A stability evaluation method for chip varistors based on polarity reversal test

Abstract

The present invention discloses a stability evaluation method of a patch varistor based on polarity reversal test, which relates to the field of resistance evaluation. The stability evaluation method of the patch varistor based on polarity reversal test includes: S1, obtaining equipment data and test factors; S2, performing current detection and collecting current change values; S3, arranging the current change values ​​in time series to obtain a current fluctuation time domain set; S4, calculating the time domain set stability value and current floating threshold of the current fluctuation time domain set; S5, evaluating the stability level of the patch varistor according to the resistance stability threshold. The present invention collects and monitors current changes, and uses time series arrangement and time domain segmentation methods to accurately capture and analyze the pattern of current fluctuations, enhance the comprehensiveness of the test and the accuracy of the results, and help identify the stability and fluctuation behavior of the current through real-time data analysis, providing in-depth insights into the performance of the resistor, thereby providing data support for optimization.

Description

Technical Field

[0001] The present invention relates to the field of resistance evaluation, and in particular to a stability evaluation method for a chip varistor based on a polarity reversal test. Background Art

[0002] Polarity reversal testing is a test commonly used in electronic devices, especially in audio technology and signal processing, to detect whether the polarity of the signal will be reversed when the device or system processes the signal. Signal polarity reversal means that the positive and negative levels of the signal are swapped. In electronic signal processing, polarity reversal testing ensures that during circuit design and debugging, the signal processing does not accidentally reverse the polarity of the signal, ensuring the integrity of the signal and the correctness of the function.

[0003] Stability assessment of chip varistors is critical to ensuring that electronic devices maintain performance and safety during long-term use. A varistor is a resistor whose resistance changes with voltage. It is usually used to protect circuits from voltage spikes or overvoltage. Chip varistors protect sensitive components in circuits, such as transistors and IC chips, by absorbing possible voltage spikes. Stability assessment ensures that these varistors can operate normally within the expected voltage range and will not cause damage to the entire circuit due to failure.

[0004] However, the existing chip varistor stability assessment method based on polarity reversal test does not consider the time domain fluctuation analysis of current data when it is used, which makes it impossible to accurately assess the stability of the chip varistor when it is used. At the same time, the existing chip varistor stability assessment method based on polarity reversal test cannot accurately capture and analyze the current fluctuation pattern, resulting in the chip varistor stability assessment method based on polarity reversal test not being ideal when it is used.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In view of the problems in the related art, the present invention proposes a stability evaluation method for a chip varistor based on a polarity reversal test to overcome the above-mentioned technical problems existing in the existing related art.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] A method for evaluating the stability of a chip varistor based on a polarity reversal test comprises the following steps:

[0009] S1. Obtain device data of chip varistor and test factors in polarity reversal test;

[0010] S2. Perform current detection on the chip varistor according to the test factors in the polarity reversal test, and collect the current change value of the chip varistor device data during the current detection process;

[0011] S3, arranging the current change values ​​in time series, analyzing the fluctuation time threshold of the current change values ​​according to the time series arrangement result, and performing time domain segmentation on the current change value time series arrangement result based on the fluctuation time threshold to obtain a current fluctuation time domain set;

[0012] S4, comparing the current fluctuation time domains in the current fluctuation time domain set, calculating the time domain set stability value of the current fluctuation time domain set based on the comparison result, and calculating the current floating threshold value of the current change value in each current fluctuation time domain;

[0013] S5, summarizing the time domain set stability value and the current floating threshold, setting the resistance stability threshold based on the summary result, and evaluating the stability level of the chip varistor according to the resistance stability threshold;

[0014] The current change value is arranged in time series, the fluctuation time threshold of the current change value is analyzed according to the time series arrangement result, and the time domain is segmented based on the fluctuation time threshold to obtain the current fluctuation time domain set, including:

[0015] S31, performing data cleaning on the current change values, adding timestamps to the cleaned current change values, and sorting the current change values ​​in chronological order according to the timestamps of the current change values;

[0016] S32, based on the time series arrangement result, analyzing the current fluctuation data of the current change value by using a moving average algorithm, and setting a fluctuation time threshold according to the current fluctuation data;

[0017] S33, dividing the current change value time series arrangement result according to the fluctuation time threshold value to obtain the current fluctuation time domain, summarizing the current fluctuation time domain, and obtaining the current fluctuation time domain set.

[0018] As a preferred solution, based on the time series arrangement result, analyzing the current fluctuation data of the current change value by a moving average algorithm, and setting the fluctuation time threshold according to the current fluctuation data includes the following steps:

[0019] S321, setting a moving average window, and calculating a moving average of the time series arrangement results based on the moving average window;

[0020] S322, summarizing the moving average values ​​to obtain a moving average sequence, and comparing the moving average sequence with the time series arrangement result, and analyzing the fluctuation characteristic value of the moving average sequence based on the comparison result;

[0021] S323. Preset a fluctuation threshold rule, set a fluctuation time threshold based on the fluctuation characteristic value and the fluctuation threshold rule, and verify and optimize the fluctuation time threshold.

[0022] As a preferred solution, a moving average window is set, and the calculation formula for calculating the moving average of the time series arrangement result based on the moving average window is:

[0023] ;

[0024] in, It is the moving average of the time series arrangement results at time point t;

[0025] is the current change value at time point t in the time series arrangement result;

[0026] is the current change value at time point t-1 in the time series arrangement result;

[0027] For the time series results, in t-( N -1) Current change value at a time point;

[0028] N is the window value of the moving average window.

[0029] As a preferred solution, the moving average values ​​are summarized to obtain a moving average sequence, and the moving average sequence is compared with the time series arrangement result. The fluctuation characteristic value of the moving average sequence is analyzed based on the comparison result, including the following steps:

[0030] S3221, denoising the moving average value, summarizing the denoised moving splicing values ​​according to the time series, and obtaining a moving average sequence;

[0031] S3222, comparing the moving average sequence with the time series arrangement result, and analyzing the current deviation value between the moving average value in the moving average sequence and the current change value in the time series arrangement result;

[0032] S3223. Set a fluctuation characteristic rule, and extract a fluctuation characteristic value of the current deviation value based on the fluctuation characteristic rule.

[0033] As a preferred solution, comparing the moving average sequence with the time series arrangement result, and analyzing the current deviation value between the moving average value in the moving average sequence and the current change value in the time series arrangement result includes the following steps:

[0034] S32221. Based on the comparison result of the moving average sequence and the time series arrangement result, calculate the moment deviation value of the moving average value and the current change value at the same time point;

[0035] S32222. Identify the deviation characteristic value of the moment deviation value by using a statistical indicator algorithm according to the moment deviation value;

[0036] S32223. Setting deviation adjustment rules, and optimizing and adjusting the moment deviation value based on the deviation adjustment rules and the deviation characteristic value;

[0037] S32224. Summarize the optimized and adjusted moment deviation values ​​and output them as current deviation values.

