Microswitch contact life test method and system

Abstract

The invention relates to the technical field of microswitches, and discloses a microswitch contact life test method and system, and the method comprises the steps: carrying out the contact preheating operation and resistance reference measurement of a microswitch, and obtaining a contact resistance reference value and an initial fluctuation characteristic spectrum; setting a contact degradation monitoring threshold value based on the contact resistance reference value and the initial fluctuation characteristic spectrum; placing the microswitch in a test device for loop test, and recording a contact state data set in each test period; performing weighting processing on the contact state data set according to a test period, and calculating an average resistance fluctuation metric value and a contact degradation evaluation index; and performing dual-mode contact failure detection based on the average resistance fluctuation metric value and the contact degradation evaluation index, and generating target early warning signals of resistance abrupt change type and gradual change type failure modes, thereby improving the identification capability of different types of microswitch contact failure modes, and effectively avoiding sudden faults of equipment caused by contact failure.

Description

Technical Field

[0001] The present invention relates to the technical field of microswitches, and particularly to a method and system for testing the contact life of microswitches. Background Art

[0002] The existing methods for testing the contact life of microswitches mainly rely on a specified number of actuation tests or simple resistance measurements, which cannot detect early signs of contact deterioration in a timely manner, resulting in a long test cycle and an inability to predict the actual service life. Traditional testing methods usually treat all test cycles equally, ignoring the differential role of test cycles at different stages in the evaluation of contact performance, resulting in insufficient accuracy in predicting contact life.

[0003] Contact failure modes are diverse and complex, mainly including two types: resistance mutation type and gradual change type. However, existing testing methods are difficult to effectively detect these two failure modes simultaneously, and can often only detect abnormalities after the contacts have severely deteriorated, without providing an early warning mechanism. Due to the inevitable presence of electrical noise and contact bounce interference during the contact testing process, the measurement data fluctuates greatly, increasing the difficulty of analysis and resulting in insufficient validity of the test data. Summary of the Invention

[0004] The present invention provides a method and system for testing the contact life of microswitches, which improves the ability to identify different types of contact failure modes of microswitches and effectively avoids sudden failures of equipment caused by contact failures.

[0005] In a first aspect, the present invention provides a method for testing the contact life of a microswitch, the method for testing the contact life of the microswitch comprising: Performing a contact preheating operation and a resistance reference measurement on the microswitch to obtain a contact resistance reference value and an initial fluctuation characteristic spectrum; Setting a contact deterioration monitoring threshold based on the contact resistance reference value and the initial fluctuation characteristic spectrum; Placing the microswitch in a test device for cyclic testing, and recording a contact state data set in each test cycle according to the contact deterioration monitoring threshold; Performing a weighted processing on the contact state data set according to the test cycle, and calculating an average resistance fluctuation metric value and a contact deterioration evaluation index; Performing a dual-mode contact failure detection based on the average resistance fluctuation metric value and the contact deterioration evaluation index, and generating a target warning signal for resistance mutation type and gradual change type failure modes.

[0006] In a second aspect, the present invention provides a system for testing the contact life of a microswitch, the system for testing the contact life of the microswitch comprising: A reference measurement module for performing contact preheating operation and resistance reference measurement on a microswitch to obtain a contact resistance reference value and an initial fluctuation characteristic spectrum; A setting module for setting a contact deterioration monitoring threshold based on the contact resistance reference value and the initial fluctuation characteristic spectrum; A cyclic test module for placing the microswitch in a test device for cyclic testing and recording a contact state data set in each test cycle according to the contact deterioration monitoring threshold; A weighted processing module for performing weighted processing on the contact state data set according to the test cycle, and calculating an average resistance fluctuation metric value and a contact deterioration evaluation index; A failure detection module for performing dual-mode contact failure detection based on the average resistance fluctuation metric value and the contact deterioration evaluation index, and generating a target warning signal for resistance mutation type and gradual change type failure modes.

[0007] In the technical solution provided by the present invention, the present invention can achieve ultra-fast detection of resistance mutation type and gradual change type contact failure modes, greatly advancing the contact failure warning time, and effectively avoiding sudden failures of equipment caused by contact failures. The test cycle weight information is introduced and a weighted contact deterioration evaluation model is constructed. The cycle weight is adaptively learned through a weight optimization function based on the contact resistance change rate, significantly improving the accuracy of contact deterioration state evaluation and the precision of contact life prediction. The existing LC filter circuit is utilized and an innovative sample and hold circuit is integrated into the test device to achieve synchronous contact deterioration detection and ensure the validity of test data without changing the main function of the filter circuit, reducing the system cost and enhancing the system's ability to adapt to different working load conditions. The neighborhood rough set theory is applied to process the noise and uncertainty in the contact test data, and an optimal life prediction parameter set is generated through a forward addition heuristic algorithm, making the life prediction result have a larger reliability boundary and providing a more accurate time expectation for equipment maintenance. Through the dual-mode contact failure detection mechanism and adaptive threshold adjustment, the system's ability to identify different types of microswitch contact failure modes is significantly improved, with a wider applicable range and highly competitive performance in terms of test efficiency and prediction accuracy. Description of the Drawings

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0009] Figure 1 It is a step schematic diagram of the microswitch contact life test method in the embodiments of the present invention; Figure 2 This is a schematic structural diagram of a microswitch contact life test system in an embodiment of the present invention. Detailed implementation manners

[0010] The embodiments of the present invention provide a method and a system for testing the life of microswitch contacts. Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present invention are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order different from that shown or described herein. In addition, the term "comprising" or "having" and any deformation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0011] For ease of understanding, the specific process of the embodiments of the present invention will be described below. Please refer to Figure 1 , an embodiment of the method for testing the life of microswitch contacts in the embodiments of the present invention includes: Step S1: Perform contact preheating operation and resistance reference measurement on the microswitch to obtain a contact resistance reference value and an initial fluctuation characteristic spectrum; It can be understood that the execution subject of the present invention can be a microswitch contact life test system, or a terminal or a server. Specifically, no limitation is made here. The embodiments of the present invention will be described by taking the server as the execution subject as an example.

[0012] Specifically, a test environment is constructed to enable the microswitch to perform continuous opening and closing actions under controlled conditions. The system presets a fixed number of contact opening and closing cycle operations as the preheating stage, which is set to about 100 times to eliminate the instability of the initial mechanical contact of the microswitch. At the same time, the microscopic irregular structure on the contact surface gradually tends to be stable, ensuring the repeatability and representativeness of subsequent measurements. When the preheating process is completed, the contact resistance of the microswitch in the closed state is measured continuously multiple times. During this measurement process, the current and voltage are controlled. A stable excitation is provided by a 100 mA constant current source and a 5 V voltage source, and the measurement value is obtained through a high-speed and high-precision resistance sampling circuit. 50 consecutive data points are collected and executed at fixed time intervals to ensure that each measurement reflects the true electrical contact state. The multiple consecutive resistance measurement values are preliminarily screened and processed. By removing the abnormal extreme values in the data to reduce noise interference, the method of removing the highest and lowest 5 values each is used to retain 40 effective measurement values in the middle. Then, the 40 effective resistance values are statistically processed, and the reference value of the contact resistance is obtained by calculating their arithmetic mean. This value is used as the reference baseline for all subsequent resistance change analyses. According to these 40 effective measurement values, their standard deviation is calculated. The standard deviation is a key parameter for evaluating the fluctuation level of the initial state of the microswitch contact, reflecting the resistance stability and the consistency of the microscopic contact state of the contact under static conditions. After the microswitch completes the first test cycle, that is, the first 1000 opening and closing actions, the resistance measurement data corresponding to 100 consecutive closing operations are extracted from it. According to the set interval division rules, all resistance values fall into the corresponding discrete resistance intervals, and the occurrence frequency of each interval is counted to form the initial resistance fluctuation characteristic spectrum, that is, the resistance value interval frequency distribution matrix.

