Motorcycle one-key start system operation state monitoring method

Abstract

The invention provides a motorcycle one-key start system operation state monitoring method, which comprises the following steps: identifying a surface concave-convex depth difference according to surface concave-convex morphology distribution, evaluating a difference between the surface concave-convex depth difference and a preset wear limit, and determining an adjusted effective contact area proportion; classifying and identifying the effective contact area proportion and the response duration of each starting record, dividing the starting conditions into three starting response types, namely a quick response type, a normal response type and a delayed response type, and identifying starting abnormal conditions of each type at the same time; recognizing a wear depth change track according to previous surface concave-convex depth difference data, predicting a wear depth increment of the contact surface in a future operation cycle through the wear depth change track, and determining a starting performance degradation trend; and analyzing the deviation degree between the output current starting response efficiency and the initial starting response efficiency, generating an alarm signal when the deviation degree exceeds a safety range, and obtaining a final reliability evaluation result.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology, and in particular to a motorcycle one-key starting system running state monitoring method. BACKGROUND

[0002] The motorcycle one-key starting system is the core mechanism to ensure the quick and reliable ignition of the vehicle, and its running state directly affects the user experience and safety performance. Especially in the frequent use scenario, the stability of the system is crucial. The current evaluation method mainly relies on the contact area size to judge the conduction quality of the relay contact, considering that the larger the area is, the smaller the resistance is and the faster the response is. This method is intuitive, but it ignores the dynamic changes of the contact in the use process. The contact area provides the initial judgment basis, and the surface topography distribution reflects the actual state after wear. Both of them determine the conduction reliability. However, when only relying on the contact area value for evaluation, the determination result deviates from the true conduction ability due to the failure to consider the topography changes, and the actual working state of the contact after wear cannot be accurately reflected. The contact will be worn in repeated attraction, and the surface topography will change. The contact with a large area may form uneven areas due to uneven wear distribution, which reduces the actual contact points and increases local oxidation, making the resistance rise instead. For example, in daily riding, the contact area of a motorcycle measured at the time of starting may seem sufficient, but due to long-term friction, one side is severely worn and has pits, and the other side is relatively flat. This makes the current only pass through a small number of raised points, and the low-lying areas cannot form effective contact and cause oxidation in the air to form an oxide layer. Since the resistivity of metal oxides is much higher than that of the metal body, the accumulation of the oxide layer in the low-lying area blocks the conduction path of the area, causing a sudden increase in resistance and delaying the starting for several seconds. Although the contact area has not decreased in geometric size, the actual conduction is only the raised part due to the uneven topography, and the recessed area is completely lost in conduction ability after being covered by the oxide layer. The uneven distribution of the surface topography causes the reliability of the motorcycle one-key starting to decrease. SUMMARY

[0003] The present application provides a motorcycle one-key starting system running state monitoring method, mainly including:

[0004] Obtaining contact data and starting response data when the relay contact is attracted, constructing a unified data set, and determining the initial starting response efficiency;

[0005] Evaluating the wear condition according to the contact data when the relay contact is attracted, and adjusting the effective contact area ratio according to the difference between the surface concave-convex depth difference and the preset wear limit;

[0006] Classifying and identifying the starting response data to determine the starting response type and abnormal condition;

[0007] Predict the performance degradation trend through the wear depth change trajectory;

[0008] Update the conduction capability evaluation rule according to the performance degradation trend, and output the current starting response efficiency;

[0009] Analyze the deviation degree of the current starting response efficiency from the initial starting response efficiency, generate an alarm signal, and form a reliability evaluation result;

[0010] Feed back the reliability evaluation result to the motorcycle one-key starting system, adjust the contact parameters according to the deviation degree, and complete the starting operation state calibration.

[0011] Further, the contact data and starting response data when the relay contact is attracted are obtained, a unified data set is constructed, and the initial starting response efficiency is determined, including:

[0012] The contact area and surface concave-convex topography distribution of the contact area of the contact are obtained through an optical imaging sensor, and the response time from the starting instruction to the ignition success is recorded;

[0013] The effective contact ratio is calculated according to the contact area, the roughness parameter of the surface concave-convex topography distribution is extracted, and the standard time length coefficient is obtained by normalizing the response time;

[0014] The actual conduction coefficient is calculated by using the effective contact ratio and the roughness parameter, the conduction coefficient is corrected according to the comparison result of the standard time length coefficient and the preset threshold, and the initial starting response efficiency is determined by the ratio of the corrected conduction coefficient to the standard time length coefficient.

[0015] Further, the contact data when the relay contact is attracted is used to evaluate the wear condition, and the effective contact area ratio is adjusted according to the difference between the surface concave-convex depth difference and the preset wear limit, including:

[0016] The surface height data of the contact is measured, and the surface concave-convex depth difference is calculated;

[0017] The surface concave-convex depth difference is compared with the preset wear limit, and if it exceeds the limit, the wear overrun ratio is calculated;

[0018] A weight adjustment function is constructed according to the wear overrun ratio, and a weight decay coefficient is calculated;

[0019] The initial weight of the contact area is adjusted by using the weight decay coefficient to obtain an adjusted area weight coefficient;

[0020] The adjusted effective contact area ratio is determined by the operation of the adjusted area weight coefficient and the original contact area.

[0021] Furthermore, the classification and identification of startup response data to determine startup response type and abnormal conditions includes:

[0022] A two-dimensional feature space is constructed based on the effective contact area ratio and response time. A clustering algorithm is used to classify the historical startup data to obtain cluster centers corresponding to different response types.

