A testing method and device for insulating adhesive

Abstract

The present invention relates to the technical field of insulating glue testing, and discloses a testing method and device for insulating glue. The method comprises the following steps: synchronously collecting multidimensional performance parameters of the insulating glue to be tested, extracting characteristic vectors of an initial data set, performing a dynamic loading experiment on the insulating glue to be tested based on a performance characteristic matrix, constructing a time series curve of the dynamic response data set, performing multidimensional evaluation on a degradation parameter set based on information entropy to obtain a comprehensive score, and performing anomaly detection on the comprehensive score; synchronously collecting multidimensional performance parameters of the insulating glue to be tested, constructing a performance characteristic matrix, so that the test data can comprehensively reflect the performance of the insulating glue in different dimensions; analyzing the comprehensive score based on anomaly detection, effectively identifying potential quality risks, avoiding misjudgment caused by abnormal local indicators, and iteratively dividing quality grades to ensure that the quality evaluation results can more reasonably guide the selection and application of the insulating glue, thereby improving the reliability of the insulating glue.

Description

Technical Field

[0001] The present invention relates to the technical field of insulating adhesive testing, and in particular to a method and device for testing insulating adhesive. Background Art

[0002] Insulating adhesive is a key material widely used in electronics, power generation, aerospace, and other fields, primarily for providing electrical insulation, protective sealing, and mechanical support. As electronic devices evolve toward higher power and smaller sizes, the performance requirements for insulating adhesives are increasing. Not only must they possess excellent electrical insulation properties, but they must also possess good mechanical strength, environmental stability, and long-term reliability. For example, in high-voltage transmission equipment, insulating adhesives must withstand high electric field strengths while maintaining stable dielectric properties. In electronic packaging applications, insulating adhesives must exhibit excellent adhesion and resistance to thermal expansion to prevent cracking or delamination during long-term use.

[0003] In the existing technology, the testing methods for insulating adhesives mostly use static measurement methods, that is, independently testing parameters such as resistivity and dielectric constant, and then making quality judgments through data comparison or empirical formulas. However, during the implementation process, it is usually difficult to fully capture the performance degradation characteristics of the material under dynamic use conditions, especially the performance under long-term or high-stress environments. As a result, the test results cannot accurately reflect the long-term stability of the insulating adhesive in actual use.

[0004] Therefore, it is necessary to provide a testing method and device for insulating adhesive to solve the problem that the existing testing method for insulating adhesive cannot accurately reflect the long-term stability of the insulating adhesive in actual use. Summary of the Invention

[0005] The main purpose of the present invention is to provide a testing method and device for insulating glue, aiming to solve the technical problems mentioned in the above background technology.

[0006] The present invention adopts the following technical solutions:

[0007] A testing method for insulating adhesive, comprising:

[0008] Synchronously collecting multi-dimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence;

[0009] Extracting eigenvectors of the initial data set, and constructing a matrix based on the eigenvectors to obtain a performance characteristic matrix;

[0010] Performing a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set;

[0011] Constructing a time series curve of the dynamic response data set, performing degradation trend fitting processing on the time series curve, and obtaining a degradation parameter set;

[0012] Performing a multi-dimensional evaluation on the degradation parameter set based on information entropy to obtain a comprehensive score;

[0013] Anomaly detection is performed on the comprehensive score, contribution analysis is performed on the comprehensive score after detection and the degradation parameter set, and the analysis result is iteratively divided into quality levels to obtain a quality judgment result.

[0014] Furthermore, the step of synchronously collecting multi-dimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence, includes:

[0015] Perform multi-channel physical excitation on the insulating adhesive to be tested, and perform time stamp alignment processing on the excitation signal to obtain the original measurement signal;

[0016] Performing signal segmentation on the original measurement signal and performing multi-dimensional parameter conversion processing on the segmented signal to obtain preliminary performance data, wherein the preliminary performance data includes resistivity, dielectric constant and tensile strength;

[0017] Performing dynamic filtering on the preliminary performance data to obtain a smoothed data sequence, and performing statistical analysis on the smoothed data sequence based on a sliding window to obtain a filtered data set;

[0018] Segmenting the filtered data set into time series, and performing feature labeling processing on the segmentation results to obtain a segmented feature data set;

[0019] The segmented feature data sets are integrated in the order of timestamps to obtain an initial data set.

[0020] Furthermore, the step of extracting the eigenvectors of the initial data set and constructing a matrix based on the eigenvectors to obtain a performance characteristic matrix includes:

[0021] Performing multi-scale hierarchical processing on the initial data set based on the resistivity sequence, the dielectric constant sequence, and the tensile strength sequence to obtain a hierarchical sequence group;

[0022] Performing nonlinear mapping on the layered sequence group to obtain a primary feature vector set, and performing weighted fusion processing on the layered sequence group based on the primary feature vector set to obtain a fused feature matrix;

[0023] Performing singular value decomposition on the fused feature matrix to obtain a reduced-dimensional feature matrix;

[0024] Based on the projection value of each dimension in the dimension-reduced feature matrix, performing residual correction processing on the primary feature vector set to obtain an optimized feature vector set;

[0025] The optimized feature vector set is subjected to tensor reconstruction processing to obtain a performance feature matrix.

[0026] Furthermore, the step of performing a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set includes:

[0027] Performing loading parameter mapping processing on the performance characteristic matrix to obtain a loading condition sequence;

[0028] Performing a progressive multi-field loading experiment on the insulating adhesive to be tested according to the loading condition sequence to obtain an initial response data set;

[0029] Performing multi-scale time-frequency decomposition processing on the initial response data set to obtain a time-frequency feature component set;

[0030] Performing adaptive noise reduction processing on the initial response data set according to the time-frequency feature component set to obtain a smoothed response data set;

[0031] Performing state transition probability modeling on the smooth response data set, constructing a state transition relationship between parameters in the smooth response data set, and obtaining a dynamic evolution feature matrix;

[0032] The smooth response data set is subjected to multi-parameter fusion processing according to the dynamic evolution characteristic matrix to obtain a dynamic response data set.

[0033] Furthermore, the step of constructing a time series curve of the dynamic response data set and performing degradation trend fitting processing on the time series curve to obtain a degradation parameter set includes:

[0034] Performing multidimensional time series decomposition processing on the dynamic response data set to obtain a primary time series group;

[0035] Performing curve smoothing processing on the primary time series group, and marking the smoothing results in segments to obtain a segmented trend series set;

[0036] Setting an initialization decay rate for each trend segment in the segmented trend sequence set, and iteratively optimizing each of the initialization decay rates based on an optimization algorithm to obtain an initial value set of fitting parameters;

[0037] Performing dynamic weight fitting processing on the segmented trend sequence set according to the initial value set of the fitting parameters to obtain an optimized time series curve;

[0038] Extracting degradation features of the optimized time series curve, and reorganizing the degradation features according to performance dimensions to obtain a degradation feature matrix;

[0039] A multi-scale residual analysis is performed on the degradation characteristic matrix, and the degradation characteristic matrix is optimized based on the distribution characteristics of the residual results to obtain a degradation parameter set.