[0013] Step S2: Set the contact deterioration monitoring threshold based on the reference value of the contact resistance and the initial fluctuation characteristic spectrum; Specifically, three coefficient factors corresponding to three deterioration levels are preset, which are respectively used to characterize the slight, moderate, and severe performance degradation trends. Among them, the first coefficient factor is set to 1.2, the second coefficient factor is set to 1.5, and the third coefficient factor is set to 2.0. Multiply the reference value of the contact resistance by these three coefficient factors respectively to obtain the slight deterioration threshold, the moderate deterioration threshold, and the severe deterioration threshold in sequence. Perform interval division and frequency statistics on the resistance data recorded in the initial fluctuation characteristic spectrum, that is, map all initial closing resistance measurement values into resistance intervals with a fixed width, and count the frequency distribution of resistance values in each interval to obtain the frequency distribution model under the normal state of the contact. This distribution model is regarded as a probability density approximate expression reflecting the fluctuation behavior of the contact resistance in the healthy state, and can depict the intrinsic fluctuation range of the contact performance under the condition of no external disturbance. In order to quantify the deviation degree between the resistance fluctuation distribution in each subsequent test cycle and this normal distribution, the KL divergence is introduced as the core index for measuring the difference between distributions. Compare the resistance fluctuation distribution matrix recorded in the test cycle with the initial characteristic spectrum, and calculate the KL divergence value between the two element by element to obtain a numerical index describing the change speed of the contact state. The larger the KL divergence, the more significant the deviation of the contact behavior from the normal state during the test cycle, reflecting a stronger potential deterioration trend. Allocate weight values to the test cycle according to the slight deterioration threshold of the contact, the moderate deterioration threshold of the contact, the severe deterioration threshold of the contact, and the contact state change rate reference. Dynamically calculate the weight value of this cycle according to the deterioration level (that is, whether it exceeds the slight, moderate, or severe threshold) of the resistance measurement result in the current cycle and the change range of its KL divergence value relative to the initial reference value. Cycles within the normal range and with a low KL divergence will be given a higher weight, while cycles in the middle and late stages of the deterioration level or with a sudden increase in the KL divergence will be given a relatively low weight. A contact deterioration monitoring threshold setting mechanism with dynamic adaptability and multi-index fusion ability is formed.

[0014] Step S3: Place the microswitch in the test device for cyclic testing, and record the contact state data set in each test cycle according to the contact deterioration monitoring threshold; Specifically, install the micro switch to be tested in the central positioning module of the test device, and ensure that its installation posture conforms to the preset contact direction and load conditions. Continuously drive the micro switch according to the set action frequency, which is set to 20 times per minute. The action stroke and acting force are strictly controlled within the specified error range to form a stable-structured and continuous-operation opening and closing action sequence. The drive system synchronously records the action number and execution time of each operation in real time to ensure the consistency of subsequent measurement and timing tracking. On this basis, select the closed states in the continuous opening and closing sequence according to the preset sampling rule. The sampling rule is based on a random or equally spaced sampling strategy. Select 100 closed stable sections as measurement points from every 1000 complete actions to construct a contact closed state sequence and reduce the cumulative interference of the measurement process on the contacts themselves. Use a high-precision resistance detection module to collect the contact resistance at each closed moment. The sampling rate is 10 kHz. The sampling results are temporarily stored by the sample and hold circuit and then enter the data buffer module to form the original contact resistance measurement data. To ensure data quality, perform stability screening on the original resistance data according to the set contact degradation monitoring threshold. Only the measured values that fall within the stable test area, that is, the data with a change range within ±2% of the reference resistance, are regarded as valid measurement data. All data in the electrical disturbance interval or with transient resistance peaks will be marked as unstable samples and excluded from the effective statistical range. After the screening is completed, accurately bind all valid resistance data with the corresponding sampling timestamps, and synchronously read the environmental parameter information at that moment, including the temperature and humidity data in the test cavity, and record the corresponding cumulative number of operations to construct a multi-dimensional data structure of contact states including five dimensions: the number of operations, resistance value, relative resistance change rate, environmental conditions, and sampling time. Reorganize and group the data according to the test cycle. Each 1000 complete opening and closing operations form a test cycle. The system automatically extracts all the multi-dimensional data of contact states within this cycle and classifies them into the data block of the current cycle to form a dataset with a standardized structure. Each periodic dataset contains all valid contact resistance values and their environmental information, as well as the statistical indicators, time information, and corresponding sequence numbers within the cycle.

[0015] Step S4: Perform weighted processing on the contact state dataset according to the test cycle, and calculate the average resistance fluctuation metric value and the contact degradation evaluation index; Specifically, data within multiple consecutive test cycles is extracted from the contact state dataset. Structurally, this data consists of multiple subsets of cycle resistances. Each cycle resistance dataset corresponds to a complete cycle of 1000 opening and closing operations, containing all valid contact resistance measurement values within that cycle, along with their corresponding environmental parameters and operation timing information. For the resistance sequence in each test cycle, the deviation value relative to the average resistance of that cycle is calculated. The original characteristics of resistance fluctuations in that cycle are obtained through point-by-point calculation. These characteristics constitute highly sensitive characterization indicators for changes in the microscopic contact state of the contact, used to reflect the severity of resistance fluctuations and the overall dispersion level within that cycle. A conditional attribute set for the deteriorated state of the contact is constructed based on representative statistics extracted from the resistance sequence and organized into a standardized feature attribute matrix. Each row of this matrix corresponds to a test cycle and contains four key attributes: namely, the average resistance value of the cycle, the standard deviation of the resistance, the maximum resistance value, and the resistance change slope. The change slope is obtained by dividing the difference in average resistance values between cycles by the difference in the number of operations. This feature attribute matrix reflects the deterioration dynamics of the contact in the electrical performance dimension. The cycle weight vector is calculated based on the original characteristics of resistance fluctuations. The cycle weight vector is calculated through a preset weight optimization function, and this vector is used to adjust the contribution intensity of different cycles to the judgment of the overall deterioration trend. The weight calculation process comprehensively considers the amplitude of resistance fluctuations, the degree of resistance distribution offset, and the KL divergence distance from the initial state. A cycle with a higher weight value indicates that its resistance state is closer to the deterioration edge or in the transition critical region. The resistance deviation value is weighted and squared according to the weight value of each cycle, and the square root of the sum of weighted deviation squares is taken to obtain the weighted average resistance fluctuation metric value. The feature attribute matrix is weighted based on the cycle weight vector. Based on the neighborhood rough set theory, by defining the decision attribute as the contact deterioration state level (such as normal, slight, moderate, severe) and using the weighted feature attributes as the conditional set, a weighted Euclidean distance metric is constructed between similar samples. The upper and lower approximation sets of each sample are calculated under a given neighborhood radius, generating a decision rule set for contact state classification, establishing a mapping relationship from the observed data to the state label, and extracting the contact deterioration evaluation index.

