Vehicle brake pad health monitoring method, device and equipment and vehicle

Abstract

The invention discloses a health monitoring method, device and equipment for a vehicle brake pad and a vehicle, is applied to the technical field of vehicle safety, and can monitor the health state of the brake pad in real time. The health monitoring method for the vehicle brake pad comprises the steps that by collecting time sequence data of any single braking event, according to the preset weight of the time sequence data in the single braking event, the abrasion score of the single braking event is obtained through calculation; adding the wear fractions of the plurality of single braking events to obtain a total wear fraction; and calculating the health degree of the brake pad according to the total wear fraction.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle safety, and particularly to a method, device, equipment and vehicle for health monitoring of vehicle brake pads. Background Art

[0002] With the rapid development of the automotive industry and the continuous progress of intelligent technologies, the safety and reliability of vehicles have become the focus of attention in manufacturing and use. As one of the core components of vehicle safety, the health status of the braking system directly affects driving safety. Brake pads are key components in the braking system, and their health status directly affects the braking effect. However, traditional methods for health detection of brake pads mainly rely on regular manual inspections or mileage-based empirical estimations. Such methods cannot real-time monitor the wear status of brake pads, posing significant safety hazards during driving. Therefore, how to timely evaluate the health status of brake pads has become a technical problem urgently to be solved by those skilled in the art. Summary of the Invention

[0003] To solve the above technical problems, the present application is proposed. Embodiments of the present application provide a method, device, equipment and vehicle for health monitoring of vehicle brake pads that can overcome or at least partially solve the above problems and can real-time monitor the health status of brake pads.

[0004] According to a first aspect of the present application, there is provided a method for health monitoring of vehicle brake pads. The method for health monitoring of vehicle brake pads includes: collecting time series data of any single braking event during a preset time period for brake pad health monitoring; wherein the time series data represents data reflecting the vehicle state and behavior during the braking action, and the preset time period for brake pad health monitoring includes the period from the start of replacing the brake pads to any time point; calculating a wear score for a single braking event according to a preset weight of the time series data in the single braking event; adding up the wear scores of single braking events during the preset time period for brake pad health monitoring to obtain a total wear score; calculating the health degree of the brake pads according to the total wear score; wherein the health degree reflects the health level of the brake pads.

[0005] Optionally, calculating a wear score for a single braking event according to a preset weight of the time series data in the single braking event includes: in a single braking event, multiplying the time series data by a preset weight to obtain a calculation result with the weight assigned; when there are at least two calculation results with the weight assigned, adding up the calculation results with the weight assigned to calculate the wear score for the single braking event.

[0006] In a single braking event, the timing data may include braking pressure, braking temperature, number of brakings, braking duration, vehicle speed, etc. For multiple single braking events, due to different driving scenarios, the dimensions of the timing data affecting the braking event may be different, and the degree of influence each time is also different. Therefore, by performing weighted calculations on the timing data with preset weights, not only can parameters with different dimensions and magnitudes be made comparable, but also the weight coefficients of the preset weights can quantify the contribution degrees of the respective timing data to the braking wear score, making the score calculation more in line with the wear situation in actual braking events. Additionally, for different driving scenarios, the same timing data can be multiplied by different preset weights to make the calculation results of the assigned weights more in line with the current driving scenario, thereby improving the monitoring accuracy of the brake pads. If there are multiple calculation results of the assigned weights, each calculation result of the assigned weights represents an evaluation dimension, and multiple evaluation dimensions jointly affect the wear of the brake pads. Therefore, by adding up the calculation results of multiple evaluation dimensions, there is a non-monotonic relationship between the contributions of the timing data of different dimensions to the wear of the brake pads. Considering the impacts of multiple timing data comprehensively can avoid the limitations of single-dimension evaluation, making the obtained wear score more consistent with the actual wear situation and more accurately reflecting the wear of the brake pads.

[0007] Optionally, in a single braking event, multiplying the timing data by a preset weight to obtain the calculation result of the assigned weight includes: when there are interacting timing data in the timing data, multiplying the interacting timing data to obtain the product of the timing data; multiplying the product of the timing data by the preset weight to obtain the calculation result of the assigned weight; wherein, the interacting timing data refers to the timing data with a preset interaction relationship based on historical experience.