[0023] Calculate the distance from the newly initiated data point to each cluster center, and classify the data point to the initiation response type corresponding to the nearest cluster center;

[0024] The startup response type and abnormal conditions are determined by the characteristics of the response type and the historical response duration sequence.

[0025] Furthermore, the method of predicting performance degradation trends through wear depth variation trajectories includes:

[0026] Obtain the surface unevenness depth difference data for each time, arrange them in chronological order, and fit the wear depth change trajectory.

[0027] The frequency of abnormal situations in the stated trajectory is statistically analyzed, and the correlation coefficient between the frequency of abnormal situations and the growth rate of depth difference is calculated.

[0028] If the correlation coefficient exceeds the preset threshold, it is determined that there is an accelerated wear trend, and the slope of the change trajectory is adjusted.

[0029] Construct a wear prediction sequence and calculate the wear depth increment within future operating cycles;

[0030] The degradation curve is fitted based on the relationship between the wear depth increment and time within the future operating cycle to determine the performance degradation trend.

[0031] Furthermore, the step of updating the conduction capability evaluation rules based on the performance degradation trend and outputting the current startup response efficiency includes:

[0032] The startup response efficiency data within multiple operating cycles is obtained, the efficiency difference between adjacent cycles is calculated, and the rate of decline of the degradation trend is obtained by the ratio of the efficiency difference to the time interval.

[0033] The preset contact quality judgment limit is dynamically adjusted according to the descent rate. If the descent rate exceeds the preset rate threshold, the limit value is reduced proportionally to obtain the updated evaluation rule.

[0034] Using the updated evaluation rules, the current contact area, surface morphology distribution, and response time are input, normalized, and then weighted to output the current startup response efficiency.

[0035] Furthermore, the analysis of the deviation between the current startup response efficiency and the initial startup response efficiency, generating alarm signals and forming reliability assessment results, includes:

[0036] The deviation percentage is obtained by comparing the difference between the current startup response efficiency and the initial startup response efficiency with the ratio of the initial startup response efficiency.

[0037] If the deviation percentage exceeds a preset safety threshold, an alarm signal will be triggered.

[0038] Based on the trigger status and deviation percentage of the alarm signal, combined with the cumulative usage data, the final reliability assessment result is determined;

[0039] The evaluation results reflect the operating status of the startup system.

[0040] Furthermore, the reliability assessment result is fed back to the motorcycle's one-button start system, and the contact parameters are adjusted according to the degree of deviation to complete the calibration of the start-up and operation status, including:

[0041] The reliability assessment results are transmitted to the startup system via the communication interface;

[0042] Based on the correspondence between the degree of deviation and the preset range, the contact pressure adjustment amount and the pull-in current correction value are obtained, and parameter adjustment instructions are generated.

[0043] The adjusted combination of operating parameters is obtained by adjusting the preload of the contact spring and the drive current of the coil through the actuator.

[0044] The feature similarity is calculated using the parameter combination and the identified startup response type. If the similarity reaches a preset threshold, the startup running status calibration is determined to be complete.

[0045] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0046] This invention discloses a method for monitoring the operational status of a motorcycle one-button start system. It addresses the problem of reduced contact area, prolonged response time, and decreased start-up efficiency due to surface wear of relay contacts during use. This problem further leads to start-up anomalies such as delayed response or contact adhesion, affecting the overall reliability and safety of the motorcycle. This invention obtains contact area, surface unevenness distribution, and response time from the contact engagement point, integrates them into a dataset to identify the initial start-up response efficiency, and assesses the degree of wear based on the difference in surface unevenness depth. When the wear exceeds a preset limit, the effective contact area ratio weight is adjusted to mitigate the impact of wear. Simultaneously, the method classifies and identifies start-up response types and abnormal conditions, predicts future performance degradation trends based on the wear depth change trajectory, updates the conduction capability evaluation rules based on the rate of decline, outputs the current start-up response efficiency, and generates an alarm if the efficiency deviates from the initial efficiency beyond a safe range. Finally, the feedback results are used to adjust the contact pressure and engagement current, achieving start-up status calibration. The technical advantages of this invention are real-time monitoring of contact wear, prediction of degradation, and dynamic parameter optimization, improving the start-up reliability and lifespan of the motorcycle and reducing the risk of failure. Attached Figure Description

[0047] Fig. 1 This is a flowchart of a method for monitoring the operating status of a motorcycle one-button start system according to the present invention.

[0048] Fig. 2 This is a schematic diagram of a method for monitoring the operating status of a one-button start system for motorcycles according to the present invention.

[0049] Fig. 3 This is another schematic diagram of a method for monitoring the operating status of a one-button start system for motorcycles according to the present invention. Detailed Implementation

[0050] To further understand the content of this invention, a detailed description of the invention is provided in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0051] like Figs. 1-3 This embodiment of a method for monitoring the operating status of a motorcycle one-button start system may specifically include:

[0052] S101. Obtain the contact area and surface unevenness distribution when the relay contacts are engaged, and record the response time of each start-up. Integrate the contact area, surface unevenness distribution, and response time into a unified dataset to identify the initial start-up response efficiency.