[0040] Furthermore, the step of performing a multi-dimensional evaluation on the degradation parameter set based on information entropy to obtain a comprehensive score includes:

[0041] Performing single feature extraction on the degradation parameter set to obtain a single performance indicator set, and performing probability distribution calculation on information entropy based on the single performance indicator set to obtain an entropy value sequence;

[0042] Setting an initial weight vector, and performing nonlinear mapping processing on the initial weight vector according to each entropy value in the entropy value sequence to obtain an optimized weight set;

[0043] Performing weighted fusion processing on the individual performance indicator sets according to the optimized weight set to obtain a preliminary evaluation score;

[0044] Performing residual analysis on the preliminary evaluation score to obtain an optimized evaluation score;

[0045] A comprehensive characterization function is constructed based on the optimization evaluation score, and the optimization evaluation score is subjected to comprehensive characterization processing according to the comprehensive characterization function to obtain a comprehensive score.

[0046] Furthermore, the step of performing abnormality detection on the comprehensive score, performing contribution analysis on the detected comprehensive score and the degradation parameter set, and iteratively dividing the analysis results into quality grades to obtain a quality determination result includes:

[0047] Performing multi-scale sliding window processing on the comprehensive score to obtain a preliminary abnormal distribution sequence;

[0048] Performing dynamic anomaly detection on the comprehensive score according to the statistical distribution characteristics of the abnormal preliminary distribution sequence to obtain an abnormal label set;

[0049] Performing weighted cluster analysis on the abnormal points in the abnormal mark set according to the distribution density and abnormal degree of the abnormal points in the abnormal mark set to obtain an abnormal cluster feature vector;

[0050] Performing contribution decomposition on the degradation parameter set according to the abnormal clustering feature vector to obtain a parameter influence weight matrix;

[0051] Constructing a multidimensional projection transformation matrix based on the parameter influence weight matrix, and mapping the comprehensive score and degradation parameter set to a high-dimensional feature space through the multidimensional projection transformation matrix to obtain a quality feature projection set;

[0052] The quality feature projection set is iteratively optimized based on a support vector machine classification algorithm to obtain a quality determination result, which includes an abnormality mark and a multi-level quality classification label.

[0053] The present invention also provides a testing device for insulating adhesive, comprising:

[0054] An acquisition module is used to synchronously acquire multi-dimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence;

[0055] An extraction module, configured to extract the eigenvectors of the initial data set and construct a matrix based on the eigenvectors to obtain a performance characteristic matrix;

[0056] An experimental module, configured to perform a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set;

[0057] A construction module is used to construct a time series curve of the dynamic response data set, perform degradation trend fitting processing on the time series curve, and obtain a degradation parameter set;

[0058] An evaluation module, configured to perform a multi-dimensional evaluation on the degradation parameter set based on information entropy to obtain a comprehensive score;

[0059] The determination module is used to perform anomaly detection on the comprehensive score, perform contribution analysis on the detected comprehensive score and the degradation parameter set, and iteratively divide the analysis results into quality levels to obtain a quality determination result.

[0060] The present invention further provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and running on the processor, wherein the processor implements the above method when executing the computer program.

[0061] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the above method when executed by a processor.

[0062] Beneficial effects:

[0063] In this invention, by synchronously collecting multi-dimensional performance parameters of the insulating adhesive under test and constructing a performance characteristic matrix, the test data comprehensively reflects the adhesive's performance across different dimensions. Based on this performance characteristic matrix, dynamic loading experiments are conducted to generate a dynamic response dataset. This allows the testing method to not only focus on the initial performance indicators of the insulating adhesive but also capture the material's response characteristics under dynamic usage conditions, thereby more accurately reflecting its long-term stability. A time series curve of the dynamic response dataset is then constructed and fitted with degradation trends to form a degradation parameter set. This enables quantitative analysis of the insulating adhesive's performance trends and facilitates accurate durability prediction. Furthermore, information entropy is used to perform a multi-dimensional evaluation of the degradation parameter set, generating a comprehensive score. This allows quality assessment to evaluate the overall quality level of the insulating adhesive through the integration of multi-dimensional information, improving the scientific nature and accuracy of the test. Furthermore, anomaly detection-based analysis of the comprehensive score, combined with the contribution of the degradation parameter set, effectively identifies potential quality risks and avoids misjudgments caused by localized indicator anomalies. Furthermore, through iterative quality grading, the quality assessment results can more effectively guide the selection and application of insulating adhesives, improving their reliability and service life in practical projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a schematic diagram of the steps of a testing method for insulating adhesive of the present invention;

[0065] Figure 2 It is a schematic structural diagram of a testing device for insulating glue of the present invention;

[0066] Figure 3 is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention;

[0067] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0068] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0069] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like, indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0070] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections, direct connections, or indirect connections through an intermediate medium; they may refer to internal communication between two components or the interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0071] In the present invention, unless otherwise expressly specified or limited, a first feature being "above" or "below" a second feature may include the first and second features being in direct contact, or may include the first and second features not being in direct contact but being in contact via another feature between them. Furthermore, a first feature being "above," "above," and "above" a second feature may include the first feature being directly above or obliquely above the second feature, or may simply mean that the first feature is higher in level than the second feature. A first feature being "below," "below," and "below" a second feature may include the first feature being directly below or obliquely below the second feature, or may simply mean that the first feature is lower in level than the second feature.

[0072] Reference Figure 1 The present invention proposes a testing method for insulating glue, comprising the following steps:

[0073] S1: synchronously collecting multi-dimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence;

[0074] In step S1, the sample preparation of the insulating glue to be tested is clarified, and the insulating glue is prepared into standardized specimens, such as rectangular sheets or cylindrical test blocks, to ensure the consistency of the test conditions. Use high-precision testing equipment to synchronously collect multi-dimensional performance parameters, that is, complete the measurement of multiple parameters within the same time period to avoid data deviations caused by time differences. Specifically, a precision resistivity tester is used to measure the resistivity sequence. The resistivity is calculated by applying a constant voltage and recording the current. It is necessary to repeat the measurement multiple times under different temperature or humidity conditions to form a resistivity sequence containing multiple data points. At the same time, the acquisition of the dielectric constant sequence is tested by a dielectric spectrometer. The trend of the dielectric constant change is obtained by measuring the capacitance characteristics of the insulating glue at different frequencies. Assuming that the test frequency range is set to 1Hz to 1MHz, a dielectric constant sequence that changes with frequency can be obtained. The measurement of the tensile strength sequence depends on a universal material testing machine. By applying a gradually increasing tensile force to the sample until it breaks, the maximum strength value of each tensile test is recorded. After repeated multiple times, a tensile strength sequence is formed. Multidimensional data consisting of multiple time series is generated. This data is then subjected to a sliding window process for outlier removal, with the window size set to five consecutive sampling points. Outliers are identified and removed based on a statistical standard deviation threshold, resulting in an initial dataset consisting of screened resistivity, dielectric constant, and tensile strength series. For example, when testing a piece of silicone-based insulating adhesive, the resistivity series might show 10^12 Ω·m at 25°C, decreasing to 10^11 Ω·m at 50°C. The dielectric constant series might be 3.5 at 1kHz and 3.2 at 100kHz, while the tensile strength series might include five measurements, such as 20MPa, 22MPa, and so on. The test data is then consolidated into the initial dataset. The entire process requires strict control of environmental variables and instrument calibration to ensure data reproducibility and scientific validity.