[0008] In a single braking event, if there are dimensional differences in the timing data, direct weighting will lead to dimensional conflicts. Through the product operation, dimensionless interaction terms can be generated, making the subsequent weighting physically meaningful. Moreover, the timing data usually does not act independently on this braking event. There are usually associations or interactions among the timing data. For example, speed and time jointly affect the braking distance, or temperature and pressure jointly affect the tire wear. Multiplying the interacting timing data can capture the joint effects among the timing data and more accurately evaluate the wear of the brake pads. Secondly, the preset weight is used to adjust the importance of different timing data. After multiplying the interacting data and then multiplying by the weight, the importance of these interacting timing data in the overall operation can be adjusted. Multiplication can capture the non-linear relationships among the timing data, while the weight adjusts the influence degree of these non-linear relationships. Finally, through the methods of multiplication and weighting, the interaction in the physical world can be more realistically reflected, and the calculation result of the assigned weight provides a reliable data basis for calculating the wear score.

[0009] Optionally, according to the preset weight of the timing data in a single braking event, the wear score of the single braking event is calculated, including: extracting features from the timing data in the single braking event to obtain feature parameters in the single braking event; calculating the wear score of the single braking event according to the preset weight of the feature parameters in the single braking event.

[0010] In the timing data of a braking event, the importance of features may change dynamically over time. For example, the instantaneous value of the vehicle speed before emergency braking is more valuable for decision-making than the steady-state value. Therefore, key feature parameters (such as mean, variance, peak value, trend, etc.) are extracted from the timing data. These feature parameters can reflect the core information of the timing data, reduce data redundancy, reduce the amount of data processing, and ignore noise data. Feature weighting is to evaluate the importance of the extracted features, and assigning different preset weights to the feature parameters can reflect the contribution of the feature parameters to the wear score. Through the configuration of preset weights, the contribution of each feature parameter to the wear score is quantified, avoiding a single feature dominating the calculation result of the wear score. Therefore, feature weighting can focus on key information, and the configuration of preset weights can quantify the contribution of each feature parameter. The combination of the two can accurately evaluate the wear state of the brake pads and improve the fit between the wear score and the actual wear state.

[0011] Optionally, the method for health monitoring of the vehicle brake pads further includes: normalizing the feature parameters to a preset range to obtain normalized feature parameters; wherein, calculating the wear score of the single braking event according to the preset weight of the feature parameters in the single braking event includes: calculating the wear score of the single braking event according to the normalized feature parameters and the preset weight of the normalized feature parameters in the single braking event.

[0012] There are usually differences in the measurement or units of different feature parameters. To solve the problem of dimensional differences, through normalization, the value ranges of different feature parameters can be scaled to the same preset range, thereby eliminating the influence of dimensional differences. Moreover, normalizing the feature parameters can eliminate the influence of dimensional differences on the weighting operation. On the basis of normalization, assigning preset weights to each feature parameter can clarify the contribution of each feature parameter to the wear score, avoiding certain features dominating the calculation result of the wear score due to their large magnitudes. When the preset weights can be adjusted and configured, according to different scenarios or requirements, the preset weights can be dynamically adjusted, which can improve the matching degree between the wear score and the actual braking scenario, and the wear score can more accurately reflect the wear state of the brake pads in the actual braking scenario. This improvement in the matching degree makes the evaluation result more in line with the actual situation and enhances the practicability and reliability of the health monitoring method.

[0013] Optionally, the timing data includes vehicle speed, longitudinal acceleration, lateral acceleration, and brake pedal depth; feature extraction is performed on the timing data in a single braking event to obtain feature parameters in the single braking event, including: extracting the initial speed in the single braking event according to the vehicle speed; extracting the average longitudinal acceleration in the single braking event according to the longitudinal acceleration of the vehicle; extracting the standard deviation of the lateral acceleration in the single braking event according to the lateral acceleration of the vehicle; extracting the average brake depth and duration in the single braking event according to the brake pedal depth of the vehicle; using the initial speed, the average longitudinal acceleration, the standard deviation of the lateral acceleration, the average brake depth, and the duration as the feature parameters in the single braking event.