[0053] At the instant the relay contacts engage, an optical imaging sensor scans the contact area to acquire the contact area value and surface roughness distribution map. Simultaneously, the response time from pressing the start button to successful engine ignition is recorded. The effective contact ratio is calculated based on the contact area value, and the surface roughness parameter is extracted from the surface roughness distribution map. The response time is normalized to obtain a standard duration coefficient. A unified dataset is constructed by combining the effective contact ratio, surface roughness parameter, and standard duration coefficient. The actual initial conduction coefficient is obtained by multiplying the effective contact ratio from the unified dataset by the surface roughness parameter. If the standard duration coefficient is less than a preset threshold, it indicates a fast response and high efficiency; therefore, it is multiplied by a correction factor of 1.2 to amplify the impact. If it is greater than or equal to the threshold, it indicates a slow response and low efficiency; therefore, it is multiplied by a correction factor of 0.8 to reduce the impact. The correction factors of 1.2 and 0.8 are based on analysis of multiple experimental data, reflecting the positive and negative impact ratio of response time on conduction efficiency. The initial start-up response efficiency is determined by dividing the corrected actual conduction coefficient by the standard duration coefficient.

[0054] Specifically, the optical imaging sensor uses a CCD camera in conjunction with a ring-shaped LED light source to complete the scan within 10 milliseconds of the relay contact engaging. The optical imaging sensor uses 45-degree oblique illumination to acquire an image of the light and dark distribution on the contact surface through vertical imaging, forming a morphology distribution map based on the differences in light reflection on the uneven surface. The contact area value is obtained by counting the number of white pixels after image binarization, with each pixel corresponding to an actual area of ​​0.01 square millimeters. The effective contact ratio is calculated based on the ratio of the nominal contact area to the measured contact area. The nominal contact area of ​​a motorcycle relay contact is typically 25 square millimeters; when the measured contact area is 20 square millimeters, the effective contact ratio is 0.8. The surface roughness parameter is extracted by performing a Fourier transform on the morphology distribution map to obtain the root mean square value Ra of the surface undulations. This parameter reflects the degree of microscopic unevenness of the contact surface. The Ra value of a new contact is typically in the range of 0.8 to 1.2 micrometers, while the Ra value of a worn contact increases to over 2.5 micrometers, directly affecting the continuity of the current conduction path. The normalization process employs a minimum-maximum normalization method, mapping the response time to a range of 0 to 1. A normal startup response time of 0.5 seconds is set as the baseline value of 1. For every 0.1-second increase in response time, the standard duration coefficient increases by 0.2. The actual conductivity coefficient is obtained by multiplying the effective contact ratio by the reciprocal of the surface roughness parameter. As surface roughness increases, the reciprocal decreases, reflecting the negative impact of rough surfaces on conductivity.

[0055] In one possible implementation, a preset threshold is set to 0.6. When the standard duration coefficient is less than 0.6, it indicates a fast startup response and good contact condition; in this case, a correction factor of 1.2 is used to increase the weight of the actual continuity coefficient. When the standard duration coefficient is greater than or equal to 0.6, it indicates a startup delay, and the contacts may be worn or oxidized; in this case, a correction factor of 0.8 is used to decrease the weight. The corrected actual continuity coefficient is divided by the standard duration coefficient to obtain the initial startup response efficiency. This efficiency value ranges from 0 to 2; a higher value indicates better startup performance.

[0056] S102. Identify the surface unevenness depth difference based on the surface unevenness distribution, evaluate the difference between the surface unevenness depth difference and the preset wear limit, and determine the adjusted effective contact area ratio.

[0057] The surface unevenness distribution is measured point-by-point using a laser profile scanner to obtain the height data of each sampling point on the contact surface. The vertical distance between the highest and lowest points is calculated to obtain the surface unevenness depth difference. This surface unevenness depth difference is compared with a preset wear limit, which is determined by a preset multiple of the new contact surface unevenness depth difference. When the surface unevenness depth difference exceeds the preset wear limit, the percentage of the excess portion relative to the preset wear limit is calculated to obtain the wear exceedance ratio. A weight adjustment function is constructed using this wear exceedance ratio. If the wear exceedance ratio is greater than zero, a weight attenuation coefficient is calculated using a calculation with the natural constant e as the base and the negative value of the wear exceedance ratio as the exponent. The initial weight value of the contact area is multiplied by this weight attenuation coefficient to obtain the adjusted area weight coefficient. The adjusted area weight coefficient is multiplied by the original contact area and then divided by the rated contact area designed for the relay contact to determine the adjusted effective contact area ratio.

