follows: calculate the normalized mean square error between the feedback signal and the standard response data, and use the result as the distortion level of the target component output feedback signal under the current interference particles; set multi-polarization judgment rules, and calibrate the failure level of the obtained feedback signal according to the distortion level, interruption status and logic error of the feedback signal. The distortion and failure level of the feedback signal are weighted and fused to obtain the interference level of the current particle.

5. The method for testing the electromagnetic radiation immunity of a component according to claim 4, characterized in that, The calculation method for individual deviation and global deviation is as follows: based on the current interference degree of each particle, the optimal interference degree and optimal position of all particles in the particle swarm are updated, and the global interference degree and global optimal position of the current test round are determined. For each particle, calculate the difference between its optimal position and current position features in terms of frequency and amplitude components. The results are individual frequency difference and individual amplitude difference, respectively. The vector formed by these two differences is the individual deviation of the particle. For each particle, calculate the difference between the optimal position in the current test round and the particle's current position features in terms of frequency and amplitude components. The results are the global frequency difference and the global amplitude difference, respectively, and the vector formed by these two differences is the global deviation of the particle.

6. The method for testing the electromagnetic radiation immunity of a component according to claim 5, characterized in that, The new velocity characteristics are determined based on the particle's current velocity characteristics, individual deviations, and global deviations. The specific calculation process includes: Set the inertia weight coefficient, individual adjustment coefficient, and global adjustment coefficient respectively; Multiply the current velocity feature by the inertia weighting coefficient to obtain the frequency component and amplitude component of the current velocity feature, respectively. Multiplying the individual deviation by the individual adjustment coefficient yields the individual frequency adjustment component and the individual amplitude adjustment component, respectively. Multiplying the global deviation by the global adjustment coefficient yields the global frequency adjustment component and the global amplitude adjustment component, respectively. The frequency component, individual frequency adjustment component, and global frequency adjustment component of the current velocity feature are added together, and the result is taken as the frequency change; the amplitude component, individual amplitude adjustment component, and global amplitude adjustment component of the current velocity feature are added together, and the result is taken as the amplitude change. A new velocity characteristic is obtained by performing boundary correction on the vector formed by the obtained frequency change and amplitude change.

7. The method for testing the electromagnetic radiation immunity of a component according to claim 6, characterized in that, The logic for obtaining the new velocity and position features is as follows: The calculated frequency change and amplitude change are corrected to integer multiples of the frequency step size and the amplitude step size, respectively. The absolute value of the difference between the corrected frequency change and its original value is less than the frequency step size, and the absolute value of the difference between the corrected amplitude change and its original value is less than the amplitude step size. The corrected frequency change and amplitude change constitute a new velocity characteristic. The current position feature is added to the corrected new velocity feature. If the resulting frequency value exceeds the range of electromagnetic frequency of the interference signal, it is corrected to the boundary value of the electromagnetic frequency range. If the resulting amplitude value exceeds the range of electromagnetic amplitude of the interference signal, it is corrected to the boundary value of the electromagnetic amplitude range. This yields the new position feature after boundary correction.

8. The method for testing the electromagnetic radiation immunity of a component according to claim 7, characterized in that, The logic for determining the electromagnetic interference sensitive range of a target component using cluster analysis is as follows: Treat each location feature in the high-sensitivity sample set as an independent cluster; calculate the inter-cluster distance between any two clusters by mapping the samples contained in any two clusters pairwise, traversing all sample combinations, calculating the Euclidean distance between each pair of samples, and selecting the maximum value of the obtained Euclidean distance as the inter-cluster distance between the two clusters; construct the dissimilarity matrix using all the calculated inter-cluster distances. Iterative processing of the dissimilarity matrix: Locate the two clusters with the smallest inter-cluster distance among all current cluster pairs and merge them into a new cluster; Update the dissimilarity matrix, recalculate the distance between the new cluster and other clusters, and repeat the above merging process until the minimum inter-cluster distance in the cluster pair is lower than the preset convergence threshold and the number of clusters obtained meets the value range of the preset number of clusters. Stop merging and obtain the partitioning result. For each cluster, calculate the geometric center of the positional features of all samples within the cluster. Using the obtained geometric center as a reference, calculate the standard deviation of the sample points within the cluster in terms of frequency and amplitude parameters. Weight the standard deviation of the frequency parameter and the standard deviation of the amplitude parameter to obtain the boundary of the range of electromagnetic frequency and electromagnetic amplitude values ​​corresponding to the cluster. By integrating the ranges of all cluster electromagnetic frequencies and amplitudes, the electromagnetic interference sensitive range of the target component is formed.

9. A testing system for the electromagnetic radiation immunity of a component, characterized in that, The electromagnetic radiation immunity test system for the component is used to perform the electromagnetic radiation immunity test method for the component according to any one of claims 1-8, including: Initialization module: Used to set the range of electromagnetic frequency and electromagnetic amplitude, generate multiple particles within the range, each particle corresponds to a position feature composed of frequency and amplitude, set the frequency step size and amplitude step size as the initial velocity feature of all particles, initialize the optimal disturbance degree of each particle to 0, and the optimal position is the initial position feature. Signal interference module: It is used to apply electromagnetic interference to target components based on the position characteristics of each particle and collect its feedback signal. It compares and analyzes the preset standard response with the feedback signal, and obtains the current interference degree of each particle according to the interference degree calculation rules. For particles whose current interference degree is greater than the preset interference threshold, its position characteristics are extracted and it is included in the high-sensitivity sample set. Parameter update module: For each particle, if the current disturbance degree is greater than the optimal disturbance degree, update the optimal disturbance degree to the current disturbance degree and the optimal position to the current position feature. Select the optimal disturbance degree with the largest value among all particles as the global disturbance degree and its corresponding position feature as the global optimal position. Particle Update Module: For each particle, calculate the individual deviation between its current position feature and its optimal position, as well as the global deviation from the global optimal position. The obtained deviation is weighted and fused with the current velocity feature. After boundary correction of the weighted result, a new velocity feature is obtained. The current position feature is added to the new velocity feature to obtain a new position feature. The boundary correction keeps the new position feature within the range of values ​​and constructs a new particle swarm. Clustering analysis module: used from the repetitive signal interference module to the particle update module. When the global interference degree meets the preset convergence condition and reaches the preset round limit, the iteration ends. Clustering analysis is performed on all positional features in the high-sensitivity sample set. The value boundaries of electromagnetic frequency and electromagnetic amplitude in each cluster are the electromagnetic interference sensitive range of the target component.

Images