[0016] In this embodiment, based on the feature attribute matrix, the importance degree values of each conditional attribute with respect to the decision attribute are calculated to generate a set of weight values representing the importance degrees of different attributes, and the state discrimination distance is obtained to reflect its discrimination ability in state recognition. The attribute differences between all test cycle samples are weighted and analyzed. For the feature differences between two cycle samples, attributes with higher importance will be assigned greater weights, thus playing a greater role in determining sample similarity. Through the weighting method, a distance metric system reflecting the degree of proximity between cycle samples is constructed. On this basis, a neighborhood radius value representing the discrimination scale is set to define which samples can be regarded as neighboring each other. All other samples within this radius are searched around each test sample to form the neighborhood set of this sample. These neighborhood sets reflect the local relationship structure of the test samples in the state space. Using the neighborhood rough set theory, the upper approximation set and the lower approximation set are partitioned and analyzed for each state category. The lower approximation set contains the cycle data in which only samples of the same category exist within its neighborhood, indicating that the sample has a high degree of certainty in the current category; while the upper approximation set contains the cycle data in which samples of other categories are mixed within its neighborhood, indicating the ambiguity of the state boundary. This partition can help the system identify the stable region and the critical transition region in the contact deterioration state classification and form the boundary map of state recognition. The cycle weight vector is introduced into the state discrimination model, so that cycle samples that are more representative or risk-sensitive in the deterioration state recognition have a greater impact on the classification rules. Based on the weighted neighborhood partition result, a set of formal state classification decision rules are extracted. For example, if the average resistance of a certain cycle exceeds a certain value, the resistance volatility increases significantly and the change trend continues to rise, then the contact state of this cycle is classified as moderately deteriorated. Through the comprehensive evaluation of the classification system formed by these rules, the discrimination margin index of the contact state is calculated, and this index is used to reflect the clarity of the overall classification boundary and the strength of the system's ability to distinguish different states.

[0017] Step S5: Based on the average resistance fluctuation measurement value and the contact deterioration evaluation index, perform dual-mode contact failure detection to generate target warning signals for the resistance mutation type and the gradual change type failure modes.

[0018] Specifically, during the continuous test cycle, the resistance change of the contact is continuously monitored, and combined with the resistance fluctuation metric and the deterioration assessment results, a set of identification and early warning strategies for different failure mechanisms are constructed. In terms of resistance mutation detection, a mutation detection threshold is automatically set based on the average resistance fluctuation metric value in the current test sample. This threshold is used to judge whether there is a sharp sudden increase in a certain resistance measurement. When it is detected that the change amplitude of the resistance measurement value of a certain contact relative to its reference value exceeds the mutation threshold, the system starts the high-frequency sampling mode, and the sampling frequency is increased to the millisecond level to obtain a high-density data sequence containing the continuous state transition process. During the sampling process, the contact resistance data is divided into multiple sliding windows of a fixed size, and the changes in the average resistance and standard deviation are calculated respectively within each window, and the dynamic features representing the transient fluctuation state of the contact are extracted. The dynamic features are matched with the pre-constructed deterioration pattern feature library, which contains the resistance jump feature patterns abstracted from historical mutation-type failure cases. By comparing the similarity between the current window sequence and various patterns in the feature library, a matching degree index is obtained and compared with a preset threshold; when the matching degree is higher than the first discrimination threshold, and the resistance values in multiple consecutive sliding windows during the matching process have been stably at a medium deterioration level or above, the system considers that the current contact is in the precursor stage of mutation-type failure, generates an early warning signal for the resistance mutation-type failure mode, and indicates the risk of a rapid and irreversible decline in contact performance. In terms of gradual failure detection, trend modeling is performed on the average resistance fluctuation metric value and the change sequence of the contact deterioration assessment index in multiple consecutive test cycles, and the overall resistance change slope and its fitting stability coefficient are extracted through linear regression. Among them, the slope reflects the speed of contact performance deterioration, and the stability index measures the trend consistency. These trend features are input into the constructed adaptive prediction model, which predicts the resistance state in the short-term future cycle based on the historical fluctuation sequence and outputs a time series containing the predicted average resistance fluctuation value. In the prediction result, deduce the time point when the current resistance trend first exceeds the severe deterioration threshold in the future test cycle, and compare this time point with the safety margin time set inside the system. If the predicted failure time is approaching and less than the response buffer time set by the system, and the credibility index of the prediction result exceeds the reliability threshold, the system generates an early warning signal for the gradual failure mode, indicating that the contact performance is gradually degrading and may experience functional collapse in a specific future cycle. Integrate the mutation-type failure early warning signal and the gradual failure early warning signal to form a target early warning signal and output it, realizing a dual-channel intelligent monitoring and response mechanism for the failure risk of microswitch contacts.

[0019] In the embodiments of the present invention, the present invention can achieve ultra-fast detection of resistor mutation type and gradual change type contact failure modes, greatly advancing the contact failure warning time, and effectively avoiding sudden failures of equipment caused by contact failures. Test cycle weight information is introduced and a weighted contact deterioration evaluation model is constructed. The cycle weight is adaptively learned through a weight optimization function based on the contact resistance change rate, significantly improving the accuracy of contact deterioration state evaluation and the precision of contact life prediction. The existing LC filter circuit is utilized and an innovative sample and hold circuit is integrated into the test device to achieve synchronous contact deterioration detection and ensure the validity of test data without changing the main functions of the filter circuit, reducing the system cost and enhancing the system's ability to adapt to different workload conditions. The neighborhood rough set theory is applied to process the noise and uncertainty in contact test data, and an optimal life prediction parameter set is generated through a forward addition heuristic algorithm, making the life prediction result have a larger reliability margin and providing a more accurate time expectation for equipment maintenance. Through a dual-mode contact failure detection mechanism and adaptive threshold adjustment, the system's ability to identify different types of microswitch contact failure modes is significantly improved, with a wider application range and highly competitive performance in terms of test efficiency and prediction accuracy.

[0020] In a specific embodiment, the process of executing step S1 may specifically include the following steps: Perform continuous opening and closing operations on the microswitch contact a preset number of times for preheating; After the preheating is completed, perform multiple consecutive resistance measurements on the contact closed state to obtain multiple consecutive resistance measurement values; Perform screening processing on the multiple consecutive resistance measurement values to obtain multiple valid measurement values; Calculate the arithmetic mean based on the multiple valid measurement values to obtain the contact resistance reference value; Calculate the standard deviation based on the multiple valid measurement values to obtain the initial contact state evaluation parameter; Analyze the resistance data of the continuous opening and closing operations of the microswitch contact after the first test cycle based on the initial contact state evaluation parameter, and calculate the resistance value interval frequency distribution matrix to obtain the initial fluctuation characteristic spectrum.