[0038] As a preferred solution, comparing the current fluctuation time domains in the current fluctuation time domain set, calculating the time domain set stability value of the current fluctuation time domain set based on the comparison result, and calculating the current floating threshold value of the current change value in each current fluctuation time domain includes the following steps:

[0039] S41, analyzing the fluctuation similarity value and the fluctuation difference value of the current fluctuation time domain according to the current fluctuation time domain in the current fluctuation time domain set;

[0040] S42, calculating the current floating threshold of the time domain set stability value and the current change value of the current fluctuation time domain set based on the fluctuation similarity value and the fluctuation difference value;

[0041] S43, verifying and adjusting the time domain set stable value and the current floating threshold, and outputting the verified and adjusted time domain set stable value and the current floating threshold.

[0042] As a preferred solution, according to the current fluctuation time domain in the current fluctuation time domain set, analyzing the fluctuation similarity value and the fluctuation difference value of the current fluctuation time domain includes the following steps:

[0043] S411, setting a current fluctuation extraction rule, and extracting a fluctuation characteristic value in the current fluctuation time domain based on the current fluctuation extraction rule;

[0044] S412, presetting a correlation model library and a model matching rule, matching the fluctuation characteristic value with the correlation model in the correlation model library according to the model matching rule, and calculating the fluctuation similarity value in the current fluctuation time domain based on the matched correlation model;

[0045] S413, based on the fluctuation characteristic value, calculating the fluctuation difference value between different current fluctuation time domains by using a distance algorithm;

[0046] S414, verify and optimize the fluctuation similarity value and the fluctuation difference value, and output the fluctuation similarity value and the fluctuation difference value after verification and optimization.

[0047] As a preferred solution, based on the fluctuation similarity value and the fluctuation difference value, calculating the current floating threshold of the time domain set stable value and the current change value of the current fluctuation time domain set includes the following steps:

[0048] S421, calculating the fluctuation average value of the fluctuation similarity value by a weighted average algorithm, and taking the fluctuation average value as the time domain set stable value;

[0049] S422, calculating the percentile value of the fluctuation difference value by using a standard deviation algorithm, and generating a current floating threshold value based on the percentile value;

[0050] S423, setting a stable value classification rule and a floating threshold rule, and adding classification labels to the time domain set stable value and the current floating threshold according to the stable value classification rule and the floating threshold rule;

[0051] S424, verifying the time domain set stability value and the current floating threshold value after adding the graded annotation, and outputting the verified time domain set stability value and the current floating threshold value.

[0052] As a preferred solution, the weighted average algorithm is used to calculate the fluctuation average of the fluctuation similarity value, and the fluctuation average is used as the calculation formula of the time domain set stability value:

[0053] ;

[0054] Among them, W is the fluctuation average value of the fluctuation similarity value;

[0055] V is the total number of fluctuation similarity values;

[0056] is the fluctuation similarity value of the i-th fluctuation similarity value;

[0057] is the weight value of the ith fluctuation similarity value.

[0058] As a preferred solution, the time domain set stability value and the current floating threshold are summarized, and the resistance stability threshold is set based on the summary result. The stability level of the patch varistor is evaluated according to the resistance stability threshold, including the following steps:

[0059] S51, integrating the time domain set stable value and the current floating threshold value to generate a summary target, and summarizing the time domain set stable value and the current floating threshold value after data integration based on the summary target;

[0060] S52, generating a resistance stability threshold based on the summary result of the time domain set stability value and the current floating threshold, and dividing the stability level of the resistance stability threshold;

[0061] S53, comparing the resistance stability threshold with the current change value of the chip varistor equipment data, and judging and evaluating the stability level of the chip varistor based on the comparison result.

[0062] The beneficial effects of the present invention are:

[0063] 1. The present invention collects and monitors current changes and uses time series arrangement and time domain segmentation methods to accurately capture and analyze current fluctuation patterns, enhance the comprehensiveness of the test and the accuracy of the results, and help identify the stability and fluctuation behavior of the current through real-time data analysis, providing in-depth insights into resistor performance, thereby providing data support for optimization.

[0064] 2. The present invention identifies fluctuation characteristics through a moving average algorithm, sets a reasonable fluctuation time threshold, promptly detects current anomalies, prevents possible equipment failures, obtains detailed current fluctuation data, adjusts the design and production process, ensures the stable performance of the resistor component, and simultaneously ensures the stable operation of the varistor under various working conditions by accurately calculating the stable value of the current fluctuation time domain set and the current floating threshold, thereby improving the safety and reliability of the entire device.

[0065] 3. The present invention combines modern signal processing technology and data analysis methods to improve the scientificity and accuracy of the test, and through real-time monitoring and analysis, it enhances the understanding and control of the performance of the varistor, so that the stability of the varistor is not limited to the traditional static test, but is expanded to dynamic and real-time performance monitoring, thereby providing a more efficient and systematic technical support for the design, manufacture and quality control of electronic products. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0067] Figure 1 The present invention is a flowchart of a method for evaluating the stability of a chip varistor based on a polarity reversal test according to an embodiment of the present invention. DETAILED DESCRIPTION

[0068] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0069] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] According to an embodiment of the present invention, a method for evaluating the stability of a chip varistor based on a polarity reversal test is provided.

[0071] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, the chip varistor stability evaluation method based on polarity reversal test according to an embodiment of the present invention includes the following steps:

[0072] S1. Obtain device data of chip varistor and test factors in polarity reversal test;

[0073] Specifically, obtain complete technical specifications and product data sheets from the varistor manufacturer, including key information such as the resistor's electrical characteristics, size specifications, operating voltage and current range, and read and analyze these technical documents in detail to understand the performance and limitations of the resistor under different conditions. Then build a test platform including a power supply, signal generator, oscilloscope, and other necessary measuring instruments to perform polarity reversal tests on the varistor, and use a signal generator to generate specific test signals, such as sine waves or square waves, to simulate voltage spikes that may be encountered in actual applications.

[0074] In actual testing, observe and record the response of the varistor when the polarity is reversed. Use an oscilloscope to monitor and record the voltage across the resistor and the current through the resistor. Make sure to collect all key current and voltage data during the test, including the peak values ​​of current and voltage, waveform changes, etc. Then use an oscilloscope or other data recording device to analyze in detail the changes in current and voltage during the polarity reversal test, focusing on observing the changes in current peak values ​​and any abnormal waveforms. Based on the current and voltage data, evaluate the response of the varistor to polarity reversal, especially its ability and stability to change rapidly and dynamically. Then organize and analyze the collected data, use data analysis software or written scripts to process the data, look for regularity and anomalies in the waveform, and document the test results and analysis as a basis for subsequent evaluation and decision-making.

[0075] S2. Perform current detection on the chip varistor according to the test factors in the polarity reversal test, and collect the current change value of the chip varistor device data during the current detection process;

[0076] Specifically, ensure that the test environment meets safety standards and prepare all necessary test equipment, such as signal generator, oscilloscope, multimeter, power supply, etc. According to the test requirements, select the appropriate test signal, such as sine wave or square wave, and set the amplitude and frequency of the signal. Connect the chip varistor to the test circuit, ensure that the connection is correct, and connect it with other test equipment (such as oscilloscope), and apply the test signal to the circuit through the signal generator to simulate the voltage changes in the actual working environment.