[0058] Specifically, the laser profilometer uses the triangulation principle to perform non-contact measurement of the contact surface. A laser emits a 50-micrometer diameter beam, incident at a 30-degree angle onto the contact surface. The reflected light is received by a CCD sensor located at another angle. As the laser beam scans across the uneven surface, the image position of the reflected light on the sensor shifts. The height of each sampling point relative to a reference surface is calculated using trigonometric relationships. The scanner performs a raster scan of the entire contact area at 0.1-millimeter intervals, acquiring a height data matrix containing 2500 sampling points. The calculation of the surface unevenness depth difference involves statistical processing of the height data matrix. The top 10 points with the highest height values ​​are selected from all sampling points, and their average is calculated as the representative value of the highest point; similarly, the top 10 points with the lowest height values ​​are selected, and their average is calculated as the representative value of the lowest point. The difference between the two representative values ​​is the surface unevenness depth difference. This multi-point averaging method avoids errors caused by individual outliers or measurement noise, improving the stability of the depth difference measurement. The surface unevenness difference of new contacts is typically in the range of 2 to 5 micrometers. After 100,000 engagements, the surface unevenness difference of severely worn contacts will increase to 15 to 20 micrometers. The preset wear limit is determined based on historical wear data statistics of relay contacts. By conducting accelerated aging tests on relays of the same model, the changes in surface unevenness of contacts at different stages of use are recorded. When the unevenness reaches a preset multiple of the initial value of the new contact, the relay's starting success rate will significantly decrease. The preset multiple for motorcycle relays is usually set to 3 times, meaning that when the current unevenness exceeds 3 times the initial unevenness, the contact is considered to have entered a state of severe wear. The wear excess ratio is obtained by calculating the percentage of the excess portion relative to the preset wear limit, reflecting the severity of contact wear. The weight attenuation coefficient is calculated using a negative exponential function of the natural constant e, and its physical meaning lies in simulating the nonlinear effect of contact wear on conductivity. When the wear excess ratio is 0.1, the weight attenuation coefficient is e to the power of -0.1, approximately equal to 0.905; when the wear excess ratio increases to 0.5, the weight attenuation coefficient decreases to approximately 0.606. This exponential decay characteristic aligns with the actual law of contact wear. Initial wear has little impact on conductivity, but as wear intensifies, conductivity drops sharply. The initial weight value of the contact area is preset according to the relay type; for motorcycle starter relays, it is typically set between 0.7 and 0.8, representing the contribution of the contact area to conductivity reliability under normal conditions.

[0059] For example, a motorcycle relay contact that has been used for 6 months has a surface unevenness depth difference of 12 micrometers as measured by laser scanning, while the initial depth difference of a new contact is 3 micrometers, and the preset wear limit is 9 micrometers. The wear excess ratio is calculated by dividing the excess of 3 micrometers by the limit value of 9 micrometers, resulting in 0.333. The weight attenuation coefficient is e raised to the power of -0.333. If the initial weight value is 0.75, then the adjusted area weight coefficient is 0.538, meaning that the contribution of the contact area to the reliability of conduction decreases by nearly one-third.

[0060] In one possible implementation, the rated contact area refers to the ideal contact area specified in the relay design, usually printed in the product's technical parameter sheet. The rated contact area of ​​a motorcycle starter relay is generally 20 to 30 square millimeters, with the specific value depending on the rated current and contact material. The original contact area is obtained through image recognition or contact resistance deduction methods, representing the current actual contact area of ​​the contact. Furthermore, the adjusted effective contact area ratio comprehensively reflects the combined influence of the contact's geometric contact state and surface quality on its conductivity. When the contact surface is smooth and slightly worn, the effective contact area ratio is close to the ratio of the actual contact area to the rated area; when the surface is severely worn and uneven, even with a large geometric contact area, the effective contact area ratio will decrease due to the reduced weighting coefficient. This dynamic adjustment mechanism makes the evaluation results closer to the actual conductivity of the contact, avoiding misjudgments caused by relying solely on the contact area. The effective contact area ratio serves as a key input parameter for subsequent conductivity reliability assessment, and its accuracy directly affects the response speed and success rate of the motorcycle's one-button start. By introducing a dynamic adjustment of the contact area weight based on wear degree, a precise quantitative assessment of the contact aging process is achieved.

[0061] S103. Classify and identify the effective contact area ratio and the response time recorded for each startup, and divide the startup status into three startup response types: fast response type, normal response type, and delayed response type. At the same time, identify the startup abnormal conditions of each type.

[0062] A two-dimensional feature space is constructed based on the effective contact area ratio and response time. K-means clustering is used to cluster historical startup data, resulting in three cluster centers corresponding to the feature distributions of fast response, normal response, and delayed response types, respectively. The Euclidean distance from each newly started data point to each cluster center is calculated, and the data point is assigned to the startup response type corresponding to the nearest cluster center, obtaining the current startup response type label. The standard deviation is calculated using the type features corresponding to the current startup response type label, combined with the historical response time sequence within a preset time window. If the standard deviation exceeds a preset fluctuation threshold for the corresponding type, it is marked as an abnormal response time fluctuation. If the current contact area ratio decreases more than a preset decrease threshold relative to the previous time, it is marked as a sudden drop in contact area. If the response time of the fast response type remains below a preset minimum threshold, it is marked as a sign of contact adhesion, thus identifying an abnormal startup condition.

[0063] Specifically, the two-dimensional feature space is constructed using the effective contact area ratio as the horizontal axis and response time as the vertical axis. Historical startup data includes all startup records of the motorcycle within the past three months, with each record corresponding to a data point in the feature space. The K-means clustering algorithm determines three cluster centers through iterative optimization: the center for fast response type is located in the region with a high contact area ratio and low response time; the center for normal response type is located in the region with a medium contact area ratio and medium response time; and the center for delayed response type is located in the region with a low contact area ratio and high response time. Euclidean distance calculation uses the standard two-dimensional spatial distance formula to measure the distance between the contact area ratio and response time value of the new startup data point and the three cluster centers. The cluster center with the smallest distance determines the response type of that startup. As contact wear intensifies, data points gradually migrate from the fast response region to the delayed response region, reflecting the degradation process of relay performance. The judgment of abnormal response time fluctuations is based on the statistical characteristics of the data within a preset time window. The time window is set to the most recent 20 startup records. The standard deviation of these response times is calculated. When the standard deviation exceeds 1.5 times the preset fluctuation threshold for the corresponding type, it indicates a decrease in startup stability. The fluctuation threshold is 0.05 seconds for fast response type, 0.1 seconds for normal response type, and 0.2 seconds for delayed response type. The difference in thresholds for different types reflects their respective stability requirements. A sudden drop in contact area is determined by comparing the contact area ratio between two adjacent startups. A sudden drop flag is triggered when the current contact area ratio decreases by more than 15% relative to the previous one. This drastic change is usually caused by peeling or severe oxidation of the contact surface material.

