A method and system for quality testing of shielding materials

Abstract

This invention belongs to the field of quality testing technology, specifically a method and system for quality testing of shielding materials. The method includes collecting and extracting shielding effectiveness difference characteristics, physicochemical property difference characteristics, and a dynamic influence index of operating load. Based on these differences, a shielding material feature differentiation model is constructed. The impact of the dynamic influence index of operating load on material performance degradation is analyzed to obtain performance degradation response data. Interference signals in the testing scenario are dynamically identified to obtain interference identification results. Based on these results, a shielding effectiveness calibration algorithm is determined to calculate basic shielding effectiveness data. Finally, the current material quality status is assessed using the performance degradation response data to obtain quality testing data. This invention enables dynamic prediction of material performance, improving testing accuracy and reliability.

Description

Technical Field

[0001] This invention belongs to the field of quality testing technology, specifically a method and system for quality testing of shielding materials. Background Technology

[0002] In fields such as electronic communications, precision manufacturing, and national defense, the electromagnetic shielding performance of shielding materials directly determines the operational stability and information security of equipment, making precise quality testing crucial.

[0003] Current methods for testing the quality of shielding materials primarily focus on testing a single electromagnetic shielding effectiveness index, neglecting the impact of the material's physicochemical properties on long-term service stability, and failing to establish a correlation model between material type and performance degradation. Furthermore, testing scenarios commonly involve complex interference signals such as power frequency interference, industrial electromagnetic interference, and radio frequency signal interference. Traditional testing methods often employ uniform filtering techniques, which struggle to specifically eliminate the effects of different types of interference, leading to distorted shielding effectiveness test data that fails to reflect the material's true performance. In addition, during actual service, temperature and humidity fluctuations, mechanical vibrations, and electromagnetic radiation accelerate the performance degradation of shielding materials. However, current technologies lack the ability to quantitatively analyze the dynamic impact of these loads, making it impossible to accurately assess the correlation between operating conditions and performance degradation. Consequently, quality test results only reflect the material's instantaneous performance and are insufficient to predict its overall lifecycle quality trends.

[0004] The aforementioned problems mean that existing testing methods cannot meet the high-precision, full-cycle, and scientific requirements of high-end manufacturing for the quality testing of shielding materials. Therefore, it is urgent to develop a quality testing method and system for shielding materials. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention proposes a method and system for quality testing of shielding materials. This invention primarily addresses the problem of inaccurate quality testing of shielding materials caused by interference signals and material performance attenuation.

[0006] The technical solution adopted by this invention to solve its technical problem is: a quality testing method for shielding materials provided by this invention, comprising: Electromagnetic shielding characteristic data and physicochemical property characteristic data of different types of shielding materials were collected. The shielding effectiveness difference characteristics were extracted from the electromagnetic shielding characteristic data, and the physicochemical property difference characteristics were extracted from the physicochemical property characteristic data.

[0007] Historical application data of shielding materials were collected, and the dynamic influence index of working condition load was extracted from them. Based on the differences in shielding effectiveness and physical and chemical properties, a characteristic differentiation model of shielding materials was constructed.

[0008] Based on the characteristic differentiation model of shielding materials, the influence of the dynamic influence index of working load on the performance degradation of materials is analyzed to obtain performance degradation response data.

[0009] Interference signals in the detection scenario are dynamically identified to obtain interference identification results. Based on the interference identification results, a shielding effectiveness calibration algorithm is determined to calculate basic shielding effectiveness data. Combined with performance decay response data, the current material quality status is detected to obtain quality detection data.

[0010] This invention provides a method for quality testing of shielding materials, comprising: The steps for extracting the characteristics of shielding effectiveness differences include: We collected electric field attenuation values, magnetic field attenuation values, and electromagnetic wave reflection coefficients of different types of shielding materials at different frequency bands, and organized them according to material type, test frequency band, and test index to obtain electromagnetic shielding characteristic data.

[0011] Based on the application scenarios of shielding materials, the entire frequency band is divided into three intervals: low frequency band, mid frequency band, and high frequency band.

[0012] The minimum and average shielding effectiveness values ​​of each shielding material in the high-frequency band are calculated as high-frequency band threshold features.

[0013] The maximum magnetic field attenuation rate and the standard deviation of attenuation rate fluctuation for each shielding material in the low-frequency band are extracted as the performance stability characteristics in the low-frequency band.

[0014] Calculate the coefficient of variation of shielding effectiveness across the entire frequency band, and use the frequency range corresponding to the inflection point of shielding effectiveness changing with frequency as the effectiveness fluctuation coefficient.

[0015] The peak value and corresponding frequency point of the electromagnetic wave reflection coefficient in each frequency band are calculated as the characteristics of the electromagnetic energy reflection coefficient.

[0016] The shielding effectiveness difference characteristics are obtained by integrating the high-frequency threshold characteristics, low-frequency performance stability characteristics, effectiveness fluctuation coefficient and electromagnetic energy reflection coefficient characteristics.

[0017] The present invention provides a quality testing method for shielding materials, the steps of which include extracting differences in physicochemical properties: Thickness, coating adhesion level, tensile strength, elongation at break and abrasion resistance of different types of shielding materials were collected as physicochemical property data.