[0077] Configure an oscilloscope or other data recording device to ensure that the current changes can be accurately captured and recorded, then start the test signal and begin collecting data to ensure that the current changes can be captured when the polarity reversal occurs. During the test, monitor the current changes in real time to observe whether the current waveform is reversed or abnormal. Record the waveform and data points of the current change, including key parameters such as the peak value, valley value, and reversal time of the current. Perform a detailed analysis of the collected current change data to evaluate the performance of the varistor under polarity reversal, and check whether there are any anomalies in the current waveform, such as overshoot, oscillation, or delay, which may be signs of poor performance of the varistor. Then record the test results in detail, including all relevant current change data and waveform diagrams. Based on the test results, prepare a test report, including test conditions, test process, test data, and conclusions.

[0078] S3, arranging the current change values ​​in time series, analyzing the fluctuation time threshold of the current change values ​​according to the time series arrangement result, and performing time domain segmentation on the current change value time series arrangement result based on the fluctuation time threshold to obtain a current fluctuation time domain set;

[0079] Specifically, the current change values ​​are arranged in time series, the fluctuation time threshold of the current change values ​​is analyzed according to the time series arrangement result, and the time series arrangement result of the current change values ​​is divided into time domains based on the fluctuation time threshold, and the current fluctuation time domain set is obtained, which includes:

[0080] S31, performing data cleaning on the current change values, adding timestamps to the cleaned current change values, and sorting the current change values ​​in chronological order according to the timestamps of the current change values;

[0081] Specifically, use a filtering algorithm to remove noise in the data, such as high-frequency noise or random noise, and identify and correct outliers, such as current mutations or non-physical current values, through statistical methods or manual inspection. If there are missing values ​​in the data, use interpolation methods, such as linear interpolation or spline interpolation, to estimate the missing data points and ensure that all data points are synchronized with a unified time base, including aligning the time of the data points with a known time source and assigning an accurate timestamp to each cleaned current change value. Then use a sorting algorithm (such as quick sort, merge sort, or heap sort) to sort the current change values ​​according to the timestamp. After sorting, check the continuity and stability of the data to ensure that the order of the time series is correct and no data points are omitted or repeated. Use charts or graphical tools (such as oscilloscopes or data visualization software) to visualize the sorted data to intuitively check the order and completeness of the data, and verify the accuracy and consistency of the data by calculating statistical indicators (such as mean, standard deviation, median).

[0082] S32, based on the time series arrangement result, analyzing the current fluctuation data of the current change value by using a moving average algorithm, and setting a fluctuation time threshold according to the current fluctuation data;

[0083] Specifically, based on the time series arrangement result, analyzing the current fluctuation data of the current change value by a moving average algorithm, and setting the fluctuation time threshold according to the current fluctuation data includes the following steps:

[0084] S321, setting a moving average window, and calculating a moving average of the time series arrangement results based on the moving average window;

[0085] Specifically, a moving average window is set, and a calculation formula for calculating the moving average of the time series arrangement result based on the moving average window is:

[0086] ;

[0087] in, It is the moving average of the time series arrangement results at time point t;

[0088] is the current change value at time point t in the time series arrangement result;

[0089] is the current change value at time point t-1 in the time series arrangement result;

[0090] For the time series results, in t-( N -1) Current change value at a time point;

[0091] N is the window value of the moving average window.

[0092] S322, summarizing the moving average values ​​to obtain a moving average sequence, and comparing the moving average sequence with the time series arrangement result, and analyzing the fluctuation characteristic value of the moving average sequence based on the comparison result;

[0093] Specifically, the moving average is summarized to obtain a moving average sequence, and the moving average sequence is compared with the time series arrangement result. The fluctuation characteristic value of the moving average sequence is analyzed based on the comparison result, including the following steps:

[0094] S3221, denoising the moving average value, summarizing the denoised moving splicing values ​​according to the time series, and obtaining a moving average sequence;

[0095] Specifically, select a suitable moving average window size, and the window size will affect the degree of smoothing and denoising effect. Perform moving average processing on the original data, and for each time point, take the average of the data points (points in the window) within a certain range before and after it as the moving average of that time point. Select a suitable denoising method, such as median filtering, wavelet transform or Kalman filtering, and apply the selected denoising algorithm to the moving average to remove noise and other interference. Then, in the denoised data, identify and remove any outliers or data points that do not meet expectations. If missing values ​​are generated during the denoising process, use interpolation methods (such as linear interpolation or spline interpolation) to estimate these values.

[0096] Arrange the denoised moving averages in chronological order to ensure the continuity and order of the data, and aggregate the arranged moving averages to form a continuous time series, namely the moving average sequence. Then, visualize the moving average sequence through charts or graphic tools (such as oscilloscopes or data visualization software) to visually check the order and integrity of the data. Then, calculate statistical indicators (such as mean, standard deviation, median) to verify the accuracy and consistency of the data. According to the verification results, adjust the denoising method and moving average window size to obtain the best denoising effect and moving average sequence.

[0097] S3222, comparing the moving average sequence with the time series arrangement result, and analyzing the current deviation value between the moving average value in the moving average sequence and the current change value in the time series arrangement result;

[0098] Specifically, comparing the moving average sequence with the time series arrangement result, and analyzing the current deviation value between the moving average value in the moving average sequence and the current change value in the time series arrangement result includes the following steps:

[0099] S32221. Based on the comparison result of the moving average sequence and the time series arrangement result, calculate the moment deviation value of the moving average value and the current change value at the same time point;

[0100] Specifically, ensure that the data points in the moving average sequence and the time series arrangement result are aligned according to the timestamp, so that each time point has a corresponding value in the two data sets. For each time point, find the value in the moving average sequence and the current change value in the time series arrangement result, calculate the difference between the two values, that is, the moment deviation value, through simple subtraction, such as moment deviation value = moving average value - current change value, and form the deviation values ​​of all moments into a sequence. This sequence reflects the deviation between the moving average value and the time series arrangement result within the entire time range. Check the deviation value sequence to ensure that there are no abnormal or erroneous data points. If abnormal deviation values ​​are found, further analyze the cause and consider whether correction or elimination is needed. Analyze the deviation value sequence to understand the trend and pattern of the overall deviation, and use the moment deviation value for subsequent analysis, such as extraction of fluctuation characteristic values ​​or calculation of current floating thresholds.

[0101] S32222. Identify the deviation characteristic value of the moment deviation value by using a statistical indicator algorithm according to the moment deviation value;

[0102] Specifically, ensure that there is a complete sequence of moment deviation values, and that this sequence contains the deviation values ​​of all time points. Calculate the basic statistics of the moment deviation value sequence, including the mean, median, standard deviation, minimum value, and maximum value. Then use histograms, box plots, and other charts to visualize the distribution and dispersion of the moment deviation values. According to the characteristics of the data and the purpose of analysis, select appropriate statistical indicator algorithms, including Z scores, quartiles, coefficient of variation, etc. Apply the selected statistical indicator algorithm to the moment deviation value sequence, calculate the statistical indicator value of each moment deviation value, and identify the deviation characteristic value of the moment deviation value based on the results of the algorithm calculation. For example, use Z scores to identify outliers or use quartiles to identify extreme values.