[0064] In one possible implementation, contact adhesion is identified by repeatedly testing fast response times with durations below a minimum threshold. This minimum threshold is set to 0.05 seconds; if the duration of 10 consecutive fast responses is below this value, it indicates that the contacts may have adhered due to high-temperature melting, making separation difficult. The identification results of these three abnormal conditions form a startup abnormality flag.

[0065] S104. Identify the wear depth change trajectory based on the surface unevenness depth difference data in previous cycles, predict the wear depth increment of the contact surface in future operating cycles through the wear depth change trajectory, and determine the trend of startup performance degradation.

[0066] The surface unevenness depth difference data from each iteration is acquired and arranged chronologically. A curve showing the change in depth difference with the cumulative number of startups is fitted using the least squares method to obtain the wear depth change trajectory. The number of startup anomalies occurring within each time period of the wear depth change trajectory is statistically analyzed, and the Pearson correlation coefficient between the anomaly frequency and the depth difference growth rate is calculated. If the correlation coefficient exceeds a preset threshold, an accelerated wear trend is identified. The slope of the original change trajectory is weighted and adjusted according to the accelerated wear trend. A first-order univariate gray prediction algorithm is used to construct a wear prediction sequence. The wear depth increment at each time point within the future operating cycle is calculated using this prediction sequence. Optionally, in the first-order univariate gray prediction algorithm, historical wear depth difference data is first selected as the original sequence; then, the original sequence is accumulated once to obtain a first-order accumulated sequence; next, a differential equation is constructed based on the accumulated sequence, and the development coefficient and gray action are estimated using the least squares method; finally, the prediction sequence is obtained by solving the differential equation, and the future wear depth increment is obtained through reconstruction calculation. Based on the relationship between the wear depth increment and time, a startup performance degradation curve is fitted. When the slope of the degradation curve exceeds the degradation threshold determined based on historical failure data, the startup performance degradation trend is determined.

[0067] Specifically, the surface unevenness depth difference data was acquired through periodic testing, with measurements taken every 5000 starts or monthly, forming a time-series dataset. For the motorcycle relay, the depth difference and corresponding cumulative start count were recorded at each testing time point from the start of its service. The least squares fitting process employed a quadratic polynomial model, setting the depth difference as the dependent variable and the cumulative start count as the independent variable. The polynomial coefficients were determined by minimizing the sum of squared residuals. The fitted wear depth change trajectory exhibited characteristics of slow initial growth, stable increase in the middle stage, and accelerated deterioration in the later stage, reflecting the layer-by-layer wear process of the contact material from the surface hardened layer to the substrate material. The frequency of abnormal start-up conditions was statistically analyzed based on the aforementioned classification and identification results, accumulating the frequency of three types of abnormalities within each time period: abnormal response time fluctuations, sudden drops in contact area, and signs of contact adhesion. The abnormality frequency was defined as the ratio of the number of abnormalities per unit time to the total number of starts, and the depth difference growth rate was obtained by dividing the change in depth difference between two adjacent testing points by the time interval. The Pearson correlation coefficient is calculated as the ratio of the product of the covariance and standard deviation of two variable sequences, ranging from -1 to 1. A correlation coefficient exceeding 0.7 indicates a strong positive correlation, meaning that an increase in abnormal frequency is accompanied by faster wear, thus indicating an accelerating wear trend. The first-order univariate algorithm for grey prediction is suitable for grey systems like contact wear, which have partially known and partially unknown information. This algorithm first accumulates the original depth difference sequence, transforming the randomly fluctuating original sequence into an accumulated sequence with a clear pattern. A first-order ordinary differential equation is established on the accumulated sequence, and the time response function is obtained. Then, a subtraction and restoration operation is performed to obtain the predicted sequence. The core of the algorithm lies in mining the inherent patterns of the system through a small amount of historical data, demonstrating good adaptability to systems like motorcycle relays where the wear process is affected by multiple uncertain factors. The slope of the original change trajectory is weighted and adjusted according to the strength of the wear acceleration trend. When an acceleration trend exists, the original slope is multiplied by 1 and a correction factor for the correlation coefficient is added, enabling the prediction model to reflect the accelerating characteristics of wear deterioration.

[0068] Preferably, the future operating cycle is set to 3 months or 15,000 starts, whichever comes first. The wear depth increment is calculated by subtracting the current measured value from the predicted value at each future time point in the prediction sequence.

[0069] For example, if the current depth difference is 12 micrometers, the prediction model shows it will reach 15 micrometers after one month, 18 micrometers after two months, and 22 micrometers after three months, corresponding to wear depth increments of 3, 6, and 10 micrometers, respectively. This increasing sequence of increments indicates that the wear rate is gradually accelerating, and the wear of the contact material exhibits a non-linear growth characteristic. The startup performance degradation curve is fitted using an exponential function, with time as the horizontal axis and the wear depth increment as the vertical axis to construct a coordinate system. The exponential function accurately describes the accelerating characteristics of contact wear, i.e., the wear rate increases exponentially with prolonged use. The slope of the degradation curve is obtained by differentiating the exponential function, representing the rate of change of the wear rate.