[0021] Specifically, the microswitch contacts are preheated by performing continuous opening and closing operations a preset number of times. Through the drive system, the microswitch is continuously and rhythmically opened and closed, so that the contacts experience a series of mechanical friction, elastic recovery, and dynamic reconstruction of the microscopic contact surface under the conditions of unchanged mechanical structure, consistent electrical excitation, and stable environmental parameters. This preheating operation is controlled to be about 100 times, which fully releases the initial assembly stress, surface oxide film, and initial irregularities between the contact surfaces on the premise of ensuring the test efficiency, and makes the contact behavior of the microswitch contacts gradually tend to be stable. When the preheating process ends, the system enters the resistance measurement stage when the contacts are closed. Using a precision resistance sampling module, combined with constant current excitation and stable voltage control, multiple measurements are continuously performed in the state where the contacts are fully closed and there is no bounce, so as to capture the electrical conduction characteristics of the contacts in the initial stable state. The number of sampling times is set to 50 times, and the sampling interval is strictly controlled to ensure that the electrical state is representative in a statistical sense. These sampling data constitute the original resistance measurement sequence, reflecting the change of the conductance performance of the contacts in the non-degraded state. The multiple consecutive resistance measurement values are screened. Using the extreme value elimination method, after sorting the measured original resistance measurement values according to their magnitudes, the smallest and largest 5 measurement values are eliminated each, and the middle 40 data are retained as the effective sample set. This method effectively eliminates the error samples caused by measurement jitter, critical contact bounce, or short-term poor contact, etc., and improves the stability and credibility of subsequent statistical calculations. After screening, an arithmetic mean operation is performed on the 40 effective measurement values to obtain the reference value of the contact resistance. At the same time, the standard deviation of this set of effective values is calculated. This standard deviation reflects the degree of dispersion of the natural fluctuation range of the contact resistance under the condition of no obvious external disturbance and initial damage. The smaller the standard deviation, the more consistent the conduction performance of the contacts in the mechanical stable state; a larger standard deviation reflects the existence of initial microdamage on the surface or uneven distribution of impurities on the contact surface. Establish a dynamic behavior description of the contact resistance under the initial conditions, that is, construct the initial fluctuation characteristic spectrum. After completing the resistance reference measurement, the system allows the microswitch to continue to complete a full test cycle, usually 1000 standard opening and closing operations, and 100 consecutive resistance measurement values in the closed state during this cycle are selected as samples for analyzing the fluctuation behavior during the stable period of the contacts. These measurement values are classified into a preset number of resistance intervals, for example, several resistance ranges are divided at fixed intervals, and the number of occurrences in each range is counted to construct a frequency distribution matrix, that is, the initial resistance value interval frequency distribution matrix. This matrix is regarded as the fluctuation characteristic spectrum of the contacts in the healthy state, reflecting the distribution density and concentration degree of the resistance values in the non-degraded state.

[0022] In a specific embodiment, the process of performing step S2 may specifically include the following steps: Multiply the reference value of the contact resistance by the first coefficient factor to obtain the threshold value of slight contact degradation; Multiply the reference value of the contact resistance by a second coefficient factor to obtain the threshold for moderate deterioration of the contact; Multiply the reference value of the contact resistance by a third coefficient factor to obtain the threshold for severe deterioration of the contact; Conduct interval frequency statistics on the resistance values in the initial fluctuation characteristic spectrum to obtain the characteristic distribution of the normal state of the contact; Calculate the KL divergence value based on the characteristic distribution of the normal state of the contact and the reference distribution matrix to obtain the reference for the rate of change of the contact state; Allocate weight values to the test cycle according to the threshold for slight deterioration of the contact, the threshold for moderate deterioration of the contact, the threshold for severe deterioration of the contact, and the reference for the rate of change of the contact state to obtain the threshold for contact deterioration monitoring.

[0023] Specifically, the reference value of the contact resistance is used as the reference starting point for judging the degree of deterioration, and three gradient coefficient factors are introduced to construct the boundary threshold for classifying the contact performance. Among them, the first coefficient factor is used to identify the early trend of slightly decreased performance. Multiplying the reference resistance value by this factor gives the slightly deteriorated threshold, which is used as the criterion for evaluating the initial appearance of unstable contact or weak degradation of the conduction state of the contact; the second coefficient factor is used to determine the intermediate stage where the resistance has increased significantly but has not reached extreme degradation. Multiplying the reference value by this factor gives the moderately deteriorated threshold; the third coefficient factor represents the demarcation line of severely decreased performance or approaching failure. Multiplying the reference value by this factor gives the severely deteriorated threshold. Analyze the resistance fluctuation behavior pattern in the healthy state of the contact, and perform interval processing on the resistance measurement values included in the initial fluctuation characteristic spectrum. Divide the resistance values into several equal-width intervals, and count the frequency of the resistance values in each interval to form the resistance distribution characteristics of the contact in the non-deteriorated state. This statistical result provides the probability distribution map of the stable operation of the contact and reflects the concentration characteristics or discrete trend of the contact conduction performance. This characteristic distribution serves as the normal state distribution model of the contact as the standard reference in the system, and is used for deviation degree analysis with the actual distribution generated in subsequent test cycles. In order to measure the deviation speed and intensity of the resistance distribution of each test cycle relative to the initial normal distribution, the KL divergence is introduced as a measurement tool for the change rate. Perform frequency statistics on the resistance values obtained in each test cycle, construct a periodic distribution matrix, compare it with the initial reference distribution matrix, and calculate the information entropy of the difference between the two distributions to obtain a KL divergence value representing the state change rate. The higher this value, the more obvious the deviation of the contact resistance distribution from the healthy state in this cycle, the fluctuation behavior shows a new structural change, indicating that the deterioration trend of the contact is accelerating; if this value remains at a low level, it indicates that the resistance behavior is stable and no obvious state mutation has occurred. This divergence value is regarded as the benchmark for the state change rate of the contact and is used to dynamically perceive the degradation speed of the contact performance. Based on the slightly, moderately, and severely deteriorated thresholds, as well as the KL divergence values corresponding to the periodic state deviation speed, a cycle weight calculation mechanism is established to realize the allocation of the relative importance of each test cycle. In this mechanism, the samples of each test cycle are judged whether they are in a certain specific deterioration level interval, and the weight value for subsequent modeling and analysis is comprehensively adjusted according to the change trend of its KL divergence. If a certain cycle is in the interval of slightly rising resistance but the KL divergence suddenly increases, it means that the contact is experiencing a potential structural transition, and the system appropriately increases the weight of this cycle to enhance its influence in the deterioration trend modeling; on the contrary, if a certain cycle is in the moderately deteriorated interval but the change of the KL divergence is extremely small, it means that the contact performance has decreased but the process is slow and stable, and the system then assigns it a lower weight to reduce the interference with the identification of the deterioration trend mutation.A set of weighted evaluation structures for all test cycles is established through a weight allocation mechanism jointly determined by the resistance value and the divergence speed, which is used for the weighted input of subsequent life prediction, state classification, and failure identification models, forming a contact deterioration monitoring threshold system.