[0014] Key features that can reflect the brake wear state are extracted from vehicle speed, longitudinal acceleration, lateral acceleration, and brake pedal depth. Among them, the higher the initial speed, the greater the kinetic energy that the brake pads need to consume, and the more serious the wear may be. Therefore, extracting the initial speed can quantify the contribution of kinetic energy to wear in a braking event. When braking, the wheels are subject to the ground braking force, and the longitudinal acceleration reflects the braking force intensity. The average longitudinal acceleration characterizes the average braking intensity during the entire braking process. The greater the average longitudinal acceleration, the greater the frictional force borne by the brake pads, and the faster the wear. The instantaneous acceleration is easily affected by sensor noise and cannot reflect the braking strategy. Therefore, extracting the average longitudinal acceleration can quantify the impact of braking force on wear. The standard deviation of the lateral acceleration measures the lateral stability of the vehicle. A large standard deviation indicates frequent direction corrections during braking. The fluctuation of the lateral acceleration reflects the stability of the vehicle during braking. The greater the fluctuation, the more likely the brake pads will bear uneven frictional forces, resulting in local wear. Therefore, the standard deviation of the lateral acceleration can quantify the impact of vehicle stability on wear during braking and solve the problem of ignoring the influence of vehicle dynamics on wear distribution. The average brake depth reflects the driver's intensity of using the brakes. The greater the depth, the faster the brake pads wear, quantifying the impact of the driver's behavior on wear. The longer the braking time, the longer the brake pads continuously rub, and the more serious the wear. Therefore, the braking duration can quantify the impact of braking duration on wear. By extracting the initial speed, average longitudinal acceleration, average brake depth, and duration, the contributions of kinetic energy, braking force, driver behavior, and braking duration to wear can be accurately quantified, avoiding calculation errors caused by insufficient single data or features. In addition, in addition to quantifying the wear impact individually, the feature parameters can also jointly evaluate the wear score. For example, the brake depth and duration can be combined to evaluate the wear score of a single braking time. Braking with a low depth for a long time may cause changes in the wear mechanism due to temperature rise. The selection of multi-dimensional timing data and feature extraction improve the accuracy, comprehensiveness, and practicality of brake pad wear assessment, and solve the problems of calculation errors and poor adaptability caused by insufficient or single features.

[0015] Optionally, the health of the brake pads is calculated based on the total wear score, including: comparing the total wear score with a preset score value to obtain the health; wherein, the smaller the difference between the total wear score and the preset score value, the lower the health.

[0016] The total wear score combines the wear scores of multiple single braking events. That is to say, each wear score is closely related to a single braking event. When each wear score has a high degree of matching with a single braking event, the wear score can accurately reflect the wear impact on the brake pads during a single braking event. Therefore, based on the high reliability and accuracy of the wear score of a single braking event, the quantified value calculated from the total wear score can accurately reflect the current wear state of the brake pads and serve as an indicator reflecting the wear degree of the brake pads. However, there is a problem of lacking a quantified evaluation standard with only the total wear score. It is impossible to directly determine whether the brake pads need to be repaired or replaced through the total wear score. Therefore, to solve the problem of lacking a clear judgment standard, a preset score value is set to convert the subjective evaluation into an objective threshold judgment. The preset score value can set a fixed standard according to the design life, usage conditions and safety standards of the brake pads, and can also be dynamically adjusted according to the working conditions. For example, during track driving, due to the accelerated attenuation of the friction material caused by high temperature, the preset score value can be reduced. On winter ice and snow roads, due to the increased braking frequency, the preset score value can be increased. By comparing the wear score with the preset score value, a clear health judgment standard is provided. Users or maintenance personnel can directly judge the state of the brake pads according to the health, avoiding the difficulty of maintenance decisions caused by the lack of a clear standard. When the wear score is close to the preset score value, the system can issue a warning in time to remind the user to perform maintenance and avoid potential safety hazards caused by the failure to detect the wear of the brake pads in time. Through the comparison and judgment, users can avoid replacing the brake pads prematurely and at the same time prevent higher maintenance costs caused by excessive wear, thus optimizing the maintenance plan and reducing unnecessary maintenance costs.

[0017] Optionally, the method for monitoring the health of vehicle brake pads further includes: when the depth of the brake pedal is greater than or equal to a first preset threshold, determining the start of a single braking event; when the depth of the brake pedal is less than a second preset threshold, determining the end of a single braking event.

[0018] When the depth of the brake pedal is greater than or equal to the first preset threshold, it indicates that the driver starts to apply significant braking force and the single braking event starts. When the depth of the brake pedal is less than the second preset threshold, it indicates that the driver reduces the braking force and the single braking event ends. Through the clear threshold definition, braking events can be more accurately distinguished from other driving behaviors (such as light braking or throttle adjustment). The threshold definition provides clear criteria for the start and end of braking events, facilitating the quantitative analysis of braking behaviors and being beneficial to the correct acquisition of time-series data and the accurate quantification of wear scores.