[0070] In one possible implementation, the degradation threshold is determined based on statistical analysis of historical failure data. Failure records of relays of the same model are collected, and the slope data of the degradation curve for a period of time before failure is extracted. The mean and standard deviation are calculated. The value of the mean minus one standard deviation is set as the degradation threshold. This setting method can avoid resource waste caused by premature alarms while ensuring a certain safety margin. When the slope of the degradation curve exceeds the degradation threshold, the starting performance degradation trend is determined to be an accelerated degradation state. At this time, the contacts have entered the rapid wear stage, the reliability of the relay drops sharply, and timely maintenance or replacement is required. The determination of the degradation trend not only considers the current wear state but also predicts the future development trend, providing a scientific basis for preventive maintenance. The starting performance degradation trend, as an important part of the evaluation results, can quantitatively characterize the evolution process of the relay from a normal state to a failure state. By analyzing and predicting the wear depth change trajectory, a shift from passive maintenance to proactive prevention is achieved, improving the overall reliability of the motorcycle one-button start system.

[0071] S105. By comparing the rate of decline in startup response efficiency over multiple operating cycles, the rate of decline in startup performance degradation is evaluated. Based on the rate of decline, the contact quality judgment threshold is updated to obtain the conduction capability evaluation rule. The rule inputs are the current contact area, surface unevenness distribution, and response time, and the output is the current startup response efficiency.

[0072] Startup response efficiency data over multiple operating cycles is acquired, and the efficiency difference between adjacent cycles is calculated. The ratio of this difference to the time interval yields the rate of decline in startup performance degradation. A preset contact quality judgment threshold is dynamically adjusted based on this rate of decline. If the rate of decline exceeds a threshold calculated from historical degradation data, the judgment threshold value is reduced according to the ratio of the rate of decline to the threshold, resulting in an updated conductivity evaluation rule. Using this conductivity evaluation rule, three parameters are input: current contact area, surface unevenness distribution, and response time. After normalization, the current startup response efficiency is output by multiplying the contact area by an area weight, the unevenness distribution by an unevenness weight, and the response time by a duration weight.

[0073] Specifically, in one implementation, the operating cycle is defined as 1000 consecutive starts or one week, whichever comes first. Start-up response efficiency data is characterized by the statistical average at the end of each cycle, with the efficiency difference between adjacent cycles reflecting the degree of change in relay performance. The degradation rate is obtained by dividing the efficiency difference by the cycle duration, measured in efficiency values ​​per day or per thousand starts. This dual measurement method adapts to different usage frequency scenarios for motorcycles. Historical degradation data comes from the full lifecycle test records of the same model of relay. By analyzing the complete degradation process from new to failure of 50 samples, the degradation rate distribution at each stage is statistically analyzed, and the 75th quantile of the distribution is set as the rate threshold. When the measured degradation rate is 1.2 times the threshold, the judgment threshold is reduced by 20%; when it is 1.5 times, it is reduced by 30%; and when it exceeds 2 times, it is reduced by 50%. This proportional relationship ensures that the judgment threshold can adaptively adjust as contact wear intensifies. The weights of the three parameters in the conduction capability evaluation rules are determined based on their contribution to start-up reliability. The contact area weight is set to 0.4, as it directly determines the width of the current conduction path; the surface unevenness distribution weight is set to 0.35, reflecting the uniformity of the contact quality; and the response time weight is set to 0.25, reflecting the dynamic response characteristics of the system. The sum of the weight values ​​is 1, ensuring that the output startup response efficiency is within the standardized range of 0 to 1.

[0074] Preferably, the current contact area is derived from the pixel area obtained through image recognition, the surface unevenness distribution is characterized by the reciprocal of the roughness parameter, and the response time is a normalized value. The three parameters are multiplied by their respective weights and then summed to obtain a current startup response efficiency value between 0 and 1; the closer the value is to 1, the better the startup performance.

[0075] For example, after six months of use, a motorcycle relay was found to have a contact area of ​​18 square millimeters, corresponding to a normalized value of 0.72 (preset coefficient 0.4); a surface roughness of 2.5 micrometers, with a morphology distribution normalized value of 0.45 (preset coefficient 0.35); and a response time of 0.8 seconds, with a normalized value of 0.6 (preset coefficient 0.25). Using the evaluation rules, the calculation is: 0.72 × 0.4 + 0.45 × 0.35 + 0.6 × 0.25 = 0.5955, yielding a current starting response efficiency of 0.5955. This indicates that the contacts are in a state of moderate wear, and while the starting performance has decreased somewhat, it remains within an acceptable range.

[0076] S106. Analyze the deviation between the current startup response efficiency and the initial startup response efficiency. When the deviation exceeds the safe range, generate an alarm signal to obtain the final reliability assessment result.

[0077] The difference between the current startup response efficiency and the initial startup response efficiency is calculated. The deviation percentage is obtained by dividing this difference by the initial value. If the deviation percentage exceeds a safety threshold determined based on historical failure data, such as 10%, or the average deviation based on failure cases over the past five years, an alarm signal is triggered. Based on the alarm signal's trigger status and the deviation percentage, combined with the cumulative usage count obtained from the startup records, the final reliability assessment result is obtained.