[0018] The Z-Score standardization method was used to normalize the physicochemical indicators with different dimensions to obtain the processed dataset.

[0019] Based on the processed dataset, the average thickness and thickness uniformity deviation of each material at the detection points in the preset locations are calculated as thickness difference features.

[0020] For shielding materials with different coatings, the surface roughness value, oxide layer thickness ratio, adhesion pass rate ratio, and coating peeling level corresponding quantitative values ​​are used as adhesion difference characteristics.

[0021] The tensile strength extremes, elongation at break, elastic modulus, wear resistance threshold, thickness decay rate after wear, and tensile strength retention rate were extracted from the processed dataset as mechanical property difference features.

[0022] The physicochemical property difference characteristics are obtained by integrating the thickness difference characteristics, adhesion difference characteristics, and mechanical property difference characteristics.

[0023] The present invention provides a method for quality testing of shielding materials, the steps of which include extracting the dynamic influence index of operating load: The temperature fluctuation curves, relative humidity change data, time series data of external electromagnetic interference intensity, and mechanical vibration frequency and amplitude records of the material's service environment are collected as dynamic environmental data.

[0024] The application scenarios, installation locations, service durations, and maintenance records of the materials are extracted as equipment operation and maintenance data, which are then combined with dynamic environmental data as historical application data.

[0025] Temperature fluctuations and humidity accumulation durations are extracted from historical application data, and the peak percentage of electromagnetic interference intensity and the range of interference frequency drift are calculated as environmental stress characteristics.

[0026] For shielding materials with different working conditions and materials, the vibration frequency change rate, effective value of amplitude, number of tensile cycles, and peak value of external force load are extracted as mechanical load characteristics.

[0027] Calculate the synergistic effect coefficient of temperature, humidity and electromagnetic interference, extract the performance degradation rate and load threshold critical point to establish the load performance correlation characteristics.

[0028] The Pearson correlation coefficient was used to calculate the correlation between each feature and the material performance degradation, and the core influencing features that reached the preset correlation threshold were retained.

[0029] The dynamic influence of operating load is constructed based on the influence weights of different core influence characteristics on material properties.

[0030] The present invention provides a quality inspection method for shielding materials, the steps of which include constructing a feature differentiation model for shielding materials: By combining the differences in shielding effectiveness with the differences in physicochemical properties, a comprehensive material feature matrix is ​​formed, and then standardized to obtain a comprehensive feature dataset.

[0031] Collect sample data of different types of shielding materials and label them with the corresponding material type. Divide the comprehensive feature dataset into training set, validation set and test set according to a preset ratio.

[0032] An improved XGBoost model was selected as the main classification model, K-nearest neighbors were introduced as an auxiliary classifier, and log loss function and gradient descent optimizer were chosen.

[0033] For the hyperparameters of the main classification model, a grid search method combined with 5-fold cross-validation is used to construct the search space and select the optimal combination of hyperparameters.

[0034] The training set is input into the model for iterative training. After each round of training, the classification accuracy is evaluated using the validation set. The hyperparameters are adjusted based on the validation set until the model converges and reaches the preset accuracy threshold to obtain the fusion model.

[0035] The test set is input into the fusion model, the classification performance is evaluated according to the preset index, and the confusion matrix is ​​plotted. The classification error for different material types is analyzed to obtain a model that masks material features for differentiation.

[0036] The present invention provides a method for quality testing of shielding materials, the steps of which include obtaining performance degradation response data: The shielding effectiveness differences and physicochemical property differences of the shielding material samples to be analyzed are input into the shielding material feature differentiation model to obtain the material type determination result for each sample.

[0037] The dynamic impact index of operating load is matched with the performance degradation monitoring data of the corresponding timestamp to construct a dataset affecting performance degradation.

[0038] For each group of shielding material samples, the grey relational analysis method was used to calculate the correlation between the dynamic influence index of the working load and each performance degradation index, so as to obtain the degree of influence of the performance index.

[0039] Based on the dataset affecting performance degradation and the degree of influence of performance indicators, a performance degradation response model is constructed for different material types based on the judgment results.

[0040] The dynamic influence index of different gradient load conditions is input into the performance degradation response model, and the corresponding performance degradation prediction value is output to obtain the performance degradation response curve. The degradation threshold index, degradation quantity and saturation degradation index are extracted from it as performance degradation response data.

[0041] The present invention provides a method for quality testing of shielding materials, the steps of which include obtaining interference identification results: Based on the sampling frequency that matches the detection frequency band, continuous signals are extracted using a preset time slice to obtain multiple signal segments.

[0042] Digital filtering is used to denoise and normalize multiple signal segments, eliminating signal strength differences under different detection environments to obtain multiple processed signal segments.

[0043] The amplitude standard deviation, pulse number, and signal duty cycle are extracted from multiple processed signal segments as interference time-domain features and compared with preset standard thresholds to preliminarily distinguish interference types.

[0044] The interference time-domain features are subjected to Fast Fourier Transform to extract frequency band features, which are then matched with a frequency band feature library of known interference sources to obtain the frequency band determination result.