[0103] Verify whether the calculated deviation characteristic value is reasonable to ensure that it reflects the true characteristics of the moment deviation value sequence. If necessary, adjust the parameters of the statistical indicator algorithm according to the verification results to obtain more accurate deviation characteristic values. Use the identified deviation characteristic values ​​for further data analysis or decision-making, such as determining the stability of current changes or evaluating the performance of resistors.

[0104] S32223. Setting deviation adjustment rules, and optimizing and adjusting the moment deviation value based on the deviation adjustment rules and the deviation characteristic value;

[0105] Specifically, according to the purpose and needs of data analysis, determine the goals that need to be adjusted for the moment deviation value, including reducing the overall deviation, eliminating outliers or improving data stability, and then formulate specific deviation adjustment rules based on the statistical analysis results and deviation characteristic values. The rules should clearly state when and how to adjust the deviation value. When formulating the rules, consider the following factors: the amplitude of the deviation. The size of the deviation value determines the necessity of adjustment. The frequency of the deviation. Frequent deviations require different treatment methods. The pattern of the deviation. Identify the pattern of the deviation, such as periodicity or randomness, to determine the appropriate adjustment method. Set a fixed threshold. When the deviation value exceeds this threshold, adjust it and dynamically adjust the threshold according to the change of the data. For example, use a moving average or quartiles as the dynamic threshold, and then adjust it proportionally according to the size of the deviation value. Smaller deviations require smaller adjustments. Segment the data and formulate different adjustment rules according to different data segments.

[0106] Test the deviation adjustment rules on actual data to verify whether the adjusted data is more in line with expectations. Based on the test results, iteratively optimize the adjustment rules until a satisfactory deviation adjustment effect is achieved. Record all deviation adjustment processes and rule changes to ensure the traceability of the adjustments. Finally, generate a report summarizing the results and effects of the deviation adjustment.

[0107] S32224. Summarize the optimized and adjusted moment deviation values ​​and output them as current deviation values.

[0108] Specifically, according to the analysis requirements, a suitable aggregation method is selected, and the aggregation methods include summation, average, median, maximum value, minimum value, etc., and the optimized moment deviation values ​​are aggregated to obtain one or more aggregated values, and the aggregated values ​​are defined as current deviation values. For example, if summation is selected as the aggregation method, the current deviation value will be the sum of all moment deviation values. If necessary, the current deviation values ​​are further aggregated according to the time range (such as every hour, every day), and then the current deviation values ​​are formatted into the required format, such as a table, chart or report, and the current deviation values ​​are saved in an appropriate file or database and shared with other relevant personnel or systems. At the same time, it is verified whether the generated current deviation values ​​are accurate and reasonable, and a report is written to summarize the process and results of generating the current deviation values.

[0109] S3223. Set a fluctuation characteristic rule, and extract a fluctuation characteristic value of the current deviation value based on the fluctuation characteristic rule.

[0110] Specifically, according to the purpose and needs of data analysis, determine the fluctuation characteristics of the current deviation value that needs to be extracted, such as oscillation frequency, fluctuation amplitude, etc., and formulate specific fluctuation characteristic rules based on statistical analysis results and business needs. The rules should clearly define the calculation method and range of fluctuation characteristics. For example, the types of fluctuation characteristics include oscillation, periodicity, randomness, etc. The frequency range of fluctuations includes determining the time scale of fluctuations, such as milliseconds, seconds, minutes, etc. According to frequency domain analysis or periodicity detection methods, extract the main oscillation frequency and amplitude of the current deviation value, and calculate the amplitude range or standard deviation of the current deviation value to reflect its fluctuation amplitude. Use time series analysis methods to identify the periodicity, trend or mutation of the current deviation value.

[0111] Use programming languages ​​or data analysis tools to automatically apply fluctuation characteristic rules to extract the fluctuation characteristics of current deviation values. When necessary, perform manual analysis to confirm the extraction results of fluctuation characteristics. According to the established rules, calculate the fluctuation characteristic values ​​of each current deviation value, such as frequency, amplitude, period, etc., organize the extracted fluctuation characteristic values ​​into data sets or reports, verify whether the extracted fluctuation characteristic values ​​meet expectations and business needs, and adjust and optimize the fluctuation characteristic rules based on the verification results to improve the accuracy and reliability of the extraction results. Use the extracted fluctuation characteristic values ​​for further analysis, such as identifying abnormal fluctuations, evaluating system stability, etc. Apply the fluctuation characteristic analysis results to decision making or system optimization to improve system performance and stability.

[0112] S323. Preset a fluctuation threshold rule, set a fluctuation time threshold based on the fluctuation characteristic value and the fluctuation threshold rule, and verify and optimize the fluctuation time threshold.

[0113] Specifically, the purpose of the fluctuation threshold rule should be clarified. For example, in order to control the current fluctuation not to exceed a specific safety standard or to monitor the operating status of the equipment, specific threshold rules should be formulated according to the fluctuation characteristics of the current deviation value, such as oscillation frequency, fluctuation amplitude, etc., including the threshold value setting, fluctuation sensitivity, fluctuation frequency and amplitude, and expected response time. If the oscillation frequency exceeds a specific value, it indicates that the equipment is operating abnormally, and it is necessary to set an upper frequency threshold and an upper limit on the fluctuation amplitude to prevent the current from being too high or too low and affecting the equipment performance.

[0114] Based on the fluctuation threshold rule, the time threshold is further set. For example, if the current fluctuation exceeds the set amplitude threshold for more than 5 minutes continuously, an alarm is triggered and the time threshold is dynamically adjusted according to actual operating experience and data feedback. The effectiveness of the time threshold is tested using actual operating data to ensure that it can accurately reflect the fluctuation situation. According to the test results and operation feedback, the setting of the time threshold is adjusted to optimize its sensitivity and response time. The fluctuation time threshold rule is integrated into the monitoring system for real-time monitoring of current fluctuations, and the effectiveness of the threshold rule is regularly evaluated. Necessary adjustments are made according to changes in equipment performance or environmental changes, and all threshold settings and adjustment processes are recorded in detail to ensure the traceability of the adjustments. A report is then compiled to summarize the setting of the fluctuation time threshold and its impact on system operation.

[0115] S33, dividing the current change value time series arrangement result according to the fluctuation time threshold value to obtain the current fluctuation time domain, summarizing the current fluctuation time domain, and obtaining the current fluctuation time domain set.