[0078] Specifically, the deviation percentage is calculated using a relative deviation method, with the difference between the current startup response efficiency and the initial startup response efficiency as the numerator and the initial value as the denominator. When the initial startup response efficiency is 0.95 and the current value drops to 0.57, the deviation percentage is 40%. The safety threshold is determined based on the statistical distribution of historical failure data. Deviation percentage data before failure of relays of the same model are collected, and the average value minus one standard deviation is taken as the safety threshold, typically set within the range of 30% to 35%.

[0079] Specifically, the alarm signals are divided into three levels: a yellow alarm is triggered when the deviation percentage exceeds the safety threshold but is less than 1.5 times the threshold; an orange alarm is triggered when it exceeds 1.5 times but is less than 2 times the threshold; and a red alarm is triggered when it exceeds 2 times the threshold. Different alarm levels correspond to different maintenance strategies. The final reliability assessment result comprehensively considers three factors: alarm level, the specific value of the deviation percentage, and the cumulative number of uses. When a red alarm is triggered and the cumulative number of uses exceeds 80% of the design life, the assessment result is immediate replacement; when an orange alarm is triggered, planned maintenance is required; and when a yellow alarm is triggered, enhanced monitoring is required. Through this graded assessment mechanism, the reliability status of the relay can be accurately determined.

[0080] S107. Feedback the final reliability assessment results to the motorcycle one-button start system. Adjust the relay contact pressure and pull-in current settings according to the deviation between the current start response efficiency and the initial start response efficiency. Match the identified start response type to complete the motorcycle start-up and operation status calibration.

[0081] The final reliability assessment results are transmitted to the motorcycle's one-button start system via a communication interface. Based on the correspondence between the deviation degree and the preset deviation range, a table is consulted to obtain the contact pressure adjustment amount and the pull-in current correction value, resulting in a parameter adjustment command. According to the parameter adjustment command, the spring preload of the relay contacts is adjusted via the actuator to change the contact pressure, while simultaneously adjusting the drive current supplied to the relay coil, thus obtaining the adjusted operating parameter combination. The adjusted operating parameter combination is compared with the identified start response type, and the similarity between the response characteristics corresponding to the parameter combination and the standard characteristics of the start response type is calculated. If the similarity reaches a preset threshold, the motorcycle's start-up and operation state calibration is considered complete.

[0082] Specifically, in one implementation, the communication interface uses a serial communication protocol to achieve real-time transmission of evaluation results. The main control unit of the motorcycle one-button start system receives data packets containing deviation values, alarm levels, and abnormal condition markers via the serial port, parses them, and then proceeds to the parameter adjustment process. The preset deviation range is divided into five levels: 0-10% is the normal range, 10-20% is slight deviation, 20-30% is moderate deviation, 30-40% is severe deviation, and over 40% is extreme deviation. Each range corresponds to a different combination of adjustment parameters, stored in a lookup table.

[0083] Specifically, contact pressure adjustment is achieved by changing the preload of the spring inside the relay. The actuator includes a stepper motor and a screw drive. For every angle the stepper motor rotates, the screw pushes the spring seat 0.01 mm, thus changing the spring compression. When the deviation is slight, the spring preload increases by 10%; for moderate deviation, it increases by 20%; and for severe deviation, it increases by 35%. The increased preload provides greater contact pressure when the contacts engage, improving poor contact caused by wear. The drive current is adjusted via a pulse width modulation circuit. Under normal operation, the relay coil drive current is 200 mA, and the current value is increased proportionally according to the degree of deviation. Increased drive current strengthens the electromagnetic attraction, accelerates contact engagement, reduces the time for arc generation, and delays further ablation of the contact surface. Similarity calculation uses the cosine similarity method. The response feature vector formed by the adjusted operating parameters is compared with the standard feature vectors of the three start-up response types. The response feature vector includes three dimensions: adjusted contact pressure value, drive current value, and expected response time. The expected response time is calculated based on the relay's historical operating data and theoretical response model. It refers to the predicted time, in milliseconds, from the issuance of the command to full engagement of the contact under the current parameter adjustment. When the calculated similarity exceeds 0.85, the parameter adjustment is considered effective, and the calibration is complete.

[0084] For example, a motorcycle relay detected a deviation of 25%, which falls within the moderate deviation range. Looking up a table, the contact pressure needs to be increased by 4 Newtons, and the drive current by 50 mA. The actuator drives the stepper motor to rotate 20 steps, correspondingly increasing the spring preload; simultaneously, the pulse width modulation duty cycle is adjusted from 50% to 62.5%. The similarity between the adjusted parameter combination and the standard characteristics of the normal response type is calculated to be 0.88, exceeding the threshold of 0.85. The motorcycle's starting and running state calibration is successfully completed, and the relay returns to a reliable operating state.

[0085] The above description of the embodiments is only for the purpose of helping to understand the technical solutions and core ideas of this application; those skilled in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for monitoring the operating status of a motorcycle one-button start system, characterized in that, include: Acquire contact data and start-up response data when the relay contacts are engaged, construct a unified dataset, and determine the initial start-up response efficiency; The wear condition is assessed based on the contact data when the relay contacts are engaged, and the effective contact area ratio is adjusted according to the difference between the surface unevenness depth difference and the preset wear limit. Classify and identify startup response data to determine startup response type and abnormal conditions; Predicting performance degradation trends by tracking wear depth changes; Update the conduction capability evaluation rules based on the performance degradation trend, and output the current startup response efficiency; Analyze the deviation between the current startup response efficiency and the initial startup response efficiency, generate alarm signals, and form a reliability assessment result; The reliability assessment results are fed back to the motorcycle's one-button start system, and the contact parameters are adjusted according to the degree of deviation to complete the calibration of the start-up and operation status.