[0024] In a specific embodiment, the process of executing step S3 may specifically include the following steps: The microswitch is cyclically driven at a preset frequency to obtain a continuous opening and closing operation sequence; According to the preset sampling rule, measurement points are selected from the continuous opening and closing operation sequence to obtain a contact closing state sequence; The contact closing state sequence is measured for resistance to obtain the original contact resistance measurement data; The original contact resistance measurement data is screened for stable regions according to the contact deterioration monitoring threshold to obtain effective contact resistance data; The effective contact resistance data is associated with the measurement timestamp, environmental parameters, and the cumulative number of operations to obtain multi-dimensional contact state data; The multi-dimensional contact state data is grouped and sorted according to the test cycle to obtain a contact state data set.

[0025] Specifically, the microswitch is cyclically driven at a preset frequency. The specific action frequency is set through a high-precision servo drive module, and the stroke, speed, and loading force during the action process are kept stable and unchanged, so that the microswitch operates continuously at the preset frequency to generate a complete, uniform, and controllable opening and closing operation sequence. In this sequence, each action cycle includes a complete mechanical behavior process of "press - close - release - reset". The stability and repeatability of each closing point are ensured through a position sensor and a time controller. To avoid data redundancy and measurement interference, key measurement points are selectively extracted from the continuous operation sequence based on a sampling strategy. This strategy can be flexibly set based on fixed intervals, random selection, or dynamic changes. For example, every 1000 operations are completed, and 100 closing states are selected as resistance sampling nodes, ensuring that each round of measurement evenly covers the entire test cycle and avoiding excessive influence on the contacts. These measurement points form a contact closing state sequence, that is, a set of operation numbers of all currently prepared operations to enter the measurement process. To ensure sampling consistency, wait for the contact to enter a fully closed and electrically stable state before each sampling, excluding interference factors such as contact bounce, dynamic poor contact, or signal overshoot. Only when the closing state meets the stability criterion, the system starts the measurement process. Perform a resistance sampling operation on each measurement point in the closed state. Using a constant excitation current source and a high-resolution voltage acquisition module, convert the measured closed voltage into a resistance value to form a set of original contact resistance measurement data. These original measurement data reflect the change in the conduction ability of the current microswitch contacts during operation. Screen the stable region of the original contact resistance measurement data according to the contact degradation monitoring threshold. Compare all measured resistance values with a preset resistance reference value. If the change range is within the set allowable deviation range, it is regarded as a stable region measurement value; otherwise, if the resistance value deviates significantly from the reference value and does not have continuity and trend rationality, it is determined as unstable data and is excluded or marked. This screening step maximally retains representative and valid samples that can be used for modeling and state judgment, while isolating mutant data caused by external factors, improving the purity and credibility of the data set. Synchronously integrate each piece of valid contact resistance data with its corresponding measurement timestamp, environmental temperature and humidity parameters at the test moment, and the cumulative number of operations to construct a multi-dimensional structured data unit. Each data unit records the magnitude of the single-point resistance value and retains its exact position on the time axis, environmental background information, and the corresponding operation context, forming a multi-dimensional data model of the contact state. To improve the organizational efficiency and time management ability of data analysis, group and organize the multi-dimensional data based on a fixed number of operations as the cycle benchmark, and centrally process and store all valid resistance measurement data and their multi-dimensional affiliated information within this cycle to form a periodic contact state data set.Each dataset is consistent in structure, recording the numerical changes in contact resistance, corresponding environmental and timing characteristics during this period, and calculating additional statistical indicators such as the average resistance value, resistance standard deviation, maximum and minimum resistance values and their change ranges in this period.

[0026] In a specific embodiment, the process of executing step S4 may specifically include the following steps: Extract the periodic resistance datasets of multiple test periods from the contact state dataset; Calculate the resistance deviation value within each test period for the periodic resistance dataset to obtain the original characteristics of resistance fluctuation; Construct a feature attribute matrix based on the conditional attribute set of contact deterioration state, and the feature attribute matrix includes the average resistance value, resistance standard deviation, maximum resistance value and resistance change slope; Calculate the periodic weight vector based on the original characteristics of resistance fluctuation; According to the periodic weight vector and the original characteristics of resistance fluctuation, calculate the square root of the weighted sum of squared resistance deviations to obtain the average resistance fluctuation metric value; Perform weighted processing on the feature attribute matrix based on the periodic weight vector, and combine the neighborhood rough set theory to calculate the contact state classification decision rule to obtain the contact deterioration evaluation index.

[0027] Specifically, multiple complete test cycles are extracted from the contact state dataset. Each cycle contains a certain number of valid measured values of the closed resistance. These measured values together form the cycle resistance dataset for that cycle, reflecting the fluctuation behavior of the contact conduction performance during this stage. The extraction process is based on a unified cycle length division strategy. For example, every 1000 opening and closing operations are regarded as a cycle. The measured values contained in each cycle, along with their timestamps, environmental parameters, and other information, are extracted together to form a set of data subsets with temporal consistency. The resistance deviation value within each test cycle is calculated for the cycle resistance dataset to obtain a resistance deviation sequence, and then a set of original characteristic indicators describing the internal fluctuation amplitude of each cycle is constituted. These resistance deviation values represent the jitter level of the contact conduction state during short-term stable operation. The larger the amplitude, the more unstable the contact state and the higher the probability of deterioration. Based on the extracted cycle data and their deviation information, a conditional attribute set for the contact deterioration state is constructed, and a characteristic attribute matrix is generated therefrom. Structurally, this matrix has cycles as rows and attributes as columns, covering multiple key indicators, including the average cycle resistance value, which is used to measure the overall level of the cycle conduction state; the standard deviation of the resistance, which is used to characterize the fluctuation intensity within the cycle; the maximum resistance value, which reflects the severity of extreme contact resistance events; and the resistance change slope, which is used to analyze the growth rate of the average resistance value between adjacent cycles with the progress of the test. These attributes comprehensively reflect the static stability and dynamic change trend of the contact. While establishing the characteristic attribute matrix, a cycle weight vector is constructed based on the statistical information of the resistance deviation of each cycle, so as to assign different influence intensities to different cycles in the deterioration trend modeling, enhancing the model's response ability to key state change cycles. The calculation of the weight value refers to the concentration degree of the resistance fluctuation within the cycle, whether the deterioration threshold is reached, and the degree of difference from the initial normal state. Cycles with severe fluctuations or prominent trends will be assigned higher weights to ensure their stronger contribution ability to the overall trend judgment and state division, while cycles with stable fluctuations or close to the initial state will be assigned lower weights, thus reflecting their reference role rather than the dominant role in the overall analysis. The cycle weight vector is applied to the previously obtained resistance deviation sequence for weighted processing. After squaring the deviation value of each cycle and summing them up with weights and then taking the square root, a global average resistance fluctuation metric value is obtained. This indicator represents the weighted resistance fluctuation intensity among all cycles and is an important basis for quantifying the overall state stability of the contact. The higher the value, the more unstable the operating state of the contact and the stronger the degradation tendency. The cycle weight vector is applied to the characteristic attribute matrix, and weighted processing is performed on each attribute dimension, so that the characteristics measured in important cycles have a greater influence on the final model. This weighted matrix serves as the input data for the neighborhood rough set model.Next, based on the neighborhood rough set theory, contact state classification modeling is carried out. The contact states in each cycle are classified into "normal", "slight deterioration", "moderate deterioration", and "severe deterioration" levels according to the labels. Then, in the attribute space, the neighborhood range of each sample is defined, and the neighborhood distance structure is constructed according to the weighted attribute differences, and the neighborhood sample set is determined for each cycle. The upper approximation set and the lower approximation set of the state category are calculated using the neighborhood set respectively to identify the high-confidence samples belonging to this category and the low-confidence samples at the fuzzy zone boundary. Through these approximation structures, state classification rules are constructed to map the attribute features to the state judgment results. By statistically analyzing the tightness of the same-class samples in the feature space and the discreteness between different-class samples, the state discrimination margin of the entire model is calculated to evaluate the robustness and clarity of the classification model. This index is used as the contact deterioration evaluation index to measure the classification certainty and trend evolution speed of the current contact operating state.