[0116] Specifically, based on the previously set fluctuation time threshold, the time series arrangement of the current change value is segmented, and whenever the current change exceeds the threshold and lasts for more than the set time length, it is marked as the beginning and end of a fluctuation time domain, and each fluctuation time domain is identified according to the mark, and these time domains represent abnormal fluctuations of the current within a specific time. The fluctuation time domain of each mark is extracted from the time series arrangement, and all the identified fluctuation time domains are aggregated into a set, with each time domain as an element of the set, and the data format of these current fluctuation time domains is organized to ensure that the information of each time domain is clear and complete.

[0117] Analyze the summarized current fluctuation time domain set to identify common patterns or abnormal behaviors of current fluctuations. According to the analysis results, apply them to equipment maintenance decisions, performance monitoring, safety warning systems, etc., and verify whether the identification and aggregation of the fluctuation time domain are consistent with the actual current fluctuation performance. Then adjust the fluctuation time threshold and segmentation logic according to the verification results, optimize the identification accuracy of the fluctuation time domain, record the identification and aggregation process of the fluctuation time domain in detail, including all relevant parameters and threshold settings, and prepare a report summarizing the characteristics of the current fluctuation time domain set and how these characteristics affect the performance and safety of the system.

[0118] S4, comparing the current fluctuation time domains in the current fluctuation time domain set, calculating the time domain set stability value of the current fluctuation time domain set based on the comparison result, and calculating the current floating threshold value of the current change value in each current fluctuation time domain;

[0119] Specifically, comparing the current fluctuation time domains in the current fluctuation time domain set, calculating the time domain set stability value of the current fluctuation time domain set based on the comparison result, and calculating the current floating threshold value of the current change value in each current fluctuation time domain includes the following steps:

[0120] S41, analyzing the fluctuation similarity value and the fluctuation difference value of the current fluctuation time domain according to the current fluctuation time domain in the current fluctuation time domain set;

[0121] Specifically, according to the current fluctuation time domain in the current fluctuation time domain set, analyzing the fluctuation similarity value and the fluctuation difference value of the current fluctuation time domain includes the following steps:

[0122] S411, setting a current fluctuation extraction rule, and extracting a fluctuation characteristic value in the current fluctuation time domain based on the current fluctuation extraction rule;

[0123] Specifically, the goal of fluctuation extraction should be clarified, such as identifying the characteristics of current fluctuation frequency, amplitude, duration, etc. Based on the requirements of fluctuation characteristics, specific current fluctuation extraction rules should be formulated, and a minimum threshold for the fluctuation amplitude should be set. Only when the current change exceeds this threshold, it is considered a significant fluctuation. Frequency domain analysis tools (such as fast Fourier transform) should be used to identify the main frequency components of current fluctuations, and the shortest duration of fluctuations should be defined to ensure that only fluctuation events with sufficient duration are extracted.

[0124] Apply the above extraction rules to the identified current fluctuation time domain, extract the characteristic values ​​of the fluctuation, such as maximum amplitude, average amplitude, fluctuation frequency, total duration, etc., calculate the specific characteristic values ​​of each fluctuation time domain, such as calculating the average amplitude, or use statistical methods to determine the main frequency of the fluctuation, summarize the characteristic values ​​of all fluctuation time domains, and organize them into a table or database form, analyze the summarized fluctuation characteristic values, identify common fluctuation patterns or abnormal fluctuations, verify whether the extracted fluctuation characteristic values ​​accurately reflect the actual fluctuation of the current, and adjust the fluctuation extraction rules according to the feedback from actual applications and the accuracy of the results, and optimize the extraction effect of the fluctuation characteristics.

[0125] S412, presetting a correlation model library and a model matching rule, matching the fluctuation characteristic value with the correlation model in the correlation model library according to the model matching rule, and calculating the fluctuation similarity value in the current fluctuation time domain based on the matched correlation model;

[0126] Specifically, select or develop correlation models suitable for current fluctuation analysis, including statistical models, machine learning models or physical models, which are specifically used to describe and predict the behavior of current fluctuations. Parameterize each model to ensure that the model can cover different types of current fluctuation behaviors. Formulate matching rules to clarify how to select the most appropriate correlation model based on the fluctuation characteristic values ​​in the current fluctuation time domain. The rules should consider the statistical properties, frequency components, amplitude, etc. of the characteristic values ​​to ensure that the rules are flexible enough to adapt to different fluctuation conditions, while also being strict enough to ensure that the model selection is appropriate. Analyze the fluctuation characteristic values ​​extracted from the current fluctuation time domain to determine their main statistical and frequency domain characteristics. Compare these characteristic values ​​with the models in the model library to select the model that best reflects the current fluctuation behavior.

[0127] Using the selected correlation model, calculate the similarity between the current fluctuation time domain and the model-predicted fluctuation. This usually involves calculating the correlation coefficient, mean square error, etc. between the actual fluctuation data and the model prediction, selecting an appropriate similarity index to quantify the similarity between the current fluctuation time domain and the model, such as the Pearson correlation coefficient or cosine similarity, and then adjusting the model parameters or selecting a different model to improve the matching effect based on the calculation results of the fluctuation similarity value, testing the modified model to verify its effectiveness and accuracy in different current fluctuation time domains, preparing a detailed report describing the matching process, model selection, similarity calculation results, and their significance for current fluctuation analysis, and using the similarity value analysis results for further current fluctuation management decisions, such as adjusting equipment parameters and optimizing operation strategies.

[0128] S413, based on the fluctuation characteristic value, calculating the fluctuation difference value between different current fluctuation time domains by using a distance algorithm;

[0129] Specifically, the key step is to select a suitable distance algorithm, including Euclidean distance, Manhattan distance, cosine similarity and dynamic time warping, and then extract key fluctuation features from each current fluctuation time domain, such as peaks, troughs, average values, frequencies, etc. In order to ensure that the comparison of different feature values ​​is fair, all features are normalized so that they are compared on the same scale. Common methods include Z-score normalization or minimum-maximum normalization.

[0130] The selected distance algorithm is applied to calculate the distance between different current fluctuation time domains. For example, if the Euclidean distance is used, for each pair of fluctuation time domain feature vectors x and y, the distance d is calculated by the following formula:

[0131] ;

[0132] in, R is the number of eigenvalues, and is the value of the corresponding feature. The calculated distance value is used to analyze the differences between different current fluctuation time domains and identify which time domains are similar and which show significant differences. Based on these differences, further measures can be taken, such as adjusting the device settings to reduce unnecessary fluctuations, or more in-depth analysis of those time domains with significant differences.

[0133] S414, verify and optimize the fluctuation similarity value and the fluctuation difference value, and output the fluctuation similarity value and the fluctuation difference value after verification and optimization.

[0134] Specifically, use known benchmark data sets to verify the calculated fluctuation similarity and difference values. Compare the consistency of these calculated values ​​with the expected results, and perform statistical tests, such as hypothesis testing, to evaluate the statistical significance of the fluctuation similarity and difference values. Adjust key parameters in the distance algorithm, such as distance weights and normalization methods, and observe the impact of these changes on the fluctuation similarity and difference values. Perform sensitivity analysis to determine which parameters have the greatest impact on the final results. Use cross-validation techniques, such as K-fold cross-validation, to test the stability of the fluctuation similarity and difference values, and repeat the calculation process on different data subsets to ensure the consistency and reliability of the results.