2. The method for monitoring the operating status of a motorcycle one-button start system as described in claim 1, characterized in that, The process of acquiring contact data and start-up response data when the relay contacts are engaged, constructing a unified dataset, and determining the initial start-up response efficiency includes: The contact area and surface morphology distribution of the contact point are obtained by an optical imaging sensor, and the response time from the start command to successful ignition is recorded. The effective contact ratio is calculated based on the contact area, the roughness parameter of the surface unevenness distribution is extracted, and the response time is normalized to obtain the standard time coefficient. The actual conduction coefficient is calculated using the effective contact ratio and roughness parameters. The conduction coefficient is then corrected based on the comparison between the standard duration coefficient and the preset threshold. The initial start-up response efficiency is determined by the ratio of the corrected conduction coefficient to the standard duration coefficient.

3. The method for monitoring the operating status of a motorcycle one-button start system as described in claim 1, characterized in that, The step of assessing wear conditions based on contact data when the relay contacts are engaged, and adjusting the effective contact area ratio based on the difference between the surface unevenness depth and the preset wear limit, includes: Measure the height data of the contact surface and calculate the difference in surface unevenness depth; The surface unevenness depth difference is compared with a preset wear limit. If it exceeds the limit, the wear over-limit ratio is calculated. A weight adjustment function is constructed based on the wear exceedance ratio, and the weight decay coefficient is calculated. The initial weight of the contact area is adjusted using the weight attenuation coefficient to obtain the adjusted area weight coefficient. The adjusted effective contact area ratio is determined by calculating the adjusted area weighting coefficient with the original contact area.

4. The method for monitoring the operating status of a motorcycle one-button start system as described in claim 1, characterized in that, The process of classifying and identifying startup response data to determine startup response type and abnormal conditions includes: A two-dimensional feature space is constructed based on the effective contact area ratio and response time. A clustering algorithm is used to classify the historical startup data to obtain cluster centers corresponding to different response types. Calculate the distance from the newly initiated data point to each cluster center, and classify the data point to the initiation response type corresponding to the nearest cluster center; The startup response type and abnormal conditions are determined by the characteristics of the response type and the historical response duration sequence.

5. The method for monitoring the operating status of a motorcycle one-button start system as described in claim 1, characterized in that, The method of predicting performance degradation trends through wear depth variation trajectories includes: Obtain the surface unevenness depth difference data for each time, arrange them in chronological order, and fit the wear depth change trajectory. The frequency of abnormal situations in the stated trajectory is statistically analyzed, and the correlation coefficient between the frequency of abnormal situations and the growth rate of depth difference is calculated. If the correlation coefficient exceeds the preset threshold, it is determined that there is an accelerated wear trend, and the slope of the change trajectory is adjusted. Construct a wear prediction sequence and calculate the wear depth increment within future operating cycles; The degradation curve is fitted based on the relationship between the wear depth increment and time within the future operating cycle to determine the performance degradation trend.

6. The method for monitoring the operating status of a motorcycle one-button start system as described in claim 1, characterized in that, The step of updating the conduction capability evaluation rules based on performance degradation trends and outputting the current startup response efficiency includes: The startup response efficiency data within multiple operating cycles is obtained, the efficiency difference between adjacent cycles is calculated, and the rate of decline of the degradation trend is obtained by the ratio of the efficiency difference to the time interval. The preset contact quality judgment limit is dynamically adjusted according to the descent rate. If the descent rate exceeds the preset rate threshold, the limit value is reduced proportionally to obtain the updated evaluation rule. Using the updated evaluation rules, the current contact area, surface morphology distribution, and response time are input, normalized, and then weighted to output the current startup response efficiency.

7. The method for monitoring the operating status of a motorcycle one-button start system as described in claim 1, characterized in that, The analysis determines the deviation between the current startup response efficiency and the initial startup response efficiency, generates alarm signals, and forms a reliability assessment result, including: The deviation percentage is obtained by comparing the difference between the current startup response efficiency and the initial startup response efficiency with the ratio of the initial startup response efficiency. If the deviation percentage exceeds a preset safety threshold, an alarm signal will be triggered. Based on the trigger status and deviation percentage of the alarm signal, combined with the cumulative usage data, the final reliability assessment result is determined; The evaluation results reflect the operating status of the startup system.

8. The method for monitoring the operating status of a motorcycle one-button start system as described in claim 1, characterized in that, The reliability assessment results are fed back to the motorcycle's one-button start system, and the contact parameters are adjusted according to the degree of deviation to complete the calibration of the start-up and operation status, including: The reliability assessment results are transmitted to the motorcycle's one-button start system via a communication interface. Based on the correspondence between the degree of deviation and the preset range, the contact pressure adjustment amount and the pull-in current correction value are obtained, and parameter adjustment instructions are generated. The adjusted combination of operating parameters is obtained by adjusting the preload of the contact spring and the drive current of the coil through the actuator. The feature similarity is calculated using the parameter combination and the identified startup response type. If the similarity reaches a preset threshold, the startup running status calibration is determined to be complete.

Images