[0045] The interference identification result is output by combining the interference type and frequency band determination results.

[0046] The present invention provides a method for quality testing of shielding materials, the steps of which include obtaining basic shielding effectiveness data: For each type of interference, a corresponding shielding effectiveness calibration algorithm and parameters are preset to form a mapping table.

[0047] Extract the original shielding effectiveness data corresponding to the interference type in the interference scenario, and call the corresponding calibration algorithm in the mapping table based on the interference identification results.

[0048] The original shielding effectiveness data is corrected based on the calibration algorithm to obtain the basic shielding effectiveness data.

[0049] The present invention provides a method for quality testing of shielding materials, the steps of which include obtaining quality test data are as follows: The service life and environmental exposure time of the current material are extracted and aligned with the performance degradation response data in the time dimension. The dynamic impact index of the current environmental load is also retrieved.

[0050] The average shielding effectiveness across the entire frequency band, the minimum shielding effectiveness in the key frequency band, and the performance stability coefficient are extracted from the basic shielding effectiveness data as core performance indicators.

[0051] The current material degradation is calculated based on the core performance indicators, and compared with the predicted performance degradation value to calculate the degradation deviation rate.

[0052] By combining the performance qualification standards and attenuation deviation rate of the material application scenario, the material quality is evaluated from the aspects of performance compliance, attenuation rationality and performance stability, and the material quality is divided into multiple quality levels to obtain quality test data.

[0053] This invention provides a quality testing system for shielding materials, comprising: The feature acquisition and extraction module is used to collect electromagnetic shielding feature data and physicochemical property feature data of different types of shielding materials, extract shielding effectiveness difference features from electromagnetic shielding feature data, and extract physicochemical property difference features from physicochemical property feature data.

[0054] The model building module is used to collect historical application data of shielding materials, extract the dynamic influence index of working condition load, and build a feature differentiation model of shielding materials based on the differences in shielding effectiveness and physicochemical properties.

[0055] The performance degradation analysis module is used to analyze the impact of the dynamic influence index of the operating load on the material performance degradation based on the shielding material characteristic differentiation model, and obtain performance degradation response data.

[0056] The interference quality detection module is used to dynamically identify interference signals in the detection scenario to obtain interference identification results. Based on the interference identification results, the shielding effectiveness calibration algorithm is determined to calculate the basic shielding effectiveness data. Combined with the performance decay response data, the current material quality status is detected to obtain quality detection data.

[0057] The beneficial effects of this invention are as follows: 1. This invention effectively eliminates interference from the testing environment through precise identification and targeted calibration of interference signals, reducing the error rate of shielding effectiveness test data and enabling a true reflection of the material's core performance. The constructed performance degradation response model can predict the material's degradation trend under different service environments based on the dynamic influence index of operating load, outputting key data such as the degradation threshold index and saturation degradation index, providing a quantitative basis for shielding material lifespan prediction and maintenance optimization. It enables quantitative prediction of performance degradation due to operating load, improving the accuracy and reliability of shielding effectiveness testing. Attached Figure Description

[0058] The invention will now be further described with reference to the accompanying drawings.

[0059] Figure 1 This is a schematic diagram of the module flow of a quality testing method for shielding materials provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the process of obtaining performance degradation response data in a quality testing method for shielding materials provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of a quality testing system for shielding materials provided in an embodiment of the present invention. Detailed Implementation

[0060] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0061] like Figures 1 to 3 As shown in the figure, an embodiment of the present invention provides a method for quality testing of shielding materials, comprising: Electromagnetic shielding characteristic data and physicochemical property characteristic data of different types of shielding materials were collected. The shielding effectiveness difference characteristics were extracted from the electromagnetic shielding characteristic data, and the physicochemical property difference characteristics were extracted from the physicochemical property characteristic data.

[0062] The steps for extracting the characteristics of shielding effectiveness differences include: We collected electric field attenuation values, magnetic field attenuation values, and electromagnetic wave reflection coefficients of different types of shielding materials at different frequency bands, and organized them according to material type, test frequency band, and test index to obtain electromagnetic shielding characteristic data.

[0063] Based on the application scenarios of shielding materials, the entire frequency band is divided into three intervals: low frequency band, mid frequency band, and high frequency band.

[0064] The minimum and average shielding effectiveness values ​​of each shielding material in the high-frequency band are calculated as high-frequency band threshold features.

[0065] This characteristic reflects the material's minimum attenuation capability against high-frequency electromagnetic waves and is a key indicator for distinguishing high-frequency shielding materials from general-purpose materials.

[0066] The maximum magnetic field attenuation rate and the standard deviation of attenuation rate fluctuation for each shielding material in the low-frequency band are extracted as the performance stability characteristics in the low-frequency band.

[0067] The magnetic field attenuation rate reflects the material's ability to attenuate low-frequency magnetic fields, and the standard deviation of the attenuation rate fluctuation reflects the material's performance stability in the low-frequency range.

[0068] Calculate the coefficient of variation of shielding effectiveness across the entire frequency band, and use the frequency range corresponding to the inflection point of shielding effectiveness changing with frequency as the effectiveness fluctuation coefficient.