[0135] Based on the results of validation and sensitivity analysis, adjust the distance calculation model, including modifying the distance formula, adjusting feature selection or improving data preprocessing methods. If thresholds are used to define similarities or differences, adjust these thresholds according to actual application needs to more accurately capture important fluctuation characteristics, conduct a detailed analysis of the optimized similarity and difference values, explain their performance and significance in different situations, organize and output the verified and optimized fluctuation similarity and difference values, and display the results through visualization methods such as charts and heat maps to make them easier to understand and use. Apply the optimized results to relevant equipment monitoring and maintenance decisions, such as adjusting operating parameters or preventive maintenance, and continuously monitor the performance of fluctuation similarity and difference values ​​to ensure their effectiveness and relevance in the long term.

[0136] S42, calculating the current floating threshold of the time domain set stability value and the current change value of the current fluctuation time domain set based on the fluctuation similarity value and the fluctuation difference value;

[0137] Specifically, based on the fluctuation similarity value and the fluctuation difference value, calculating the current floating threshold of the time domain set stable value and the current change value of the current fluctuation time domain set includes the following steps:

[0138] S421, calculating the fluctuation average value of the fluctuation similarity value by a weighted average algorithm, and taking the fluctuation average value as the time domain set stable value;

[0139] Specifically, the weighted average algorithm is used to calculate the fluctuation average of the fluctuation similarity value, and the fluctuation average is used as the calculation formula of the stable value of the time domain set:

[0140] ;

[0141] Among them, W is the fluctuation average value of the fluctuation similarity value;

[0142] V is the total number of fluctuation similarity values;

[0143] is the fluctuation similarity value of the i-th fluctuation similarity value;

[0144] is the weight value of the ith fluctuation similarity value.

[0145] S422, calculating the percentile value of the fluctuation difference value by using a standard deviation algorithm, and generating a current floating threshold value based on the percentile value;

[0146] Specifically, first collect or calculate a series of current fluctuation difference values, and calculate the average value (mean) and standard deviation of these difference values. The standard deviation is an indicator of the degree of data dispersion. The formula is:

[0147] ;

[0148] in, is the standard deviation, B is the number of data points, is each data point, is the mean value of the data. The standard deviation interval is used to estimate different percentiles. For example, if you want to know the 90% percentile, you need to find the corresponding z value in the standard normal distribution. According to the sensitivity and safety requirements of the system, select the appropriate percentile as the threshold for current floating. For example, if you want the threshold to cover most cases, select the 95% percentile. According to the selected percentile, use the standard deviation and mean above to calculate the threshold. Apply the calculated current floating threshold to the monitoring and control logic of the system to ensure that the current fluctuates within the normal operating range. Regularly review and analyze the effects of the threshold, adjust the threshold setting according to the actual operating conditions, and ensure that it reflects the latest operating experience and environmental conditions.

[0149] S423, setting a stable value classification rule and a floating threshold rule, and adding classification labels to the time domain set stable value and the current floating threshold according to the stable value classification rule and the floating threshold rule;

[0150] Specifically, a grading standard for stability values ​​is defined based on system performance and safety requirements. For example, the stability values ​​can be divided into very stable, stable, generally stable and unstable, and a specific numerical range is set for each level. For example, a very stable current value may refer to a fluctuation range of ±1%, stability is ±3%, and generally stable is ±5%, while values ​​exceeding these values ​​are considered unstable. Similar to the stability value, a floating threshold is set according to the actual floating situation of the current and the operating safety standards. For example, the floating threshold can be set to low risk (±10%), medium risk (±15%), high risk (±20%), etc. The current stability value data concentrated in the time domain is analyzed, and the corresponding stability level is marked for each time domain data point according to the defined stability value grading rules. This is processed automatically through programming, such as using conditional statements to check each value and assigning a corresponding level label.

[0151] Similarly, the current floating data is analyzed, and the risk level of each data point is marked according to the set floating threshold rules. Using data analysis tools or programming methods, each floating value is compared with the threshold and its risk level is marked. A detailed report is created to show the stable value and risk level of each time domain set and current floating data. This grading information is used to optimize the operating parameters of the system, perform preventive maintenance or develop emergency response plans, and regularly evaluate the effectiveness and accuracy of the grading annotations. The grading rules and thresholds are adjusted according to actual conditions, and system operation feedback can be used to optimize future grading standards and threshold settings to ensure that they meet actual needs.

[0152] S424, verifying the time domain set stability value and the current floating threshold value after adding the graded annotation, and outputting the verified time domain set stability value and the current floating threshold value.

[0153] Specifically, collect historical data or use known standards as benchmarks, determine indicators for verifying the accuracy of graded labeling, such as accuracy, recall, and F1 score, compare the labeled stable values ​​and floating thresholds with the benchmark data, check whether the classification meets the expected standards, and use statistical methods such as t-tests or ANOVA analysis to verify whether there are significant differences in the current stable values ​​and floating thresholds of different levels.

[0154] Analyze any deviations or errors found during the verification process, identify possible causes of errors, and adjust grading rules and threshold settings based on the verification results to ensure higher accuracy and reliability. Verify the adjusted annotations again to ensure that the modifications have improved the accuracy of the annotations. Perform multiple rounds of verification and adjustment to achieve optimal accuracy and results. Create a detailed report containing the verification process, results, and final annotation data, including details of the statistical analysis and any important findings. Use charts and graphs, such as bar charts, scatter plots, and heat maps, to display the annotation data and verification results.

[0155] S43, verifying and adjusting the time domain set stable value and the current floating threshold, and outputting the verified and adjusted time domain set stable value and the current floating threshold.

[0156] Specifically, clarify which performance indicators (such as accuracy, stability, and responsiveness) will be used to evaluate the effectiveness of stable values ​​and floating thresholds, decide to use statistical analysis software, simulation models, or field tests as verification tools, collect relevant data on stable values ​​and floating thresholds from actual operations, and then compare the collected data with the expected baseline or standard to check whether it meets the preset performance standards. Analyze the data to identify any discrepancies with expectations, such as stable values ​​or floating thresholds that are too high or too low, and conduct in-depth analysis of possible causes of the problem, such as equipment aging, environmental changes, or improper parameter settings.

[0157] Adjust the grading standards or floating threshold settings based on the verification findings. For example, it is necessary to modify the grading range of the stable value or adjust the sensitivity of the floating threshold. Update the algorithm or logic used to calculate the stable value and floating threshold. Re-verify the adjusted settings to ensure that the modification has achieved the expected improvement effect. Repeat the verification and adjustment process as needed until the results are stable and meet all operating standards. Record the verification process, problems found, measures taken and final results in detail, and prepare a detailed report, including comparisons before and after adjustments, reasons for adjustments and verification results. Output the adjusted and verified stable values ​​and floating thresholds in a structured data format (such as CSV or database table), and use charts and graphs to show the changes before and after the adjustment to facilitate understanding and communication.