[0075] S2: extracting the eigenvectors of the initial data set, and constructing a matrix based on the eigenvectors to obtain a performance characteristic matrix;

[0076] In step S2, the initial data set is normalized to eliminate dimensional differences. The mean, variance, and skewness of each sequence are calculated to generate a eigenvector containing statistical features. The normalized sequence and eigenvector are then combined into a multidimensional matrix. This multidimensional matrix is then subjected to principal component analysis for dimensionality reduction, retaining principal components whose cumulative contribution reaches a preset ratio to obtain a performance characteristic matrix. This performance characteristic matrix characterizes the multidimensional performance distribution characteristics of the insulating adhesive. Specifically, the initial data set needs to be preprocessed, including removing outliers or noise, extracting eigenvectors for each parameter sequence, and calculating the mean, variance, and maximum value of the resistivity sequence to obtain a three-dimensional eigenvector such as [10^12, 10^2, 10^13]. Similarly, the frequency response slope and mean value may be extracted for the dielectric constant sequence, while the peak value and coefficient of variation may be extracted for the tensile strength sequence. After extracting the eigenvectors, they are combined into a matrix. For example, if there are 10 specimens, and eigenvectors for three parameters are extracted for each specimen, with each vector having three components, a 10×9 performance characteristic matrix can be constructed, where each row represents a characteristic of a specimen, and each column corresponds to a specific eigenvalue of a parameter. For example, if five silicone samples were tested, the resistivity eigenvector might be [mean 10^12, variance 10^3, maximum 10^13], the dielectric constant eigenvector [mean 3.5, slope 0.01, minimum 3.0], and the tensile strength eigenvector [mean 21, variance 2, peak 23]. Arranging these vectors in rows and expanding them into columns creates a 5×9 matrix. After constructing the matrix, principal component analysis (PCA) is used to further reduce dimensionality or verify data significance. The entire process is implemented using programming tools such as MATLAB or Python to improve computational efficiency and accuracy.

[0077] S3: performing a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set;

[0078] In step S3, the performance characteristic matrix is subjected to loading parameter mapping to determine the range of electric field and stress increments. Progressive electric field and mechanical stress loading experiments are performed on the insulating adhesive sample, and multi-dimensional response data of resistivity change, dielectric constant offset, and deformation are recorded in real time. The response data is then smoothed by Kalman filtering, and the influence of noise is eliminated by iteratively updating the state estimation to obtain a dynamic data set containing a sequence of multi-parameter changes in each loading stage. Specifically, the implementation process requires the design of a dynamic loading experiment plan. Periodic stress or electric field loading is applied to the sample through a servo control system. The loading conditions can be sinusoidal waveforms or step-by-step increments, depending on the application scenario of the insulating adhesive. The performance characteristic matrix is used as an input parameter to guide the selection of experimental conditions. If the matrix shows that the tensile strength characteristics of a sample are low, the loading frequency is reduced to avoid premature failure. During the experiment, the sample response data, such as the change in resistivity over time, the drift of the dielectric constant with frequency, or the decay of tensile strength with the number of cycles, are recorded in real time. These data constitute the dynamic response data set. For example, if a silicone specimen is subjected to 1000 tensile cycles, each with a tensile force of 15 MPa, the resistivity will decrease gradually from 10^12 Ω·m to 10^10 Ω·m, and the dielectric constant will change from 3.5 to 3.8, indicating possible microstructural damage within the material. Dynamic loading experiments require high-precision sensors and data acquisition systems to ensure the continuity and resolution of the response data. Furthermore, the experimental design should consider the influence of variables such as loading rate and ambient temperature. Ultimately, the dynamic response dataset is stored as a multidimensional time series.

[0079] S4: constructing a time series curve of the dynamic response data set, performing degradation trend fitting processing on the time series curve, and obtaining a degradation parameter set;

[0080] In step S4, the dynamic data set is processed to generate time series curves, constructing curves for resistivity, dielectric constant, and tensile strength. Each curve is then fitted with an exponential decay model. Fitting parameters are iteratively calculated using the least squares method. Adaptive weighting factors are then introduced based on the variance of each parameter change to adjust the fitting process. This yields a set of degradation parameters, including the decay rates and stable value parameters for each performance indicator, revealing the material's stability over long-term use. The dynamic response data set is first organized into curves in chronological order. The resistivity versus loading cycle curve can be plotted using a line graph. Assuming that after 1000 cycles, the resistivity decreases from 10^12 Ω·m to 10^10 Ω·m, the curve exhibits an exponential decay trend. These time series curves are then fitted using a combination of linear regression, exponential decay, or polynomial fitting. The choice depends on the curve's shape. For example, an exponential model may be more suitable for resistivity decay, while a quadratic polynomial fit may be required for dielectric constant. Through fitting, a set of degradation parameters is extracted, including the decay rate, steady-state value, or inflection point time. Assuming that the resistivity fitting result is R(t)=10^12·e^(-0.002t), the decay rate of 0.002 becomes one of the degradation parameters. The fitting process needs to be completed with the help of mathematical software Origin or R language, and the goodness of fit is verified by residual analysis.

[0081] S5: performing a multi-dimensional evaluation on the degradation parameter set based on information entropy to obtain a comprehensive score;

[0082] In step S5, a scoring function is calculated for the degradation parameter set. A single score for each performance indicator is generated based on the decay rate and stability value. A multidimensional evaluation function with initial weights is designed to combine the single scores. The degradation parameter set is then subjected to information entropy analysis. Entropy values are calculated using a probability distribution and the weights of each indicator are dynamically adjusted. The adjusted weights are then applied to the evaluation function to obtain a comprehensive score, which represents the overall performance level of the insulating adhesive. Specifically, the information entropy of the degradation parameter set is calculated. This reflects the degree of dispersion of the parameter distribution. For example, if the resistivity decay rate fluctuates slightly between [0.002, 0.0018, 0.0022], the entropy value is low, indicating high degradation consistency. The entropy values of each parameter are weighted and combined. The weights can be determined based on application requirements. For example, resistivity can account for 40%, while dielectric constant and tensile strength can each account for 30%. The weighted summation yields the comprehensive score. Assuming the entropy values of five samples are [0.1, 0.15, 0.12, 0.09, 0.11], the weighted composite score might be [0.85, 0.78, 0.82, 0.88, 0.84]. If a silicone sample has low resistivity degradation entropy but high dielectric constant change entropy, its composite score might be low, reflecting uneven performance. The entire evaluation process requires a clear entropy calculation formula and ensured data standardization. This entire step achieves unified quantification of multidimensional parameters through information entropy.

[0083] S6: performing anomaly detection on the comprehensive score, performing contribution analysis on the comprehensive score after detection and the degradation parameter set, and iteratively dividing the analysis result into quality levels to obtain a quality judgment result.

[0084] In step S6, a threshold comparison process is performed on the comprehensive score, a quality threshold is set based on historical data statistics and anomalies are marked, a contribution analysis is performed on the degradation parameter set and the comprehensive score, and the relative influence ratio of each parameter on the score is calculated. The degradation parameter set and the comprehensive score are then input into the support vector machine classification model, and the quality level is divided iteratively through hyperplane optimization to obtain a quality judgment result, which includes anomaly marking and quality level classification.