[0069] The smaller the coefficient of variation, the more stable the shielding effectiveness of the material across different frequency bands. Simultaneously, the frequency sensitive range is extracted, i.e., the frequency range corresponding to the inflection point of the shielding effectiveness changing with frequency. This feature can distinguish between broadband shielding materials and narrowband dedicated materials.

[0070] The peak value and corresponding frequency point of the electromagnetic wave reflection coefficient in each frequency band are calculated as the characteristics of the electromagnetic energy reflection coefficient.

[0071] A higher peak reflection coefficient indicates a stronger ability of the material to reflect electromagnetic waves in the corresponding frequency band. By combining the absorption coefficient (absorption coefficient = 1 - reflection coefficient - transmission coefficient) and extracting the proportion of high absorption frequencies, this characteristic can distinguish between reflective and absorptive shielding materials.

[0072] The shielding effectiveness difference characteristics are obtained by integrating the high-frequency threshold characteristics, low-frequency performance stability characteristics, effectiveness fluctuation coefficient and electromagnetic energy reflection coefficient characteristics.

[0073] The steps for extracting differences in physicochemical properties include: Thickness, coating adhesion level, tensile strength, elongation at break and abrasion resistance of different types of shielding materials were collected as physicochemical property data.

[0074] The Z-Score standardization method was used to normalize the physicochemical indicators with different dimensions to obtain the processed dataset.

[0075] Based on the processed dataset, the average thickness and thickness uniformity deviation of each material at the detection points in the preset locations are calculated as thickness difference features.

[0076] For shielding materials with different coatings, the surface roughness value, oxide layer thickness ratio, adhesion pass rate ratio, and coating peeling level corresponding quantitative values ​​are used as adhesion difference characteristics.

[0077] The tensile strength extremes, elongation at break, elastic modulus, wear resistance threshold, thickness decay rate after wear, and tensile strength retention rate were extracted from the processed dataset as mechanical property difference features.

[0078] The differences in physicochemical properties are obtained by integrating the differences in back lake characteristics, adhesion characteristics, and mechanical properties.

[0079] Historical application data of shielding materials were collected, and the dynamic influence index of working condition load was extracted from them. Based on the differences in shielding effectiveness and physical and chemical properties, a characteristic differentiation model of shielding materials was constructed.

[0080] The steps for extracting the dynamic impact index of operating load include: The temperature fluctuation curves, relative humidity change data, time series data of external electromagnetic interference intensity, and mechanical vibration frequency and amplitude records of the material's service environment are collected as dynamic environmental data.

[0081] The application scenarios, installation locations, service durations, and maintenance records of the materials are extracted as equipment operation and maintenance data, which are then combined with dynamic environmental data as historical application data.

[0082] Temperature fluctuations and humidity accumulation durations are extracted from historical application data, and the peak percentage of electromagnetic interference intensity and the range of interference frequency drift are calculated as environmental stress characteristics.

[0083] It reflects the aging-accelerating effect of temperature and humidity changes on materials and demonstrates the dynamic impact characteristics of external electromagnetic loads.

[0084] For shielding materials with different working conditions and materials, the rate of change of vibration frequency, effective value of amplitude, number of tensile cycles, and peak value of external force load are extracted as mechanical load characteristics. This reflects the dynamic load intensity of mechanical vibration and embodies the cumulative effect of dynamic mechanical stress.

[0085] Calculate the synergistic effect coefficient of temperature, humidity and electromagnetic interference, extract the performance degradation rate and load threshold critical point to establish the load performance correlation characteristics.

[0086] The Pearson correlation coefficient was used to calculate the correlation between each feature and the material performance degradation, and the core influencing features that reached the preset correlation threshold were retained.

[0087] The dynamic influence index of operating load is constructed based on the influence weights of different core influencing characteristics on material properties, and the formula is expressed as follows: In the formula, It is the dynamic impact index of operating load. It is the temperature and humidity coupling coefficient. It is the percentage of peak electromagnetic interference. It is the rate of change of vibration frequency. , , These are feature weights.

[0088] The steps for constructing a feature differentiation model for shielding materials include: By combining the differences in shielding effectiveness with the differences in physicochemical properties, a comprehensive material feature matrix is ​​formed, and then standardized to obtain a comprehensive feature dataset.

[0089] Collect sample data of different types of shielding materials and label them with the corresponding material type. Divide the comprehensive feature dataset into training set, validation set and test set according to a preset ratio.

[0090] An improved XGBoost model was selected as the main classification model, K-nearest neighbors were introduced as an auxiliary classifier, and log loss function and gradient descent optimizer were chosen.

[0091] For the hyperparameters of the main classification model, a grid search method combined with 5-fold cross-validation is used to construct the search space and select the optimal combination of hyperparameters.

[0092] The training set is input into the model for iterative training. After each round of training, the classification accuracy is evaluated using the validation set. The hyperparameters are adjusted based on the validation set until the model converges and reaches the preset accuracy threshold to obtain the fusion model.

[0093] The test set is input into the fusion model, and the classification performance is evaluated according to preset metrics. A confusion matrix is ​​plotted, and the classification error for different material types is analyzed to obtain a material feature-masked distinguishing model. Preset metrics may include classification accuracy, precision, recall, and F1 score.