[0158] S5, summarizing the time domain set stability value and the current floating threshold, setting the resistance stability threshold based on the summary result, and evaluating the stability level of the chip varistor according to the resistance stability threshold;

[0159] Specifically, the time domain set stability value and the current floating threshold are summarized, and the resistance stability threshold is set based on the summary result. The stability level of the chip varistor is evaluated according to the resistance stability threshold, including the following steps:

[0160] S51, integrating the time domain set stable value and the current floating threshold value to generate a summary target, and summarizing the time domain set stable value and the current floating threshold value after data integration based on the summary target;

[0161] Specifically, clarify the purpose of integrating data, such as whether it is to improve monitoring efficiency, optimize system performance, or for subsequent data analysis and decision support. Set specific summary goals based on the purpose, such as generating a unified report or creating a comprehensive data set to facilitate trend analysis or the establishment of a predictive model. Check the time domain set stable value and current floating threshold data and remove any errors or abnormal values. Ensure that all data formats are consistent, such as date and time format, value range, etc., to facilitate data integration, and design a data structure to accommodate the merged data, including all required fields, such as timestamp, device ID, stable value level, floating threshold level, etc. Use database query or data processing tools to merge stable value data and floating threshold data according to timestamps or other relevant key values.

[0162] Aggregate data as needed, such as calculating averages, maximums, minimums, or counts, to obtain more useful information, create summary tables or views that include all important statistical indicators and aggregated results, conduct in-depth analysis of the integrated data, explore trends, patterns, and potential problems in the data, present analysis results and aggregated data in the form of charts, graphs, and reports for easy understanding and sharing, ensure that the integrated and aggregated data accurately reflects the actual situation, verify by comparing with original data or other benchmark data, and adjust the methods or parameters for data integration and aggregation based on the verification results and user feedback.

[0163] S52, generating a resistance stability threshold based on the summary result of the time domain set stability value and the current floating threshold, and dividing the stability level of the resistance stability threshold;

[0164] Specifically, an in-depth analysis is performed on the aggregated time domain set stability values ​​and current floating thresholds to determine the correlation between these data and resistance stability. It is key to understand how current stability values ​​and floating thresholds affect resistance changes. Based on current stability and floating data, the resistance stability threshold is calculated. For example, regression analysis or other statistical methods are used to find the mathematical relationship between current stability and resistance value, and the resistance stability threshold is set accordingly. Different stability levels are defined based on the calculation results of the resistance stability threshold. For example, resistance stability is divided into four levels: very stable, stable, relatively unstable, and unstable. A specific resistance value range is set for each stability level. For example, very stable may mean that the resistance value fluctuates within ±1% of the set threshold, stable may be ±3%, and so on. Data analysis software or programming tools are used to establish a resistance stability threshold model based on the above definitions and data relationships, and the model should be able to predict the resistance stability level based on the input current stability value and floating threshold.

[0165] Verify the accuracy and reliability of the resistor stability threshold model through experimental data or historical data, check the consistency between the stability level predicted by the model and the actual observation results, adjust the model parameters or logic according to the verification results, improve the accuracy and practicality of the prediction, and apply the resistor stability threshold model to actual equipment and systems to monitor and adjust resistor-related operating conditions. Regularly check the performance of the model to ensure that it can accurately reflect the stability of the resistor, and update the model and stability level division in a timely manner according to changes in equipment and environment.

[0166] S53, comparing the resistance stability threshold with the current change value of the chip varistor equipment data, and judging and evaluating the stability level of the chip varistor based on the comparison result.

[0167] Specifically, determine the specific values ​​of the resistance stability threshold, which represent the dividing points of different stability levels, clarify the current change parameters in the chip varistor device data, such as the maximum current change, the average current change, etc., and then collect the current change data from the chip varistor device to ensure the accuracy and timeliness of the data. Perform necessary cleaning and formatting on the collected current data to make it suitable for subsequent analysis and comparison. Compare the resistance stability threshold with the current change data, and use an algorithm to calculate the difference between the current change and the threshold, for example, calculate whether the current change value is within the threshold range.

[0168] Analyze the difference between the current change value and the threshold to determine whether there is a situation that exceeds the predetermined threshold. According to the comparison result of the current change value and the stability threshold, judge the stability level of the chip varistor. For example, if the current change value is continuously within the threshold range, it is rated as high stability; if it often exceeds the range, it may be rated as low stability or unstable. Apply decision logic to classify the stability of the resistor, record the evaluation results in a database or report, including the stability level of each device and the related current change data, and generate a detailed evaluation report describing the current change of each chip varistor and its corresponding stability level, including any observed trends or abnormalities.

[0169] In summary, with the aid of the above-mentioned technical scheme of the present invention, the present invention collects and monitors current changes and uses timing arrangement and time domain segmentation methods to accurately capture and analyze the pattern of current fluctuations, thereby enhancing the comprehensiveness of the test and the accuracy of the results. It also helps identify the stability and fluctuation behavior of the current through real-time data analysis, provides in-depth insights into resistor performance, and thus provides data support for optimization.

[0170] In addition, the present invention identifies fluctuation characteristics through a moving average algorithm, sets a reasonable fluctuation time threshold, promptly detects current anomalies, prevents possible equipment failures, obtains detailed current fluctuation data, adjusts the design and production process, and ensures the stable performance of the resistor component. At the same time, by accurately calculating the stable value of the current fluctuation time domain set and the current floating threshold, the stable operation of the varistor under various working conditions is ensured, thereby improving the safety and reliability of the entire device.

[0171] In addition, the present invention combines modern signal processing technology and data analysis methods to improve the scientificity and accuracy of the test, and enhances the understanding and control of the performance of the varistor through real-time monitoring and analysis, so that the stability of the varistor is not limited to traditional static testing, but is expanded to dynamic and real-time performance monitoring, thereby providing a more efficient and systematic technical support for the design, manufacturing and quality control of electronic products.