[0085] Specifically, the overall score is first detected for anomalies, using a boxplot to identify outliers. For example, if 0.5 of the five scores is significantly lower, it is marked as an anomaly. Contribution analysis is then performed to calculate the contribution of each degradation parameter to the overall score. For example, the partial regression coefficient reveals that resistivity decay has the greatest impact on the score. Finally, the quality is iteratively graded, with score thresholds set to [0.9, 0.8, 0.7] corresponding to excellent, good, and fair. The thresholds are repeatedly adjusted and the classification results verified to ultimately determine the sample quality.

[0086] In one embodiment, the step of synchronously collecting multi-dimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence, includes:

[0087] Perform multi-channel physical excitation on the insulating adhesive to be tested, and perform time stamp alignment processing on the excitation signal to obtain the original measurement signal;

[0088] Performing signal segmentation on the original measurement signal and performing multi-dimensional parameter conversion processing on the segmented signal to obtain preliminary performance data, wherein the preliminary performance data includes resistivity, dielectric constant and tensile strength;

[0089] Performing dynamic filtering on the preliminary performance data to obtain a smoothed data sequence, and performing statistical analysis on the smoothed data sequence based on a sliding window to obtain a filtered data set;

[0090] Segmenting the filtered data set into time series, and performing feature labeling processing on the segmentation results to obtain a segmented feature data set;

[0091] The segmented feature data sets are integrated in the order of timestamps to obtain an initial data set.

[0092] In the above embodiment, the insulating rubber sample to be tested is subjected to parallel processing of electric field excitation, dielectric response measurement, and mechanical tensile loading. The current response is recorded by applying a preset gradient voltage to generate an electric signal sequence, the dielectric response is recorded by high-frequency capacitance scanning to generate a capacitance signal sequence, and the deformation is recorded by gradually increasing the tensile stress to generate a stress signal sequence. The above signals are then timestamped and aligned to synchronize multi-channel data to obtain raw measurement signals, which include unprocessed current, capacitance, and stress time series. The raw measurement signals are then subjected to signal segmentation and feature calculation processing. The current series is converted into a resistivity series using Ohm's law, the capacitance series is converted into a dielectric constant series using a capacitance-dielectric model, and the stress series is converted into a tensile strength series using the stress-strain relationship. The converted series are then subjected to sampling rate normalization processing to unify the time resolution, thereby obtaining preliminary performance data including a preliminary numerical series of resistivity, dielectric constant, and tensile strength.

[0093] The preliminary performance data were subjected to wavelet transform decomposition, decomposing each sequence into high-frequency and low-frequency components. The high-frequency components were then subjected to adaptive threshold filtering to suppress noise, and the low-frequency components were reconstructed and merged to generate smooth curves. The filtering parameters were then dynamically adjusted based on the sampling point density of the preliminary performance data to produce smoothed data sequences, including denoised resistivity, dielectric constant, and tensile strength sequences. The smoothed data sequences were subjected to sliding window statistical analysis, with a window length set to seven consecutive sampling points. The mean and standard deviation within each window were calculated, and outliers were then marked and removed using the triple standard deviation criterion. The removed sequences were then interpolated and padded to maintain data continuity, resulting in a filtered data set containing the outlier-removed resistivity, dielectric constant, and tensile strength sequences.

[0094] The filtered data set was subjected to change point detection. Performance mutation points were determined by calculating the slope change rate of adjacent data points, and the sequence was segmented. Statistical features, including mean, peak, and trend slope, were then calculated for each segmented sequence. Time interval labels were then added to each sequence based on the segmented features to produce a segmented feature dataset, including labeled segmented sequences of resistivity, dielectric constant, and tensile strength. Multidimensional vector reconstruction was then performed on the segmented feature dataset, merging the segmented sequences of resistivity, dielectric constant, and tensile strength into a unified matrix in timestamp order. This matrix was then column-normalized to balance the dimensional differences between the parameters. Metadata tags, including sampling frequency and excitation parameters, were then added to the matrix based on the experimental conditions to produce the initial dataset.

[0095] In one example, the step of extracting the eigenvectors of the initial data set and constructing a matrix based on the eigenvectors to obtain a performance characteristic matrix includes:

[0096] Performing multi-scale hierarchical processing on the initial data set based on the resistivity sequence, the dielectric constant sequence, and the tensile strength sequence to obtain a hierarchical sequence group;

[0097] Performing nonlinear mapping on the layered sequence group to obtain a primary feature vector set, and performing weighted fusion processing on the layered sequence group based on the primary feature vector set to obtain a fused feature matrix;

[0098] Performing singular value decomposition on the fused feature matrix to obtain a reduced-dimensional feature matrix;

[0099] Based on the projection value of each dimension in the dimension-reduced feature matrix, performing residual correction processing on the primary feature vector set to obtain an optimized feature vector set;

[0100] The optimized feature vector set is subjected to tensor reconstruction processing to obtain a performance feature matrix.

[0101] In the above embodiment, based on the resistivity sequence, dielectric constant sequence, and tensile strength sequence in the initial data set, each sequence is decomposed by wavelet transform, and high-frequency and low-frequency components are extracted to generate corresponding multi-scale subsequences. All subsequences are then rearranged and combined according to the frequency level to obtain a hierarchical sequence group, which characterizes the distribution characteristics of the insulation adhesive performance parameters at different time scales. Kernel function mapping is performed on each subsequence in the hierarchical sequence group, and the subsequence is projected into a high-dimensional feature space using a Gaussian kernel function. Subsequently, the Euclidean distance and cosine similarity of each subsequence group in the high-dimensional space are calculated. The distance and similarity are combined to form a primary feature vector, resulting in a primary feature vector set, which characterizes the nonlinear correlation characteristics of the insulation adhesive performance.

[0102] Based on the Euclidean distance value of each eigenvector in the primary eigenvector set, dynamic weight coefficients are assigned to the corresponding subsequences in the hierarchical sequence group. A weighted average method is then used to fuse all subsequences to generate a fused sequence containing multidimensional performance parameters. The fused sequence is then reconstructed into a two-dimensional matrix based on the time and parameter dimensions to obtain a fused feature matrix, which characterizes the comprehensive distribution characteristics of the insulation adhesive's performance. Singular value decomposition is performed on the fused feature matrix to extract the first k singular values and their corresponding left and right singular vectors, where k is determined by the cumulative singular value contribution rate reaching a preset threshold. The fused feature matrix is then linearly projected based on the extracted singular vectors to generate a reduced-dimensional feature representation, resulting in a reduced-dimensional feature matrix, which characterizes the main variation patterns in the insulation adhesive's performance.

[0103] Based on the projection values of each dimension in the dimensionality-reduced feature matrix, residuals are calculated for the corresponding eigenvectors in the primary eigenvector set. The least squares method is used to fit the mapping relationship between the residuals and the projection values. The primary eigenvectors are then iteratively updated based on the mapping relationship to obtain an optimized eigenvector set. This optimized eigenvector set characterizes the multidimensional statistical characteristics of the insulating adhesive's performance. An outer product operation is performed on each eigenvector in the optimized eigenvector set to generate a corresponding feature tensor sub-block. All feature tensor sub-blocks are then stacked into a three-dimensional tensor according to the time dimension, parameter dimension, and statistical dimension. This three-dimensional tensor is then normalized to obtain a performance characteristic matrix, which characterizes the global distribution characteristics of the insulating adhesive's multidimensional performance.