[0028] In a specific embodiment, the process of performing the steps of weighting the feature attribute matrix based on the cycle weight vector and calculating the contact state classification decision rule in combination with the neighborhood rough set theory to obtain the contact deterioration evaluation index may specifically include the following steps: Calculate the importance value of each conditional attribute to the decision attribute based on the feature attribute matrix to obtain the attribute importance vector; Calculate the weighted distance matrix between samples based on the attribute importance vector and the feature attribute matrix to obtain the state discrimination distance; Set the neighborhood radius based on the state discrimination distance, construct a δ-neighborhood set for each sample, and obtain the contact state neighborhood relationship; Calculate the upper approximation set and the lower approximation set of the neighborhood rough set for the contact state neighborhood relationship to obtain the contact state classification boundary; Construct the contact state classification decision rule according to the contact state classification boundary and the cycle weight vector, and calculate the state discrimination margin to obtain the contact deterioration evaluation index.

[0029] Specifically, based on the feature attribute matrix, calculate the importance value of each conditional attribute for the decision attribute, and analyze whether this attribute has good discrimination ability among samples in different states. This calculation is based on the difference in dependence degrees in rough set theory, that is, by comparing the change in the degree of dependence of the decision result under the condition of including this attribute and excluding this attribute, the contribution degree of each attribute to classification discrimination is obtained, forming an attribute importance vector. Each value in it represents the classification sensitivity of an attribute. The higher the value, the more accurately the attribute can describe the state difference, and the more suitable it is as the main reference factor in the state classification rule. According to the attribute importance vector and the feature attribute matrix, calculate the weighted distance matrix between samples, obtain the state discrimination distance between each pair of samples, which reflects the structural difference between two cycles in the observed data, and at the same time embeds the weight expression of the contribution size of each attribute to state classification. Based on the above weighted distance matrix, set a neighborhood radius for state recognition to construct the local neighborhood structure of each sample. This neighborhood radius is set by multiplying the global average distance by an empirical coefficient, or is adaptively generated by combining sample density and state distribution. Based on this radius, for each sample, retrieve the set of other samples in the distance matrix whose distance from it is less than this radius, forming the δ-neighborhood set of this sample. By performing this operation on all samples, construct the contact state neighborhood relationship structure, that is, the local aggregation distribution of each cycle sample with its neighboring states in the feature space. Calculate the upper approximation set and the lower approximation set of the neighborhood rough set for the contact state neighborhood relationship to identify the state classification boundary. For each state label, respectively identify the samples whose neighborhood sets are completely included in this state category as the lower approximation set; and the samples whose neighborhood sets partially intersect but do not completely belong to this state category as the upper approximation set. The lower approximation set represents the high-confidence classification result, indicating that these samples show highly concentrated state characteristics in the attribute space, while the upper approximation set represents the fuzzy classification boundary area, containing overlapping samples of different states. Through the ratio of the upper and lower approximation sets, the boundary coincidence situation, and sample density analysis, obtain the boundary clarity of each state category in the attribute space, thereby identifying the effectiveness and reliability of the classification rule. On the basis of the formation of the state classification boundary, combined with the cycle weight vector, the state classification contribution intensity of each sample is weighted and corrected according to its degradation sensitivity in time evolution. For cycles with higher weights, their classification influence is strengthened, thereby enhancing the model's ability to identify state transitions in critical stages. According to the upper and lower approximation sets, neighborhood structure, and weight relationship, extract a series of contact state classification decision rules with high discrimination and high representativeness. Each rule consists of an attribute interval and a state label, in the form of "if the average resistance value of the cycle is in a certain range, the resistance standard deviation is higher than a certain value, and the resistance slope is positive, then the state of this cycle is moderately deteriorated", etc. Conduct an overall analysis of the classification structure formed under the action of all classification rules, statistically analyze the similarity degree of samples of the same category in the attribute space and the interval size between samples of different categories, and calculate the state discrimination margin to quantify the credibility of state classification.The larger the discrimination margin is, the clearer the classification boundary is, and the more discriminative the model is; if the margin is too small, it indicates that there is a large overlap between samples, and more attributes need to be introduced or the classification scale needs to be adjusted. This discrimination margin value is used as the final evaluation index for contact deterioration to reflect the classification clarity, deterioration trend significance, and data distribution reliability of the contact state changes during the current operation of the entire microswitch.

[0030] In a specific embodiment, the process of executing step S5 may specifically include the following steps: Set a resistance mutation detection threshold according to the average resistance fluctuation metric value. When the change rate of the single contact resistance value relative to the reference value exceeds the resistance mutation detection threshold, start the high-frequency sampling mode to obtain a continuous high-frequency sampling data sequence; Construct a sliding window for the continuous high-frequency sampling data sequence based on the contact deterioration evaluation index, and calculate the average resistance and the standard deviation change rate within each sliding window to obtain the dynamic characteristics of resistance fluctuation; Calculate the matching degree between the dynamic characteristics of resistance fluctuation and the preset deterioration mode feature library to obtain the matching degree. When the matching degree exceeds the first threshold and the resistance value of the continuous window exceeds the medium contact deterioration threshold, generate a warning signal for the resistance mutation type failure mode; According to the historical change trend of the average resistance fluctuation metric value and the contact deterioration evaluation index, use linear regression analysis to calculate the resistance change slope and the slope stability index to obtain the resistance change trend characteristics; Input the resistance change trend characteristics into the adaptive prediction model to calculate the average resistance fluctuation metric value for the future test cycle, and obtain the prediction result of the resistance deterioration trend; Calculate the time expectation for the contact to reach the severe contact deterioration threshold according to the prediction result of the resistance deterioration trend. When the expected time is less than the safety margin time and the prediction reliability index is greater than the reliability threshold, generate a warning signal for the gradual failure mode, and merge the warning signal for the resistance mutation type failure mode and the warning signal for the gradual failure mode into the target warning signal for output.