[0094] Based on the characteristic differentiation model of shielding materials, the influence of the dynamic influence index of working load on the performance degradation of materials is analyzed to obtain performance degradation response data.

[0095] The steps to obtain performance degradation response data include: The shielding effectiveness differences and physicochemical property differences of the shielding material samples to be analyzed are input into the shielding material feature differentiation model to obtain the material type determination result for each sample.

[0096] The dynamic impact index of operating load is matched with the performance degradation monitoring data at the corresponding timestamp to construct a dataset affecting performance degradation. Performance degradation monitoring data includes performance degradation indicators such as shielding effectiveness degradation value, adhesion reduction level, and tensile strength loss rate at the corresponding timestamp.

[0097] For each group of shielding material samples, the grey relational analysis method was used to calculate the correlation between the dynamic influence index of the working load and each performance degradation index, so as to obtain the degree of influence of the performance index.

[0098] Based on the dataset affecting performance degradation and the degree of influence of performance indicators, a performance degradation response model is constructed for different material types based on the judgment results.

[0099] The model input is set as the dynamic impact index of the operating load, and the output is the key performance degradation index, such as the shielding effectiveness degradation value.

[0100] Machine learning models, such as random forest regression and LSTM time series prediction models, are used to fit the mapping relationship between the dynamic impact index and performance degradation. The model parameters are optimized with the goal of minimizing the mean squared error (MSE) to ensure that the model accurately reflects the driving effect of changes in the dynamic impact index on performance degradation.

[0101] The dynamic influence index of different gradient load conditions is input into the performance degradation response model, and the corresponding performance degradation prediction value is output to obtain the performance degradation response curve. The degradation threshold index, degradation quantity and saturation degradation index are extracted from it as performance degradation response data.

[0102] Attenuation threshold index: The dynamic impact index corresponding to the performance degradation value reaching the qualified critical value.

[0103] Decay rate: The incremental performance degradation caused by a change in the dynamic effect exponential.

[0104] Saturation decay index: The critical value of the dynamic influence index when the performance decay tends to level off.

[0105] Interference signals in the detection scenario are dynamically identified to obtain interference identification results. Based on the interference identification results, a shielding effectiveness calibration algorithm is determined to calculate basic shielding effectiveness data. Combined with performance decay response data, the current material quality status is detected to obtain quality detection data.

[0106] The steps to obtain interference identification results include: Based on the sampling frequency that matches the detection frequency band, continuous signals are extracted using a preset time slice to obtain multiple signal segments.

[0107] Digital filtering is used to denoise and normalize multiple signal segments, eliminating signal strength differences under different detection environments to obtain multiple processed signal segments.

[0108] The amplitude standard deviation, pulse number, and signal duty cycle are extracted from multiple processed signal segments as interference time-domain features and compared with preset standard thresholds to preliminarily distinguish interference types.

[0109] Calculate the amplitude standard deviation of each processed signal segment, count the number of pulses exceeding the preset amplitude threshold within each processed signal segment, and calculate the signal duty cycle by the ratio of the duration of the high amplitude segment to the total duration of multiple processed signal segments.

[0110] Judgment rules: If the amplitude standard deviation is < 0.05, it is judged as stationary background noise. If 0.05 ≤ amplitude standard deviation ≤ 0.3, it is judged as industrial electromagnetic interference, such as pulse interference from motor start-stop. If the amplitude standard deviation is > 0.3, it is judged as radio frequency signal interference, such as frequency hopping signals from base stations.

[0111] If the number of pulses is 0, it is stationary noise. If 1 ≤ number of pulses ≤ 5, it is intermittent industrial interference. If the number of pulses > 5, it is dense radio frequency pulse interference.

[0112] If the signal duty cycle is <10%, it is considered transient spike interference, such as electrostatic discharge. A signal duty cycle of 10%–50% is considered periodic industrial interference, such as relay on / off signals.

[0113] The interference time-domain features are subjected to Fast Fourier Transform to extract frequency band features, which are then matched with a frequency band feature library of known interference sources to obtain the frequency band determination result.

[0114] Perform a Fast Fourier Transform on each signal segment to obtain the frequency-amplitude distribution spectrum of the signal, and extract the main frequency point (the frequency with the largest amplitude) and the frequency band distribution range.

[0115] Preset the frequency band characteristics of three typical types of interference: Compare the main frequency point and frequency band range of the signal with the feature database: If the main frequency falls between 45Hz and 65Hz, it is directly identified as power frequency interference.

[0116] If the frequency band is distributed between 1kHz and 10MHz and there is no obvious dominant frequency, it is judged as industrial electromagnetic interference.

[0117] If discrete main frequencies appear and fall within 800MHz–6GHz, it is determined to be radio frequency signal interference.

[0118] If the frequency band does not have the above characteristics, it is determined to be an unknown type of interference.

[0119] The interference identification result is output by combining the interference type and frequency band determination results.

[0120] The principle of "prioritizing frequency domain characteristics and verifying time domain characteristics" is adopted: Example 1: If the frequency domain is determined to be power frequency interference and the time domain amplitude fluctuation is <0.05, it is finally determined to be power frequency background interference.