[0104] In one embodiment, the step of performing a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set includes:

[0105] Performing loading parameter mapping processing on the performance characteristic matrix to obtain a loading condition sequence;

[0106] Performing a progressive multi-field loading experiment on the insulating adhesive to be tested according to the loading condition sequence to obtain an initial response data set;

[0107] Performing multi-scale time-frequency decomposition processing on the initial response data set to obtain a time-frequency feature component set;

[0108] Performing adaptive noise reduction processing on the initial response data set according to the time-frequency feature component set to obtain a smoothed response data set;

[0109] Performing state transition probability modeling on the smooth response data set, constructing a state transition relationship between parameters in the smooth response data set, and obtaining a dynamic evolution feature matrix;

[0110] The smooth response data set is subjected to multi-parameter fusion processing according to the dynamic evolution characteristic matrix to obtain a dynamic response data set.

[0111] In the above embodiment, a condition sequence suitable for dynamic loading experiments is generated using a performance characteristic matrix (containing multidimensional eigenvectors such as resistivity, dielectric constant, and tensile strength), ensuring that the loading range matches the material properties. Specifically, the statistical distribution of each eigenvector is extracted from the performance characteristic matrix. The increasing range of electric field intensity (e.g., 0 to 10 kV / mm) is determined based on the resistivity sequence, the increasing range of mechanical stress (e.g., 0 to 5 MPa) is determined based on the tensile strength sequence, and the loading time step (e.g., 0.1 second) is determined based on the dielectric constant sequence. These ranges are then discretized into increasing steps to generate the loading condition sequence.

[0112] A progressive multi-field loading experiment is conducted on the insulating adhesive to be tested based on the sequence of loading conditions. At this stage, the experiment needs to be completed under the coupling of multiple fields (such as thermal, mechanical, and electric fields) to simulate the complex working conditions of the insulating adhesive in actual application scenarios. In the experimental setting, it is necessary to accurately control the step-by-step changes of the loading conditions so that the test can be progressively advanced from low to high and from simple to complex, ensuring that the experimental results can capture the response characteristics under low load conditions and reflect the failure behavior under high load or extreme working conditions. During the experiment, the original response data of the insulating adhesive under the loading conditions are obtained through high-precision sensors and sampling devices, including multi-dimensional information such as time, displacement, temperature, and current, and finally form the initial response data set.

[0113] The initial response dataset was subjected to multiscale time-frequency decomposition to extract a set of time-frequency characteristic components. Signal decomposition techniques were used to reveal the multiscale patterns in the data. Discrete wavelet transforms (DWTs) were applied to the resistivity, dielectric constant, and deformation series, respectively. The Daubechies wavelet basis was selected and decomposed into five layers. Low-frequency trend components (such as the long-term downward trend in resistivity) and high-frequency detail components (such as instantaneous fluctuations in deformation) were obtained. The amplitude and phase of each component were extracted and integrated into a set of time-frequency characteristic components.

[0114] Based on the time-frequency characteristic component set, the initial response dataset is adaptively denoised to generate a smoothed response dataset. The amplitude of the high-frequency component is used to determine the noise threshold (for example, amplitudes less than 0.1 are considered noise). A Kalman filter is then applied to the data for iterative updates. This method suppresses the high-frequency noise in the resistivity series while preserving its main downward trend. Ultimately, a smoothed response dataset is output, eliminating random interference from the experiment. State transition probability modeling is then performed on this smoothed dataset. A hidden Markov model (HMM) is used to calculate the conditional probabilities between resistivity, dielectric constant, and deformation (for example, the probability of dielectric constant increasing when resistivity decreases is 0.8). A dynamic evolution characteristic matrix is constructed to reflect the dynamic dependencies between the parameters.

[0115] The smoothed response dataset is subjected to multi-parameter fusion processing based on the dynamic evolution characteristic matrix to generate the final dynamic response dataset. Using the transition probabilities of resistivity and dielectric constant as weights, the smoothed data is weighted and normalized, and its statistical distribution (such as mean and standard deviation) is calculated. This data is then integrated into a multidimensional dynamic response dataset that not only captures the variation sequence of each parameter but also reflects its evolutionary patterns, providing a comprehensive basis for performance evaluation of insulating adhesives.

[0116] In one embodiment, the step of constructing a time series curve of the dynamic response data set, performing degradation trend fitting processing on the time series curve, and obtaining a degradation parameter set includes:

[0117] Performing multidimensional time series decomposition processing on the dynamic response data set to obtain a primary time series group;

[0118] Performing curve smoothing processing on the primary time series group, and marking the smoothing results in segments to obtain a segmented trend series set;

[0119] Setting an initialization decay rate for each trend segment in the segmented trend sequence set, and iteratively optimizing each of the initialization decay rates based on an optimization algorithm to obtain an initial value set of fitting parameters;

[0120] Performing dynamic weight fitting processing on the segmented trend sequence set according to the initial value set of the fitting parameters to obtain an optimized time series curve;

[0121] Extracting degradation features of the optimized time series curve, and reorganizing the degradation features according to performance dimensions to obtain a degradation feature matrix;

[0122] A multi-scale residual analysis is performed on the degradation characteristic matrix, and the degradation characteristic matrix is optimized based on the distribution characteristics of the residual results to obtain a degradation parameter set.

[0123] In the above embodiment, the dynamic response data set is subjected to signal hierarchical decomposition processing, and the response data of resistivity, dielectric constant and tensile strength are extracted as independent time series respectively. Subsequently, each series is segmented in the frequency domain by short-time Fourier transform to separate the low-frequency trend component and the high-frequency fluctuation component. The low-frequency trend component is then reconstructed into a continuous curve along the time axis to obtain a primary time series group, which includes the low-frequency trend series of resistivity, dielectric constant and tensile strength.

[0124] The primary time series group was subjected to moving average filtering, and the sliding window length was set to 10 sampling points to smooth local fluctuations. Subsequently, trend turning points were detected based on the rate of change of the second-order derivative of each series. Each series was divided into multiple subintervals, and start and end time labels were added to each subinterval to obtain a segmented trend sequence set, which included time-labeled trend segments of resistivity, dielectric constant, and tensile strength.

[0125] The segmented trend series set is pre-fitted with an exponential decay model. The decay rate and stable value parameters are initialized for each trend segment. The optimal initial values for each segment are then calculated using the Newton iteration method. Constraints are introduced to adjust the parameter range based on slope continuity between segments, resulting in a set of initial fitting parameters. This set includes preliminary degradation parameters for each segment of resistivity, dielectric constant, and tensile strength. Based on this set of initial fitting parameters, the segmented trend series set is fitted using the weighted least squares method. The residual variance is calculated for each segment and an adaptive weight factor is assigned based on the variance. The fitting parameters are then iteratively updated until the residual converges. The fitted curves for each segment are then spliced in chronological order, and transitions at the junctions are smoothed to produce an optimized time series curve. This curve includes the continuously fitted degradation curves for resistivity, dielectric constant, and tensile strength.