[0031] Specifically, the average resistance fluctuation measurement value calculated in multiple test cycles is used as the global characterization basis of the current contact state stability. The resistance mutation detection threshold is set according to its numerical value and historical evolution law. The threshold has a certain adaptive ability, that is, when the fluctuation measurement value is small, the threshold is set higher to prevent false alarms; when the system detects that the overall fluctuation tends to be violent, the threshold is automatically lowered to enhance the sensitivity to mutation events. The mutation detection threshold is compared with the instantaneous contact resistance value obtained by each sampling. When the relative change amplitude of the resistance value relative to the resistance reference value at a certain moment exceeds the mutation threshold, it is immediately determined that there is a possibility of abnormal fluctuation. On this basis, the system starts the high-frequency sampling mode, increases the sampling frequency from the kilohertz level in the normal state to more than ten kilohertz, and continuously collects multiple data points in a short time to form a continuous high-frequency sampling data sequence covering the entire mutation process. After obtaining the high-frequency data sequence, based on the current degradation evaluation index of the contact, the sliding window parameters are set in combination with the sampling time and the total amount of data, and the entire high-frequency resistance sequence is divided into multiple sliding windows of equal length and overlap, ensuring that the stable capture of the time series change process is achieved without missing the microscopic change characteristics. In each sliding window, the resistance average and standard deviation of all sampling points in the window are calculated respectively, and the change rate of the average and standard deviation between multiple consecutive sliding windows is calculated to construct the dynamic characteristic curve of resistance fluctuation. These characteristics reflect the severity and change direction of resistance in a local time period, and reflect its change rate and change form, which is the basis for analyzing whether the resistance mutation conforms to the established failure mode. The dynamic characteristics of resistance fluctuation are matched and calculated with the internally established degradation mode feature library. The feature library is built based on a large amount of real failure data and contains several standard resistance mutation behavior templates, such as sudden rise type, sawtooth type, jitter type, etc. The fluctuation dynamic characteristics extracted from the current sliding window sequence are matched with all the patterns in the library, and the degree of consistency between the current contact state and the known failure type is evaluated based on the numerical similarity and morphological similarity. If the matching value exceeds the set first judgment threshold, and the phenomenon of resistance value exceeding the moderate degradation threshold continues to appear in multiple sliding windows, it is preliminarily confirmed that the contact has entered the critical interval of rapid jump from normal state to failure state, so that the system outputs the warning signal of resistance mutation failure mode, and records the triggering time and dynamic characteristic indicators of this event. At the same time, in order to achieve early identification of slow degradation trends, continuous trend analysis operations are performed based on the average resistance fluctuation measurement value and degradation evaluation index in the historical multi-cycle data to obtain the resistance degradation trend prediction results. This process calculates the resistance change slope and slope stability index by performing linear regression processing on these historical values. The resistance change slope represents the rising speed of the average resistance fluctuation measurement value with the cycle change, reflecting the contact degradation rate; and the slope stability index is used to evaluate the stability of the regression model fitting. The higher the value, the clearer the trend and the more reliable the fitting effect.Input the resistance change trend feature into the adaptive prediction model to calculate the average resistance fluctuation metric value for the future test cycle. This model combines the current state parameters and trend features to predict the resistance fluctuation metric values within several future test cycles and outputs a sequence of predicted values representing the future cycles. Determine whether there is a resistance fluctuation value in a certain future cycle in the prediction result that exceeds the severe degradation threshold for the first time, and use the time point corresponding to this cycle as the predicted failure time. Compare this predicted time point with the internally set safety margin time. If the predicted failure time is less than the preset minimum response window, that is, the current state has entered the prediction risk control boundary, and at the same time, evaluate it in combination with the fitting confidence level of the prediction process. If the prediction reliability index is higher than the reliability threshold, confirm that the contact is in the critical stage of the gradual degradation process, the system generates a warning signal for the gradual failure mode, and records its predicted failure time and degradation speed parameter. Combine the warning signal for the resistance mutation failure mode and the warning signal for the gradual failure mode to form the target warning signal for output. Set the priority mechanism or fusion strategy. When both warning signals meet the trigger conditions at the same time, determine the final warning output level according to the matching degree, prediction time, change slope, and the level weight of the trigger signal source.

[0032] The above describes the method for testing the contact life of the microswitch in the embodiment of the present invention. Next, the system for testing the contact life of the microswitch in the embodiment of the present invention will be described. Please refer to Figure 2 , an embodiment of the system for testing the contact life of the microswitch in the embodiment of the present invention includes: A reference measurement module, which is used to perform contact preheating operation and resistance reference measurement on the microswitch to obtain the contact resistance reference value and the initial fluctuation feature spectrum; A setting module, which is used to set the contact degradation monitoring threshold based on the contact resistance reference value and the initial fluctuation feature spectrum; A cyclic test module, which is used to place the microswitch in the test device for cyclic testing and record the contact state data set in each test cycle according to the contact degradation monitoring threshold; A weighted processing module, which is used to perform weighted processing on the contact state data set according to the test cycle, and calculate the average resistance fluctuation metric value and the contact degradation evaluation index; A failure detection module, which is used to perform dual-mode contact failure detection based on the average resistance fluctuation metric value and the contact degradation evaluation index, and generate the target warning signal for the resistance mutation type and the gradual failure mode.

[0033] Through the collaborative cooperation of the above-mentioned various components, the present invention can achieve ultra-fast detection of resistive mutation and gradual contact failure modes, significantly advance the contact failure warning time, and effectively avoid sudden failures of equipment caused by contact failures. The test cycle weight information is introduced and a weighted contact deterioration evaluation model is constructed. The cycle weight is adaptively learned through a weight optimization function based on the contact resistance change rate, significantly improving the accuracy of contact deterioration state evaluation and the precision of contact life prediction. The existing LC filter circuit is utilized and an innovative sample and hold circuit is integrated into the test device to achieve synchronous contact deterioration detection and ensure the validity of test data without changing the main functions of the filter circuit, reducing the system cost and enhancing the system's ability to adapt to different workload conditions. The neighborhood rough set theory is applied to process the noise and uncertainty in contact test data, and the optimal life prediction parameter set is generated through a forward addition heuristic algorithm, making the life prediction result have a larger reliability margin and providing a more accurate time expectation for equipment maintenance. Through the dual-mode contact failure detection mechanism and adaptive threshold adjustment, the system's ability to identify different types of microswitch contact failure modes is significantly improved, with a wider application range and highly competitive performance in terms of test efficiency and prediction accuracy.

[0034] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the above-described systems, systems, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0035] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0036] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for testing the life of a micro switch contact, characterized in that: include: Perform contact preheating operation and resistance benchmark measurement on the micro switch to obtain the contact resistance benchmark value and initial fluctuation characteristic spectrum; Setting a contact degradation monitoring threshold based on the contact resistance reference value and the initial fluctuation characteristic spectrum; Placing the micro switch in a test device for cyclic testing, and recording a contact state data set in each test cycle according to the contact degradation monitoring threshold; Performing weighted processing on the contact state data set according to the test cycle, and calculating an average resistance fluctuation measurement value and a contact degradation evaluation index; Dual-mode contact failure detection is performed based on the average resistance fluctuation measurement value and the contact degradation assessment index to generate target early warning signals for resistance mutation type and gradual type failure modes.