[0121] Example 2: If the frequency domain is determined to be radio frequency interference and the peak pulse count in the time domain is greater than the peak pulse count, it is ultimately determined to be dense radio frequency pulse interference.

[0122] Example 3: No matching characteristics in the frequency domain, but time domain fluctuation > 0.3 → ultimately determined to be unknown broadband interference.

[0123] The steps to obtain basic shielding effectiveness data include: for each type of interference, pre-setting the corresponding shielding effectiveness calibration algorithm and parameters, and forming a mapping table.

[0124] Mapping table: Extract the original shielding effectiveness data corresponding to the interference type in the interference scenario, and call the corresponding calibration algorithm in the mapping table based on the interference identification results.

[0125] The original shielding effectiveness data is corrected based on the calibration algorithm to obtain the basic shielding effectiveness data.

[0126] Scenario 1: Stable background noise → Direct attenuation calculation method: Calculate the mean of the original shielding effectiveness within the effective data segment. Subtract the preset background noise baseline value, and subtract the baseline value from the mean to obtain the calibrated shielding effectiveness.

[0127] Scenario 2: Power frequency interference → Notch filter + baseline subtraction method: Use a notch filter to remove power frequency interference signals in the 45–65Hz frequency band.

[0128] The filtered signal is linearly fitted to obtain the interference baseline curve. The calibration value is calculated by subtracting the interference baseline curve from the baseline value.

[0129] Scenario 3 Radio frequency signal interference → frequency band filtering + energy weighting method: effective test data outside the radio frequency characteristic frequency band is screened out. The energy proportion of each frequency band is calculated, and energy weighting is performed on the shielding effectiveness values of different frequency bands.

[0130] The formula expression of the calibration value is: In the formula, is the number of effective frequency bands, is the energy proportion of each frequency band, is the calibration value, is the interference baseline curve.

[0131] The steps of obtaining quality detection data through detection include: Align the service time and environmental exposure time of the current material with the performance attenuation response data in the time dimension, and retrieve the dynamic influence index of working condition load of the current environment.

[0132] Extract the full-band average shielding effectiveness, the minimum shielding effectiveness of key frequency bands and the performance stability coefficient from the basic shielding effectiveness data as core performance indicators.

[0133] Calculate the attenuation of the current material according to the core performance indicators, compare it with the predicted performance attenuation value, and calculate the attenuation deviation rate.

[0134] Combined with the performance qualification standard and attenuation deviation rate of the material application scenario, the evaluation is carried out from performance compliance, attenuation rationality and performance stability, and the material quality is divided into multiple quality grades to obtain quality detection data.

[0135] Performance compliance assessment: compare the full-band average shielding effectiveness (SEavg) and the minimum shielding effectiveness of key frequency bands (SEmin−key) with the preset minimum qualification threshold SEstd: if SEavg≥SEstd and SEmin−key≥SEstd → compliant. If SEavg≥SEstd but SEmin−key<SEstd → partial frequency bands are not compliant. If SEavg<SEstd → overall not compliant.

[0136] Attenuation rationality assessment: judgment based on attenuation deviation rate ηΔSE: ηΔSE≤10% → attenuation is reasonable, and performance attenuation conforms to the influence law of working conditions.

[0137] 10%<ηΔSE≤30% → attenuation slightly exceeds expectation, and environmental monitoring needs to be strengthened.

[0138] ηΔSE>30% → attenuation is abnormal, and there is a risk of material failure.

[0139] Performance stability assessment: judgment based on performance stability coefficient Cs: When Cs ≤ 0.1, good performance stability is obtained.

[0140] When 0.1 < Cs ≤ 0.3, the performance stability is moderate, and fluctuations in some frequency bands are large.

[0141] When Cs > 0.3, the performance stability is poor, and the consistency of shielding effectiveness cannot meet application requirements.

[0142] Material quality grade table: Based on the same general inventive concept, the present invention further protects a quality detection system for shielding materials, comprising: a feature acquisition and extraction module, configured to acquire electromagnetic shielding feature data and physical and chemical property feature data of different types of shielding materials, extract shielding effectiveness difference features from the electromagnetic shielding feature data, and extract physical and chemical property difference features from the physical and chemical property feature data.

[0143] a distinguishing model construction module, configured to collect historical application data of shielding materials, extract a dynamic influence index of working condition load therefrom, and construct a shielding material feature distinguishing model based on the shielding effectiveness difference features and the physical and chemical property difference features.

[0144] a performance attenuation analysis module, configured to analyze the influence of the dynamic influence index of working condition load on material performance attenuation based on the shielding material feature distinguishing model, to obtain performance attenuation response data.

[0145] an interference quality detection module, configured to dynamically identify interference signals in a detection scene to obtain an interference identification result, determine a shielding effectiveness calibration algorithm based on the interference identification result to calculate and obtain basic shielding effectiveness data, and detect the current material quality condition in combination with the performance attenuation response data to obtain quality detection data.