[0126] The optimized time series curves were subjected to characteristic decomposition, and the decay rate, stable value, and inflection point time of each curve were calculated. Subsequently, multidimensional statistical analysis was performed on each parameter, and the skewness and kurtosis indices were introduced to characterize the degradation distribution characteristics. All characteristic parameters were then reorganized into a matrix form according to the performance dimension to obtain a degradation characteristic parameter matrix, which contained the degradation characteristic vectors of resistivity, dielectric constant, and tensile strength.

[0127] Multi-scale residual analysis is performed on the degradation characteristic parameter matrix. The characteristic parameters in the matrix are substituted into the long-period and short-period degradation models for verification. The parameter weights are then adjusted according to the distribution characteristics of the verification residuals. The confidence intervals of each parameter are iteratively optimized through Monte Carlo simulation. Finally, all refined parameters are integrated to obtain the degradation parameter set, which includes the final attenuation rate and stable value parameters of resistivity, dielectric constant, and tensile strength.

[0128] In another embodiment, the calculation expression of the above content is:

[0129] ;

[0130] in, is the degradation parameter set; To accumulate the results of all K trend segments; is the dynamic weight factor, the adaptive weight of the k-th trend segment; is an exponential decay term, indicating the degradation trend of the kth trend segment, is the decay rate of the segment, is a time variable; is the inverse Fourier transform term, represents the frequency domain signal of the kth segment, represents the inverse Fourier transform, which reconstructs the low-frequency trend component into a time series; It is the Hadamard product, element-by-element multiplication, which means the element-by-element multiplication of two matrices or vectors; is the second-order derivative term, It is the moving average filtering result with a sliding window length of 10, and the second-order derivative Calculate its rate of change; F represents the residual analysis and distribution characteristic optimization function, specifically , is the mean residual calculated by multi-scale residual analysis, reflecting the deviation of long-period and short-period model verification; is the residual adjustment coefficient, used to balance the residual variance Contributions to the formula, optimized through Monte Carlo simulation; is the residual variance of each segment in the fitting process, which measures the error distribution of the fitting curve; is the statistical distribution of the degenerate characteristic matrix, is the skewness of the degradation distribution, which characterizes the asymmetry of the degradation characteristics.

[0131] In one example, the step of performing a multi-dimensional evaluation on the degradation parameter set based on information entropy to obtain a comprehensive score includes:

[0132] Performing single feature extraction on the degradation parameter set to obtain a single performance indicator set, and performing probability distribution calculation on information entropy based on the single performance indicator set to obtain an entropy value sequence;

[0133] Setting an initial weight vector, and performing nonlinear mapping processing on the initial weight vector according to each entropy value in the entropy value sequence to obtain an optimized weight set;

[0134] Performing weighted fusion processing on the individual performance indicator sets according to the optimized weight set to obtain a preliminary evaluation score;

[0135] Performing residual analysis on the preliminary evaluation score to obtain an optimized evaluation score;

[0136] A comprehensive characterization function is constructed based on the optimization evaluation score, and the optimization evaluation score is subjected to comprehensive characterization processing according to the comprehensive characterization function to obtain a comprehensive score.

[0137] In the above embodiment, based on the time series data in the degradation parameter set, eigendecomposition is performed on the time series curve of each degradation parameter to extract characteristic parameters of the attenuation rate, stability value, and fluctuation amplitude, thereby obtaining a set of individual performance indicators, wherein the individual performance indicator set includes eigenvalues of the resistivity degradation indicator, the dielectric constant degradation indicator, and the tensile strength degradation indicator. Each eigenvalue in the individual performance indicator set is normalized, and a probability distribution function is calculated for each eigenvalue. Based on the probability distribution function, the information entropy is discretized to obtain an entropy value sequence, wherein the entropy value sequence represents the information uncertainty distribution of each degradation parameter.

[0138] Based on each entropy value in the entropy value sequence, a nonlinear mapping process is performed on the corresponding weight in the initial weight vector. An adjustment coefficient is calculated using the inverse relationship between the entropy value and the weight. The initial weight vector is iteratively updated using the adjustment coefficient to obtain an optimized weight set, wherein the optimized weight set reflects the contribution of each degradation parameter to the comprehensive evaluation. Based on the weight values in the optimized weight set, a weighted summation process is performed on the eigenvalues in the individual performance indicator set to construct a multidimensional evaluation function. The individual performance indicator set is linearly combined using the multidimensional evaluation function to obtain a preliminary evaluation score, wherein the preliminary evaluation score represents a preliminary quantitative result of the performance of the insulating adhesive.

[0139] Residual calculation processing is performed on the preliminary evaluation score and the expected value of the degradation parameter set to construct a residual distribution model. The preliminary evaluation score is iteratively corrected based on the residual distribution model. The corrected score sequence is smoothed and filtered to obtain an optimized evaluation score, wherein the optimized evaluation score has higher stability and consistency. Based on the optimized evaluation score, a normalized mapping processing is performed on the multidimensional degradation characteristics of the degradation parameter set to construct a comprehensive characterization function. The optimized evaluation score is nonlinearly transformed and integrated using the comprehensive characterization function to obtain a comprehensive score, wherein the comprehensive score represents the overall performance level of the insulating adhesive.

[0140] In one embodiment, the steps of performing abnormality detection on the comprehensive score, performing contribution analysis on the detected comprehensive score and the degradation parameter set, and iteratively classifying the analysis results into quality grades to obtain a quality determination result include:

[0141] Performing multi-scale sliding window processing on the comprehensive score to obtain a preliminary abnormal distribution sequence;

[0142] Performing dynamic anomaly detection on the comprehensive score according to the statistical distribution characteristics of the abnormal preliminary distribution sequence to obtain an abnormal label set;

[0143] Performing weighted cluster analysis on the abnormal points in the abnormal mark set according to the distribution density and abnormal degree of the abnormal points in the abnormal mark set to obtain an abnormal cluster feature vector;

[0144] Performing contribution decomposition on the degradation parameter set according to the abnormal clustering feature vector to obtain a parameter influence weight matrix;

[0145] Constructing a multidimensional projection transformation matrix based on the parameter influence weight matrix, and mapping the comprehensive score and degradation parameter set to a high-dimensional feature space through the multidimensional projection transformation matrix to obtain a quality feature projection set;

[0146] The quality feature projection set is iteratively optimized based on a support vector machine classification algorithm to obtain a quality determination result, which includes an abnormality mark and a multi-level quality classification label.

[0147] In the above embodiment, based on the statistical characteristics of the historical comprehensive score, a multi-scale sliding window is constructed, the time series of the comprehensive score is segmented and scanned and local features are extracted, the mean, variance and skewness in each window are calculated, and an abnormal preliminary distribution sequence is obtained, which reflects the local fluctuation characteristics of the comprehensive score.