[0019] According to a second aspect of the present application, there is provided a health monitoring device for vehicle brake pads. The health monitoring device for vehicle brake pads includes: a collection module, configured to collect the timing data of any single braking event during a preset time period for brake pad health monitoring; wherein the timing data represents the data reflecting the vehicle state and behavior during the braking behavior, and the preset time period for brake pad health monitoring includes the time from when the brake pads are replaced to any time point; a first calculation module, configured to calculate the wear score of a single braking event according to the preset weight of the timing data in the single braking event; a second calculation module, configured to add up the wear scores of single braking events during the preset time period for brake pad health monitoring to obtain the total wear score; a third calculation module, configured to calculate the health degree of the brake pads according to the total wear score; wherein the health degree reflects the health condition of the brake pads.

[0020] According to a third aspect of the present application, there is provided a device for health monitoring of vehicle brake pads, including: one or more processors; a memory; and one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the processor to implement the method for health monitoring of vehicle brake pads according to the first aspect or any one of the implementation manners in the first aspect.

[0021] According to a fourth aspect of the present application, there is provided a vehicle, including: brake pads; and the device for health monitoring of vehicle brake pads according to the third aspect or any one of the implementation manners in the third aspect, wherein the device for health monitoring of vehicle brake pads is communicatively connected to the brake pads.

[0022] According to a fifth aspect of the present application, there is provided a computer-readable storage medium, on which program codes for executing the method for health monitoring of vehicle brake pads according to the first aspect or any one of the implementation manners in the first aspect are stored.

[0023] According to a sixth aspect of the present application, there is provided a computer program product, including program codes for executing the method for health monitoring of vehicle brake pads according to the first aspect or any one of the implementation manners in the first aspect.

[0024] The health monitoring method, device, equipment and vehicle for vehicle brake pads provided by this application are applied to the brake pads on the vehicle. In the health monitoring method of vehicle brake pads, every time a new brake pad is replaced, the start time of the preset time period for brake pad health monitoring is reset to independently monitor the health of each brake pad. Since the time series data is generated corresponding to the use of the brake pads, and there are differences in the use process and use mode of each brake pad, therefore, in the independent health monitoring, independent calculations are performed on the time series data of each brake pad, which can more accurately reflect the actual state of each brake pad and provide a data basis for improving the accuracy of wear score calculation. In a single braking event, since the data reflecting the vehicle state and behavior in each single braking event will vary according to the actual situation, therefore, arbitrarily collecting the data reflecting the vehicle state and behavior in a single braking event can ensure the comprehensiveness of data collection and provide a reliable basis for calculating the wear score. A preset weight is configured for each time series data in the braking event. The preset weight is a coefficient used to measure the contribution of each time series data to the brake pad wear assessment. And in different braking events, the influence degree of each time series data on the brake pad wear may be different. Therefore, on the premise that the preset weight can be changed, the influence of the time series data on the wear score can be dynamically adjusted to more accurately quantify the contribution of each time series data to the wear score, thereby improving the accuracy of the brake pad wear score calculation. When the wear score calculation of a single braking event is accurate, the total wear score can also more accurately reflect the wear degree of the brake pad. Given the known wear score of the brake pad, it is also necessary to evaluate whether the brake pad meets the replacement or repair standard. To save the process of manual evaluation, the health degree of the brake pad is calculated using the total wear score. The health degree is presented in an intuitive way, and users can understand the state of the brake pad in a timely manner without professional knowledge, and it provides a clear maintenance basis for users and maintenance personnel, avoiding premature or late replacement of the brake pad and reducing the maintenance cost. In addition, in the preset time period for brake pad health monitoring, the health degree is calculated. The time series data comes from each single braking event. Therefore, the time series data has timeliness and pertinence. Using accurate and effective time series data as the basis for calculating the health degree can improve the accuracy and effectiveness of the health degree calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other objects, features and advantages of the present application will become more obvious. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. They are used to explain the present application together with the embodiments of the present application and do not constitute a limitation to the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0026] Figure 1 It is a schematic flowchart of the health monitoring method for vehicle brake pads provided by an exemplary embodiment of the present application.