[0172] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A stability evaluation method for a chip varistor based on polarity reversal test, characterized in that: The following steps are involved: S1. Obtain device data of chip varistor and test factors in polarity reversal test; S2. Perform current detection on the chip varistor according to the test factors in the polarity reversal test, and collect the current change value of the chip varistor device data during the current detection process; S3, arranging the current change values ​​in time series, analyzing the fluctuation time threshold of the current change values ​​according to the time series arrangement result, and performing time domain segmentation on the current change value time series arrangement result based on the fluctuation time threshold to obtain a current fluctuation time domain set; S4, comparing the current fluctuation time domains in the current fluctuation time domain set, calculating the time domain set stability value of the current fluctuation time domain set based on the comparison result, and calculating the current floating threshold value of the current change value in each current fluctuation time domain; S5, summarizing the time domain set stability value and the current floating threshold, setting the resistance stability threshold based on the summary result, and evaluating the stability level of the chip varistor according to the resistance stability threshold; The step of arranging the current change values ​​in time series, analyzing the fluctuation time threshold of the current change values ​​according to the time series arrangement result, and performing time domain segmentation on the current change value time series arrangement result based on the fluctuation time threshold to obtain the current fluctuation time domain set comprises: S31, performing data cleaning on the current change values, adding timestamps to the cleaned current change values, and sorting the current change values ​​in chronological order according to the timestamps of the current change values; S32, based on the time series arrangement result, analyzing the current fluctuation data of the current change value by using a moving average algorithm, and setting a fluctuation time threshold according to the current fluctuation data; The method of analyzing the current fluctuation data of the current change value by a moving average algorithm based on the time series arrangement result and setting the fluctuation time threshold according to the current fluctuation data includes the following steps: S321, setting a moving average window, and calculating a moving average of the time series arrangement results based on the moving average window; The calculation formula for setting the moving average window and calculating the moving average of the time series arrangement result based on the moving average window is: ； in, It is the moving average of the time series arrangement results at time point t; is the current change value at time point t in the time series arrangement result; is the current change value at time point t-1 in the time series arrangement result; For the time series results, in t-( N -1) Current change value at a time point; N is the window value of the moving average window; S322, summarizing the moving average values ​​to obtain a moving average sequence, and comparing the moving average sequence with the time series arrangement result, and analyzing the fluctuation characteristic value of the moving average sequence based on the comparison result; S323, presetting a fluctuation threshold rule, and setting a fluctuation time threshold based on the fluctuation characteristic value and the fluctuation threshold rule, and verifying and optimizing the fluctuation time threshold; S33, dividing the current change value time series arrangement result according to the fluctuation time threshold value to obtain the current fluctuation time domain, summarizing the current fluctuation time domain, and obtaining the current fluctuation time domain set.

2. The method for evaluating the stability of a chip varistor based on a polarity reversal test according to claim 1, characterized in that: The step of summarizing the moving average values ​​to obtain a moving average sequence, comparing the moving average sequence with the time series arrangement result, and analyzing the fluctuation characteristic value of the moving average sequence based on the comparison result includes the following steps: S3221, denoising the moving average value, summarizing the denoised moving splicing values ​​according to the time series, and obtaining a moving average sequence; S3222, comparing the moving average sequence with the time series arrangement result, and analyzing the current deviation value between the moving average value in the moving average sequence and the current change value in the time series arrangement result; S3223. Set a fluctuation characteristic rule, and extract a fluctuation characteristic value of the current deviation value based on the fluctuation characteristic rule.

3. The method for evaluating the stability of a chip varistor based on a polarity reversal test according to claim 2, characterized in that: The step of comparing the moving average sequence with the time series arrangement result and analyzing the current deviation value between the moving average value in the moving average sequence and the current change value in the time series arrangement result comprises the following steps: S32221. Based on the comparison result of the moving average sequence and the time series arrangement result, calculate the moment deviation value of the moving average value and the current change value at the same time point; S32222. Identify the deviation characteristic value of the moment deviation value by using a statistical indicator algorithm according to the moment deviation value; S32223. Setting deviation adjustment rules, and optimizing and adjusting the moment deviation value based on the deviation adjustment rules and the deviation characteristic value; S32224. Summarize the optimized and adjusted moment deviation values ​​and output them as current deviation values.

4. The method for evaluating the stability of a chip varistor based on a polarity reversal test according to claim 1, characterized in that: The steps of comparing the current fluctuation time domains in the current fluctuation time domain set, calculating the time domain set stability value of the current fluctuation time domain set based on the comparison result, and calculating the current floating threshold value of the current change value in each current fluctuation time domain include the following steps: S41, analyzing the fluctuation similarity value and the fluctuation difference value of the current fluctuation time domain according to the current fluctuation time domain in the current fluctuation time domain set; S42, calculating the current floating threshold of the time domain set stability value and the current change value of the current fluctuation time domain set based on the fluctuation similarity value and the fluctuation difference value; S43, verifying and adjusting the time domain set stable value and the current floating threshold, and outputting the verified and adjusted time domain set stable value and the current floating threshold.

5. The method for evaluating chip varistor stability based on polarity reversal test according to claim 4, characterized in that: The analyzing the fluctuation similarity value and the fluctuation difference value of the current fluctuation time domain according to the current fluctuation time domain in the current fluctuation time domain set comprises the following steps: S411, setting a current fluctuation extraction rule, and extracting a fluctuation characteristic value in the current fluctuation time domain based on the current fluctuation extraction rule; S412, presetting a correlation model library and a model matching rule, matching the fluctuation characteristic value with the correlation model in the correlation model library according to the model matching rule, and calculating the fluctuation similarity value in the current fluctuation time domain based on the matched correlation model; S413, based on the fluctuation characteristic value, calculating the fluctuation difference value between different current fluctuation time domains by using a distance algorithm; S414, verify and optimize the fluctuation similarity value and the fluctuation difference value, and output the fluctuation similarity value and the fluctuation difference value after verification and optimization.

6. The method for evaluating chip varistor stability based on polarity reversal test according to claim 4, characterized in that: The method of calculating the current floating threshold of the time domain set stable value and the current change value of the current fluctuation time domain set based on the fluctuation similarity value and the fluctuation difference value comprises the following steps: S421, calculating the fluctuation average value of the fluctuation similarity value by a weighted average algorithm, and taking the fluctuation average value as the time domain set stable value; S422, calculating the percentile value of the fluctuation difference value by using a standard deviation algorithm, and generating a current floating threshold value based on the percentile value; S423, setting a stable value classification rule and a floating threshold rule, and adding classification labels to the time domain set stable value and the current floating threshold according to the stable value classification rule and the floating threshold rule; S424, verifying the time domain set stability value and the current floating threshold value after adding the graded annotation, and outputting the verified time domain set stability value and the current floating threshold value.

7. The method for evaluating chip varistor stability based on polarity reversal test according to claim 6, characterized in that: The weighted average algorithm is used to calculate the fluctuation average value of the fluctuation similarity value, and the calculation formula for using the fluctuation average value as the stability value of the time domain set is: ； Among them, W is the fluctuation average value of the fluctuation similarity value; V is the total number of fluctuation similarity values; is the fluctuation similarity value of the i-th fluctuation similarity value; is the weight value of the ith fluctuation similarity value.

8. The method for evaluating chip varistor stability based on polarity reversal test according to claim 1, characterized in that: The step of summarizing the time domain set stability value and the current floating threshold, setting the resistance stability threshold based on the summary result, and evaluating the stability level of the chip varistor according to the resistance stability threshold includes the following steps: S51, integrating the time domain set stable value and the current floating threshold value to generate a summary target, and summarizing the time domain set stable value and the current floating threshold value after data integration based on the summary target; S52, generating a resistance stability threshold based on the summary result of the time domain set stability value and the current floating threshold, and dividing the stability level of the resistance stability threshold; S53, comparing the resistance stability threshold with the current change value of the chip varistor equipment data, and judging and evaluating the stability level of the chip varistor based on the comparison result.

Images