[0148] Based on the statistical distribution characteristics of the initial anomaly distribution sequence, combined with an adaptive threshold algorithm, a dynamic anomaly detection threshold is calculated. The comprehensive score is filtered point by point, and outliers exceeding the threshold are marked to obtain an anomaly marker set, which includes the index of the outlier and the degree of anomaly. Based on the distribution density and degree of anomaly of the outliers in the anomaly marker set, the outliers are grouped and clustered using a weighted K-means clustering algorithm. The center position and weighted characteristic distance of each cluster are calculated to obtain an anomaly cluster feature vector, which represents the spatial aggregation pattern of the outliers.

[0149] Based on the cluster distribution characteristics of the anomaly clustering feature vectors, a mapping relationship model between the degradation parameter set and the comprehensive score is constructed. Using a partial least squares regression algorithm, the contribution of each parameter in the degradation parameter set is decomposed. The influence weight of each parameter on the anomaly cluster is calculated, resulting in a parameter influence weight matrix that reflects the relative contribution of each degradation parameter to the anomaly. Based on the weight distribution of the parameter influence weight matrix, a multidimensional projection transformation matrix is constructed to map the comprehensive score and degradation parameter set to a high-dimensional feature space. The feature distance and distribution density after projection are calculated to obtain a quality feature projection set that represents the joint quality characteristics of the comprehensive score and degradation parameters.

[0150] Based on the distribution characteristics of the quality feature projection set, a support vector machine (SVM) classification algorithm is employed. This algorithm iteratively optimizes a hyperplane to divide quality into different levels. This algorithm combines kernel function mapping and soft margin adjustment to calculate classification boundaries and divide quality intervals, resulting in a quality assessment result. This quality assessment result includes anomaly markers and multi-level quality classification labels. Specifically, a set of features extracted from the original data is repeatedly adjusted to a hyperplane (the boundary separating data in high-dimensional space) to achieve optimal discrimination between different quality categories. The resulting optimized classification result is used to assess data quality. If the data cannot be clearly separated by a line or plane in the original space, a kernel function (such as a Gaussian kernel or radial basis function) is used to map the data to a higher-dimensional space, making them separable in that space. Real-world data may contain noise or outliers, preventing complete linear separation. The soft margin allows for some data points to be misclassified, and a penalty parameter (such as the C parameter) is introduced to balance the optimization objectives of classification error and margin width. Simply put, quality assessment is performed using a support vector machine (SVM). Initially, a set of quality-related feature data is collected. The SVM then finds a "dividing line" (a hyperplane). This dividing line is repeatedly adjusted to best distinguish between different quality levels. If the data is difficult to separate, a kernel function is used to "pull" it into a space where it is easier to separate, while also allowing for some small errors (soft margins). Finally, the dividing line is calculated, classifying the quality into several intervals, and outliers and quality levels are then identified.

[0151] Reference Figure 2 The present invention also provides a testing device for insulating glue, comprising:

[0152] An acquisition module 100 is used to synchronously acquire multi-dimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence;

[0153] An extraction module 200 is configured to extract eigenvectors of the initial data set and construct a matrix based on the eigenvectors to obtain a performance characteristic matrix;

[0154] An experimental module 300 is configured to perform a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set;

[0155] A construction module 400 is used to construct a time series curve of the dynamic response data set, and perform degradation trend fitting processing on the time series curve to obtain a degradation parameter set;

[0156] An evaluation module 500 is configured to perform a multi-dimensional evaluation on the degradation parameter set based on information entropy to obtain a comprehensive score;

[0157] The determination module 600 is configured to perform anomaly detection on the comprehensive score, perform contribution analysis on the detected comprehensive score and the degradation parameter set, and iteratively classify the analysis results into quality levels to obtain a quality determination result.

[0158] In the above embodiment, by synchronously collecting multi-dimensional performance parameters of the insulating adhesive under test and constructing a performance characteristic matrix, the test data comprehensively reflects the performance of the insulating adhesive across different dimensions. Based on the performance characteristic matrix, dynamic loading experiments are conducted to generate a dynamic response dataset. This allows the testing method to not only focus on the initial performance indicators of the insulating adhesive but also capture the material's response characteristics under dynamic usage conditions, thereby more accurately reflecting its long-term stability. A time series curve of the dynamic response dataset is further constructed and fitted with degradation trends to form a degradation parameter set. This enables quantitative analysis of the insulating adhesive's performance trends and facilitates accurate durability prediction. Furthermore, information entropy is used to perform a multi-dimensional evaluation of the degradation parameter set to generate a comprehensive score. This allows quality assessment to evaluate the overall quality level of the insulating adhesive through the integration of multi-dimensional information, improving the scientific nature and accuracy of the test. Furthermore, anomaly detection-based analysis of the comprehensive score, combined with the contribution of the degradation parameter set, effectively identifies potential quality risks and avoids misjudgments due to localized indicator anomalies. Furthermore, through iterative quality grading, the quality assessment results can more effectively guide the selection and application of insulating adhesives, improving the reliability and service life of insulating adhesives in actual projects.

[0159] Reference Figure 3 In the embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as a test method for insulating glue. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a test method for insulating glue is implemented.

[0160] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a testing method for an insulating adhesive, comprising: synchronously collecting multidimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence; extracting the eigenvectors of the initial data set, and constructing a matrix based on the eigenvectors to obtain a performance characteristic matrix; performing a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set; constructing a time series curve of the dynamic response data set, performing degradation trend fitting processing on the time series curve to obtain a degradation parameter set; performing a multidimensional evaluation of the degradation parameter set based on information entropy to obtain a comprehensive score; performing anomaly detection on the comprehensive score, performing contribution analysis on the detected comprehensive score and the degradation parameter set, and iteratively dividing the analysis results into quality grades to obtain a quality judgment result.

[0161] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media provided herein and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0162] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for testing insulating adhesive, characterized in that: include: Synchronously collecting multi-dimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence; Extracting eigenvectors of the initial data set, and constructing a matrix based on the eigenvectors to obtain a performance characteristic matrix; Performing a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set; Constructing a time series curve of the dynamic response data set, performing degradation trend fitting processing on the time series curve, and obtaining a degradation parameter set; Performing a multi-dimensional evaluation on the degradation parameter set based on information entropy to obtain a comprehensive score; Performing anomaly detection on the comprehensive score, performing contribution analysis on the comprehensive score after detection and the degradation parameter set, and iteratively dividing the analysis results into quality grades to obtain a quality judgment result; The step of extracting the eigenvectors of the initial data set and constructing a matrix based on the eigenvectors to obtain a performance characteristic matrix includes: Performing multi-scale hierarchical processing on the initial data set based on the resistivity sequence, the dielectric constant sequence, and the tensile strength sequence to obtain a hierarchical sequence group; Performing nonlinear mapping on the layered sequence group to obtain a primary feature vector set, and performing weighted fusion processing on the layered sequence group based on the primary feature vector set to obtain a fused feature matrix; Performing singular value decomposition on the fused feature matrix to obtain a reduced-dimensional feature matrix; Based on the projection value of each dimension in the dimension-reduced feature matrix, performing residual correction processing on the primary feature vector set to obtain an optimized feature vector set; Performing tensor reconstruction processing on the optimized feature vector set to obtain a performance feature matrix; The step of performing a multi-dimensional evaluation on the degradation parameter set based on information entropy to obtain a comprehensive score includes: Performing single feature extraction on the degradation parameter set to obtain a single performance indicator set, and performing probability distribution calculation on information entropy based on the single performance indicator set to obtain an entropy value sequence; Setting an initial weight vector, and performing nonlinear mapping processing on the initial weight vector according to each entropy value in the entropy value sequence to obtain an optimized weight set; Performing weighted fusion processing on the individual performance indicator sets according to the optimized weight set to obtain a preliminary evaluation score; Performing residual analysis on the preliminary evaluation score to obtain an optimized evaluation score; A comprehensive characterization function is constructed based on the optimization evaluation score, and the optimization evaluation score is subjected to comprehensive characterization processing according to the comprehensive characterization function to obtain a comprehensive score.