2. The micro switch contact life testing method according to claim 1, characterized in that: The step of performing contact preheating operation and resistance reference measurement on the micro switch to obtain a contact resistance reference value and an initial fluctuation characteristic spectrum includes: Preheat the micro switch contacts by performing a preset number of continuous opening and closing operations; After the preheating is completed, multiple continuous resistance measurements are performed on the contact closed state to obtain multiple continuous resistance measurement values; Screening the multiple continuous resistance measurement values ​​to obtain multiple valid measurement values; Calculate an arithmetic mean value according to the multiple valid measurement values ​​to obtain a contact resistance reference value; Calculate a standard deviation based on the multiple valid measurement values ​​to obtain an initial contact state evaluation parameter; The continuous opening and closing operation resistance data of the micro switch contact after completing the first test cycle is analyzed based on the initial contact state evaluation parameters, and the resistance value interval frequency distribution matrix is ​​calculated to obtain the initial fluctuation characteristic spectrum.

3. The micro switch contact life testing method according to claim 1, characterized in that: The step of setting a contact degradation monitoring threshold based on the contact resistance reference value and the initial fluctuation characteristic spectrum includes: Multiplying the contact resistance reference value by a first coefficient factor to obtain a contact slight degradation threshold value; Multiplying the contact resistance reference value by a second coefficient factor to obtain a contact moderate degradation threshold value; Multiplying the contact resistance reference value by a third coefficient factor to obtain a contact severe degradation threshold; Performing interval frequency statistics on the resistance values ​​in the initial fluctuation characteristic spectrum to obtain a characteristic distribution of a contact in a normal state; Calculate the KL divergence value according to the contact normal state characteristic distribution and the reference distribution matrix to obtain the contact state change rate benchmark; A weight value is allocated to the test cycle according to the contact slight degradation threshold, the contact moderate degradation threshold, the contact severe degradation threshold and the contact state change rate benchmark to obtain a contact degradation monitoring threshold.

4. The micro switch contact life testing method according to claim 1, characterized in that: Placing the micro switch in a test device for cyclic testing, and recording a contact state data set in each test cycle according to the contact degradation monitoring threshold, includes: Cyclic driving of the micro switch according to a preset frequency to obtain a continuous opening and closing operation sequence; According to a preset sampling rule, a measurement point is selected from the continuous opening and closing operation sequence to obtain a contact closure state sequence; Performing resistance measurement on the contact closure state sequence to obtain original contact resistance measurement data; Performing stable region screening on the original contact resistance measurement data according to the contact degradation monitoring threshold to obtain effective contact resistance data; Associating the effective contact resistance data with a measurement timestamp, environmental parameters, and cumulative operation times to obtain multi-dimensional contact state data; The contact state multi-dimensional data is grouped and sorted according to the test cycle to obtain a contact state data set.

5. The micro switch contact life testing method according to claim 1, characterized in that: The step of performing weighted processing on the contact state data set according to the test cycle to calculate the average resistance fluctuation measurement value and the contact degradation evaluation index includes: Extracting a periodic resistance data set of a plurality of test cycles from the contact state data set; Calculating the resistance deviation value in each test cycle for the periodic resistance data set to obtain the original characteristics of resistance fluctuation; Constructing a characteristic attribute matrix based on the contact degradation state condition attribute set, wherein the characteristic attribute matrix includes an average resistance value, a resistance standard deviation, a maximum resistance value, and a resistance change slope; Calculating a periodic weight vector based on the original characteristics of the resistance fluctuation; Calculate the square root of the sum of squares of weighted resistance deviations according to the periodic weight vector and the original resistance fluctuation characteristics to obtain an average resistance fluctuation measurement value; The characteristic attribute matrix is ​​weighted based on the periodic weight vector, and the contact state classification decision rule is calculated in combination with the neighborhood rough set theory to obtain the contact degradation evaluation index.

6. The micro switch contact life testing method according to claim 5, characterized in that: The characteristic attribute matrix is ​​weighted based on the periodic weight vector, and the contact state classification decision rule is calculated in combination with the neighborhood rough set theory to obtain the contact degradation evaluation index, including: Calculate the importance value of each condition attribute to the decision attribute based on the characteristic attribute matrix to obtain an attribute importance vector; Calculating a weighted distance matrix between samples according to the attribute importance vector and the feature attribute matrix to obtain a state discrimination distance; The neighborhood radius is set based on the state discrimination distance, a delta-neighborhood set is constructed for each sample, and a contact state neighborhood relationship is obtained; Calculating an upper approximation set and a lower approximation set of a neighborhood rough set for the contact state neighborhood relationship to obtain a contact state classification boundary; A contact state classification decision rule is constructed according to the contact state classification boundary and the period weight vector, and the state discrimination margin is calculated to obtain a contact degradation evaluation index.

7. The micro switch contact life testing method according to claim 1, characterized in that: The dual-mode contact failure detection is performed based on the average resistance fluctuation measurement value and the contact degradation evaluation index to generate target warning signals of resistance mutation type and gradual type failure mode, including: A resistance mutation detection threshold is set according to the average resistance fluctuation measurement value, and when it is detected that the change rate of a single contact resistance value relative to a reference value exceeds the resistance mutation detection threshold, a high-frequency sampling mode is started to obtain a continuous high-frequency sampling data sequence; Based on the contact degradation evaluation index, a sliding window is constructed for the continuous high-frequency sampling data sequence, and the resistance average value and standard deviation change rate in each sliding window are calculated to obtain the resistance fluctuation dynamic characteristics; The resistance fluctuation dynamic characteristics are matched with a preset degradation mode feature library to obtain a matching degree, and when the matching degree exceeds a first threshold and the continuous window resistance value exceeds a contact moderate degradation threshold, a resistance mutation failure mode warning signal is generated; According to the historical change trend of the average resistance fluctuation measurement value and the contact degradation evaluation index, a resistance change slope and a slope stability index are calculated by linear regression analysis to obtain a resistance change trend characteristic; Inputting the resistance variation trend characteristics into the adaptive prediction model to perform average resistance fluctuation measurement value calculation for future test cycles to obtain a resistance degradation trend prediction result; The expected time for the contact to reach the contact severe degradation threshold is calculated based on the resistance degradation trend prediction result. When the expected time is less than the safety margin time and the predicted reliability index is greater than the reliability threshold, a gradual failure mode warning signal is generated, and the resistance mutation failure mode warning signal and the gradual failure mode warning signal are combined into a target warning signal output.

8. A micro switch contact life test system, characterized in that: Used to perform the micro switch contact life test method according to any one of claims 1 to 7, the micro switch contact life test system comprises: A reference measurement module is used to perform contact preheating operation and resistance reference measurement on the micro switch to obtain a contact resistance reference value and an initial fluctuation characteristic spectrum; A setting module, used for setting a contact degradation monitoring threshold based on the contact resistance reference value and the initial fluctuation characteristic spectrum; A cycle test module, used for placing the micro switch in a test device for cycle test, and recording a contact state data set in each test cycle according to the contact degradation monitoring threshold; A weighted processing module, used to perform weighted processing on the contact state data set according to the test cycle, and calculate the average resistance fluctuation measurement value and the contact degradation evaluation index; A failure detection module is used to perform dual-mode contact failure detection based on the average resistance fluctuation measurement value and the contact degradation evaluation index, and generate target warning signals for resistance mutation type and gradual change type failure modes.

Images