[0146] In summary, the shielding material quality detection method and system provided by this embodiment reduce the influence of environmental noise on measurement results through the interference identification and calibration algorithm, significantly improve the accuracy of material type identification, and avoid evaluation deviation caused by material misjudgment; realize environment adaptive calibration of shielding effectiveness, making on-site detection results closer to the real electromagnetic protection capability; reveal the quantitative relationship between working condition load and performance attenuation, and support life prediction and failure early warning; construct a dynamic, multi-dimensional and interpretable quality evaluation system, output refined quality grades, support differentiated operation and maintenance strategies, and achieve high precision, strong adaptability and engineering practicability of shielding material quality detection.

[0147] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for quality testing of shielding materials, comprising: Electromagnetic shielding characteristic data and physicochemical property characteristic data of different types of shielding materials are collected. Shielding effectiveness difference characteristics are extracted from the electromagnetic shielding characteristic data, and physicochemical property difference characteristics are extracted from the physicochemical property characteristic data. Collect historical application data of shielding materials, extract the dynamic influence index of working condition load, and construct a feature differentiation model of shielding materials based on the differences in shielding effectiveness and the differences in physicochemical properties. The steps for extracting the dynamic impact index of the operating condition load include: The temperature fluctuation curves, relative humidity change data, time series data of external electromagnetic interference intensity, and mechanical vibration frequency and amplitude records of the material's service environment are collected as dynamic environmental data. The application scenarios, installation locations, service durations, and maintenance records of the materials are extracted as equipment operation and maintenance data, which are then combined with the dynamic environmental data to form the historical application data. Temperature fluctuations and humidity accumulation durations are extracted from the historical application data, and the peak percentage of electromagnetic interference intensity and the frequency drift range of interference are calculated as environmental stress characteristics. For shielding materials with different working conditions and materials, the vibration frequency change rate, effective value of amplitude, number of tensile cycles and peak value of external force load are extracted as mechanical load characteristics. Calculate the synergistic effect coefficient of temperature-humidity-electromagnetic interference, extract the performance degradation rate and load threshold critical point to establish load performance correlation characteristics; The Pearson correlation coefficient was used to calculate the correlation between each feature and the material performance degradation, and the core influencing features that reached the preset correlation threshold were retained. The dynamic influence of working condition load is constructed based on the influence weights of different core influence characteristics on material properties. Based on the shielding material characteristic differentiation model, the influence of the dynamic influence index of the operating load on the material performance degradation is analyzed to obtain performance degradation response data; the steps to obtain the performance degradation response data include: The shielding effectiveness difference characteristics and physicochemical property difference characteristics of the shielding material samples to be analyzed are input into the shielding material feature differentiation model to obtain the material type determination result for each sample; The dynamic impact index of the operating condition load is matched with the performance degradation monitoring data of the corresponding timestamp to construct a dataset affecting performance degradation. For each group of shielding material samples, the grey relational analysis method is used to calculate the correlation between the dynamic influence index of the working condition load and each performance degradation index, so as to obtain the degree of influence of the performance index. Based on the dataset of performance degradation and the degree of influence of the performance indicators, a performance degradation response model is constructed for the judgment results of different material types. The dynamic influence index of different gradient load conditions is input into the performance degradation response model, and the corresponding performance degradation prediction value is output to obtain the performance degradation response curve. The degradation threshold index, degradation rate and saturation degradation index are extracted from it as the performance degradation response data. Interference signals in the detection scenario are dynamically identified to obtain interference identification results. Based on the interference identification results, a shielding effectiveness calibration algorithm is determined to calculate basic shielding effectiveness data. Combined with the performance decay response data, the current material quality status is detected to obtain quality detection data.

2. The quality testing method for a shielding material according to claim 1, characterized in that: The steps for extracting the shielding effectiveness difference features include: The electric field attenuation value, magnetic field attenuation value, and electromagnetic wave reflection coefficient of different types of shielding materials at different frequency bands are collected and organized according to material type-test frequency band-test index to obtain the electromagnetic shielding characteristic data; Based on the application scenarios of shielding materials, the entire frequency band is divided into three intervals: low frequency band, mid frequency band, and high frequency band. The minimum and average shielding effectiveness values ​​of each shielding material in the high-frequency band are calculated as high-frequency band threshold features; The maximum magnetic field attenuation rate and the standard deviation of attenuation rate fluctuation of each shielding material in the low-frequency band are extracted as the low-frequency band performance stability characteristics. Calculate the coefficient of variation of the shielding effectiveness across the entire frequency band, and use the frequency range corresponding to the inflection point of the shielding effectiveness change with frequency as the effectiveness fluctuation coefficient. The peak value and corresponding frequency point of the electromagnetic wave reflection coefficient in each frequency band are calculated as the characteristics of the electromagnetic energy reflection coefficient. The shielding effectiveness difference characteristics are obtained by integrating the high-frequency threshold characteristics, the low-frequency performance stability characteristics, the effectiveness fluctuation coefficient, and the electromagnetic energy reflection coefficient characteristics.