2. The method for testing insulating adhesive according to claim 1, characterized in that: The step of synchronously collecting multi-dimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence, comprises: Perform multi-channel physical excitation on the insulating adhesive to be tested, and perform time stamp alignment processing on the excitation signal to obtain the original measurement signal; Performing signal segmentation on the original measurement signal and performing multi-dimensional parameter conversion processing on the segmented signal to obtain preliminary performance data, wherein the preliminary performance data includes resistivity, dielectric constant and tensile strength; Performing dynamic filtering on the preliminary performance data to obtain a smoothed data sequence, and performing statistical analysis on the smoothed data sequence based on a sliding window to obtain a filtered data set; Segmenting the filtered data set into time series, and performing feature labeling processing on the segmentation results to obtain a segmented feature data set; The segmented feature data sets are integrated in the order of timestamps to obtain an initial data set.

3. The method for testing insulating adhesive according to claim 1, characterized in that: The step of performing a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set includes: Performing loading parameter mapping processing on the performance characteristic matrix to obtain a loading condition sequence; Performing a progressive multi-field loading experiment on the insulating adhesive to be tested according to the loading condition sequence to obtain an initial response data set; Performing multi-scale time-frequency decomposition processing on the initial response data set to obtain a time-frequency feature component set; Performing adaptive noise reduction processing on the initial response data set according to the time-frequency feature component set to obtain a smoothed response data set; Performing state transition probability modeling on the smooth response data set, constructing a state transition relationship between parameters in the smooth response data set, and obtaining a dynamic evolution feature matrix; The smooth response data set is subjected to multi-parameter fusion processing according to the dynamic evolution characteristic matrix to obtain a dynamic response data set.

4. The method for testing insulating adhesive according to claim 1, characterized in that: The step of constructing a time series curve of the dynamic response data set, performing degradation trend fitting processing on the time series curve, and obtaining a degradation parameter set includes: Performing multidimensional time series decomposition processing on the dynamic response data set to obtain a primary time series group; Performing curve smoothing processing on the primary time series group, and marking the smoothing results in segments to obtain a segmented trend series set; Setting an initialization decay rate for each trend segment in the segmented trend sequence set, and iteratively optimizing each of the initialization decay rates based on an optimization algorithm to obtain an initial value set of fitting parameters; Performing dynamic weight fitting processing on the segmented trend sequence set according to the initial value set of the fitting parameters to obtain an optimized time series curve; Extracting degradation features of the optimized time series curve, and reorganizing the degradation features according to performance dimensions to obtain a degradation feature matrix; A multi-scale residual analysis is performed on the degradation characteristic matrix, and the degradation characteristic matrix is optimized based on the distribution characteristics of the residual results to obtain a degradation parameter set.

5. The method for testing insulating adhesive according to claim 1, characterized in that: The step of performing abnormality detection on the comprehensive score, performing contribution analysis on the detected comprehensive score and the degradation parameter set, and iteratively dividing the analysis results into quality grades to obtain a quality determination result includes: Performing multi-scale sliding window processing on the comprehensive score to obtain a preliminary abnormal distribution sequence; Performing dynamic anomaly detection on the comprehensive score according to the statistical distribution characteristics of the abnormal preliminary distribution sequence to obtain an abnormal label set; Performing weighted cluster analysis on the abnormal points in the abnormal mark set according to the distribution density and abnormal degree of the abnormal points in the abnormal mark set to obtain an abnormal cluster feature vector; Performing contribution decomposition on the degradation parameter set according to the abnormal clustering feature vector to obtain a parameter influence weight matrix; Constructing a multidimensional projection transformation matrix based on the parameter influence weight matrix, and mapping the comprehensive score and degradation parameter set to a high-dimensional feature space through the multidimensional projection transformation matrix to obtain a quality feature projection set; The quality feature projection set is iteratively optimized based on a support vector machine classification algorithm to obtain a quality determination result, which includes an abnormality mark and a multi-level quality classification label.

6. A testing device for insulating glue, characterized in that: include: An acquisition module is used to synchronously acquire multi-dimensional performance parameters of the insulating adhesive to be tested to obtain an initial data set, wherein the initial data set includes a resistivity sequence, a dielectric constant sequence, and a tensile strength sequence; An extraction module, configured to extract the eigenvectors of the initial data set and construct a matrix based on the eigenvectors to obtain a performance characteristic matrix; An experimental module, configured to perform a dynamic loading experiment on the insulating adhesive to be tested based on the performance characteristic matrix to generate a dynamic response data set; A construction module is used to construct a time series curve of the dynamic response data set, perform degradation trend fitting processing on the time series curve, and obtain a degradation parameter set; An evaluation module, configured to perform a multi-dimensional evaluation on the degradation parameter set based on information entropy to obtain a comprehensive score; a determination module, configured to perform anomaly detection on the comprehensive score, perform contribution analysis on the detected comprehensive score and the degradation parameter set, and iteratively classify the analysis results into quality grades to obtain a quality determination result; The extracting of the characteristic vectors of the initial data set and constructing a matrix based on the characteristic vectors to obtain a performance characteristic matrix includes: Performing multi-scale hierarchical processing on the initial data set based on the resistivity sequence, the dielectric constant sequence, and the tensile strength sequence to obtain a hierarchical sequence group; Performing nonlinear mapping on the layered sequence group to obtain a primary feature vector set, and performing weighted fusion processing on the layered sequence group based on the primary feature vector set to obtain a fused feature matrix; Performing singular value decomposition on the fused feature matrix to obtain a reduced-dimensional feature matrix; Based on the projection value of each dimension in the dimension-reduced feature matrix, performing residual correction processing on the primary feature vector set to obtain an optimized feature vector set; Performing tensor reconstruction processing on the optimized feature vector set to obtain a performance feature matrix; The multi-dimensional evaluation of the degradation parameter set based on information entropy to obtain a comprehensive score includes: Performing single feature extraction on the degradation parameter set to obtain a single performance indicator set, and performing probability distribution calculation on information entropy based on the single performance indicator set to obtain an entropy value sequence; Setting an initial weight vector, and performing nonlinear mapping processing on the initial weight vector according to each entropy value in the entropy value sequence to obtain an optimized weight set; Performing weighted fusion processing on the individual performance indicator sets according to the optimized weight set to obtain a preliminary evaluation score; Performing residual analysis on the preliminary evaluation score to obtain an optimized evaluation score; A comprehensive characterization function is constructed based on the optimization evaluation score, and the optimization evaluation score is subjected to comprehensive characterization processing according to the comprehensive characterization function to obtain a comprehensive score.

7. A computer device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and running on the processor, and the processor implements the method according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images