[0027] Figure 2 It is a schematic flow chart of a method for health monitoring of vehicle brake pads provided by another exemplary embodiment of the present application.

[0028] Figure 3 It is a schematic flow chart of an exemplary method for health monitoring of vehicle brake pads provided by an embodiment of the present application.

[0029] Figure 4 It is a schematic structural diagram of a device for health monitoring of vehicle brake pads provided by an exemplary embodiment of the present application.

[0030] Figure 5 It is a structural diagram of an electronic device provided by an exemplary embodiment of the present application. Detailed implementation manners

[0031] Next, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0032] With the rapid development of the automotive industry and the continuous progress of intelligent technologies, the safety and reliability of vehicles have become the focus of attention of consumers and manufacturers. As a core component of vehicle safety, the performance of the braking system is directly related to the driving safety of the vehicle. Brake pads are key components in the braking system, and their health status directly affects the braking effect. However, traditional methods for health monitoring of brake pads mainly rely on regular manual inspections or experience-based judgments. This method lacks real-time performance and cannot monitor the wear status of brake pads in real time. It can only be inspected at specific time points, making it difficult to detect problems in a timely manner. In some methods for non-manual judgment of brake pad health, the overall health of brake pads can be monitored by combining parameters such as the total usage time of brake pads, braking time, and the number of braking operations. However, in this monitoring method, the wear condition of brake pads is evaluated overall without considering the actual usage situation of brake pads during each braking time. The obtained health degree cannot accurately reflect the actual wear status and can only reflect the approximate wear situation of brake pads. Prompting the driver or maintenance personnel with an inaccurate health degree will lead to premature or late prompting for replacing brake pads, increasing potential safety hazards.

[0033] To solve the problem of difficulty in timely and accurately evaluating the wear degree of brake pads, the present application provides a method for health monitoring of vehicle brake pads, as Figure 1 shown Figure 1The figure is a flow chart of a vehicle brake pad health monitoring method provided by an exemplary embodiment of the present application. The method can be applied to monitoring vehicle brake pads. First, in the preset time period of brake pad health monitoring, the time series data of any single braking event is collected (see Figure 1 Step 101); wherein, the time series data represents data reflecting the vehicle status and behavior in the braking behavior, and the preset time period for brake pad health monitoring includes the time from the replacement of the brake pad to any time point. After each new replacement of the brake pad, the time point of replacing the brake pad can be used as the starting point of the preset time period for brake pad health monitoring, the time point when the brake pad health needs to be checked each time can be used as the end point of the preset time period for brake pad health monitoring, and the time point when the current brake pad is replaced can be used as the end point of the preset time period for brake pad health monitoring of the brake pad. The time series data refers to all the data related to the braking behavior in the braking behavior, including multi-dimensional data. The data related to the braking behavior can be used to evaluate the degree of use of the brake pad and further analyze the degree of wear of the brake pad. Then, according to the preset weight of the time series data in a single braking event, the wear score of the single braking event is calculated (see Figure 1 Step 102). By setting a preset weight according to the actual use of the vehicle, the influence of the time series data on the wear score can be dynamically modified, thereby scoring the vehicle brake pad more accurately and specifically. Then, the wear scores of a single braking event in the preset time period of the brake pad health monitoring are added together to obtain the total wear score (see Figure 1 The total wear score can directly reflect the degree of wear of the vehicle's brake pads, and can then be used to analyze whether the vehicle needs to be replaced. Since the wear threshold of the brake pads of different vehicles is different, finally, the health of the brake pads is calculated based on the total wear score (see Figure 1 Step 104); wherein the health degree reflects the health degree of the brake pad. The same wear score may have different effects on the health degree of the brake pads of different vehicles. Therefore, further calculating the health degree of the brake pad by the total wear score can be applicable to a variety of vehicle models and driving conditions, with high accuracy and universality.

[0034] Combined with the following Figure 1 , a more detailed introduction is given to the vehicle brake pad health monitoring method provided in the embodiment of the present application.

[0035] In S101, in a preset time period for health monitoring of brake pads, time series data of any single braking event is collected, wherein the time series data refers to data reflecting the state and behavior of the vehicle in the braking behavior, and the preset time period for health monitoring of brake pads includes from the start of replacing the brake pads to any time point.