3. The quality testing method for a shielding material according to claim 1, characterized in that: The steps for extracting the differences in the physicochemical properties include: The thickness, coating adhesion level, tensile strength, elongation at break, and abrasion resistance of different types of shielding materials were collected as the physicochemical property data. The Z-Score standardization method was used to normalize the physicochemical indicators with different dimensions to obtain the processed dataset. Based on the processed dataset, the average thickness and thickness uniformity deviation of each material at the detection points in the preset locations are calculated as thickness difference features. For shielding materials with different coatings, the surface roughness value, oxide layer thickness ratio, adhesion pass rate ratio, and coating peeling level corresponding quantitative values ​​are used as adhesion difference characteristics. The tensile strength extreme value, elongation at break, elastic modulus, wear resistance threshold, thickness decay rate after wear, and tensile strength retention rate are extracted from the processed dataset as mechanical property difference features. The physicochemical property difference characteristics are obtained by integrating the thickness difference characteristics, the adhesion difference characteristics, and the mechanical property difference characteristics.

4. The quality testing method for a shielding material according to claim 1, characterized in that: The steps for constructing the feature differentiation model of the shielding material include: The shielding effectiveness difference features and the physicochemical property difference features are concatenated to form a material comprehensive feature matrix, which is then standardized to obtain a comprehensive feature dataset. Collect sample data of different types of shielding materials and label them with corresponding material type tags. Divide the comprehensive feature dataset into training set, validation set and test set according to a preset ratio. An improved XGBoost model was selected as the main classification model, K-nearest neighbors were introduced as an auxiliary classifier, and log loss function and gradient descent optimizer were chosen. For the hyperparameters of the main classification model, a search space is constructed using a grid search method combined with 5-fold cross-validation to select the optimal hyperparameter combination; The training set is input into the model for iterative training. After each round of training, the classification accuracy is evaluated using the validation set. The hyperparameters are adjusted according to the validation set until the model converges and reaches the preset accuracy threshold to obtain the fusion model. The test set is input into the fusion model, the classification performance is evaluated according to preset indicators, a confusion matrix is ​​plotted, and the classification error for different material types is analyzed to obtain the feature discrimination model of the shielding material.

5. The quality testing method for a shielding material according to claim 1, characterized in that: The steps to obtain the interference identification result include: Based on the sampling frequency that matches the detection frequency band, continuous signals are extracted using a preset time slice to obtain multiple signal segments; Digital filtering is used to denoise and normalize multiple signal segments, eliminating signal strength differences under different detection environments to obtain multiple processed signal segments; The amplitude standard deviation, pulse number, and signal duty cycle are extracted from multiple processed signal segments as interference time-domain features and compared with preset standard thresholds to preliminarily distinguish interference types. The interference time-domain features are subjected to Fast Fourier Transform to extract frequency band features and matched with the frequency band feature library of known interference sources to obtain the frequency band determination result; The interference identification result is output by combining the interference type and the frequency band determination result.

6. The quality testing method for a shielding material according to claim 1, characterized in that: The steps for obtaining the basic shielding effectiveness data include: For each type of interference, a corresponding shielding effectiveness calibration algorithm and parameters are preset to form a mapping table; Extract the original shielding effectiveness data under the interference scenario corresponding to the interference type, and call the corresponding calibration algorithm in the mapping table according to the interference identification result; The original shielding effectiveness data is specifically corrected based on the calibration algorithm to calculate the basic shielding effectiveness data.

7. The quality testing method for a shielding material according to claim 1, characterized in that: The steps for obtaining the quality inspection data include: The service life and environmental exposure time of the current material are extracted and aligned with the performance degradation response data in terms of time dimension, and the dynamic impact index of the current environmental load is retrieved. The average shielding effectiveness across the entire frequency band, the minimum shielding effectiveness in the key frequency band, and the performance stability coefficient are extracted from the basic shielding effectiveness data as core performance indicators. The attenuation of the current material is calculated based on the core performance indicators, and compared with the predicted performance attenuation value to calculate the attenuation deviation rate. By combining the performance qualification standards of the material application scenario with the attenuation deviation rate, the material quality is evaluated from the perspectives of performance compliance, attenuation rationality and performance stability, and the quality test data is obtained by classifying the material quality into multiple quality levels.

8. A quality inspection system for shielding materials, applied to a quality inspection method for shielding materials as described in any one of claims 1 to 7, characterized in that, The quality inspection system includes: The feature acquisition and extraction module is used to acquire electromagnetic shielding feature data and physicochemical property feature data of different types of shielding materials, extract shielding effectiveness difference features from the electromagnetic shielding feature data, and extract physicochemical property difference features from the physicochemical property feature data. The model construction module is used to collect historical application data of shielding materials, extract the dynamic influence index of working condition load, and construct a shielding material feature differentiation model based on the shielding effectiveness difference characteristics and the physicochemical property difference characteristics. The performance degradation analysis module is used to analyze the impact of the dynamic influence index of the working condition load on the material performance degradation based on the characteristic differentiation model of the shielding material to obtain performance degradation response data. The interference quality detection module is used to dynamically identify interference signals in the detection scenario to obtain interference identification results. Based on the interference identification results, a shielding effectiveness calibration algorithm is determined to calculate basic shielding effectiveness data. Combined with the performance decay response data, the current material quality status is detected to obtain quality detection data.

Images