[0036] For the brake pads on a vehicle, each time a new brake pad is replaced is the moment when the preset time period for brake pad health monitoring needs to be reset. Therefore, with the newly replaced brake pad as the starting time point of the preset time period for brake pad health monitoring, the health of the brake pad can be obtained in real time for any time period starting from the starting time point, realizing real-time monitoring of the brake pad. After the vehicle newly replaces the brake pad, at least one braking event will occur during the normal use of the vehicle, and usually multiple braking events will occur when the vehicle does not malfunction. Therefore, a single braking event can be used as a unit to collect and integrate time-series data. The definition of a single braking event can be that when the depth of the brake pedal is greater than or equal to the first preset threshold, it is determined that a single braking event starts, and when the depth of the brake pedal is less than the second preset threshold, it is determined that the single braking event ends. The values of the first preset threshold and the second preset threshold can be the same or different. For example, both the first preset threshold and the second preset threshold are set to 10%, or the first preset threshold is set to 10% and the second preset threshold is set to 5%. The value of the second preset threshold should be less than or equal to the first preset threshold.

[0037] For each single braking event, the time-series data represents the data that is chronologically recorded during the braking behavior and reflects the vehicle's state and behavior during braking. The time-series data is usually collected by vehicle sensors, control systems, or external devices, has timestamps, and can reflect the dynamic changes of the vehicle during braking. Collecting the time-series data for each single braking event can more accurately analyze the wear state of the brake pads. The braking force may be different in each single braking event. For example, the braking pressure difference between hard braking and gentle braking is relatively large. The braking force directly affects the wear speed of the brake pads. Hard braking will result in higher instantaneous wear. Collecting the time-series data of the braking pressure for each braking event can specifically evaluate the impact of the braking force on the wear of the brake pads in each braking event. In each single braking event, the duration of each braking may be different. For example, the difference in duration between long downhill braking and short urban road braking. Long-term braking will cause continuous friction of the brake pads, temperature rise, and accelerated wear. Collecting the duration data for each braking event can specifically evaluate the wear characteristics of the brake pads under long-term friction in this braking event. In each single braking event, the initial vehicle speed at the time of each braking may be different. For example, braking at high speed and braking at low speed. When braking at high speed, the brake pads need to bear greater kinetic energy, and the wear mode may be different from that at low speed. Therefore, collecting the time-series data of the vehicle speed for each braking event can specifically analyze the wear law of the brake pads at the braking event speed. In each single braking event, the driving behavior of each braking may be different. For example, hard braking, gentle braking, continuous point braking, etc. Different driving behaviors will result in different wear modes of the brake pads. Hard braking will accelerate wear. Therefore, collecting the driving behavior data for each braking event can specifically analyze the impact of driving habits on the wear of the brake pads in the braking event. It can be understood that the differences in each braking event will result in different wear modes of the brake pads. Collecting the time-series data separately can more accurately evaluate the impact of each braking on the brake pads, detect abnormal situations in single braking events earlier, provide a reliable data basis for the subsequent calculation of wear scores, and thus improve the consistency between the wear scores and the actual wear of the brake pads.

[0038] The time-series data can include one or more of brake input data, vehicle dynamics data, braking system data, environment and auxiliary data, and driver behavior data, covering multiple dimensions such as the physics, environment, driving behavior, and system status of brake pad wear, ensuring the comprehensiveness of data collection. Selecting appropriate time-series data according to the calculation requirements can more accurately evaluate the degree of brake pad wear. The brake input data can include the brake pedal depth, such as the opening degree of the brake pedal (e.g., percentage or angle); the brake pedal force, the force applied by the driver on the brake pedal; the brake switch status, whether the brake pedal is depressed. The vehicle dynamics data can include the vehicle speed, the speed change of the vehicle during braking; the longitudinal deceleration, the deceleration value of the vehicle during braking; the wheel speed, the rotation speed of each wheel, used to detect whether there is skidding or locking; the acceleration, including longitudinal acceleration and lateral acceleration, reflecting the motion state of the vehicle. The braking system data can include the braking pressure, the hydraulic or pneumatic value in the braking system; the braking torque, the braking torque applied by the braking system on the wheel; the ABS (antilock braking system) status, whether the ABS is activated and its working frequency; the EBD (electronic brake force distribution) status, the distribution of the braking force between the front and rear wheels. The environment and auxiliary data can include the road surface conditions, such as wet or dry, which can be obtained through sensors or external systems; the vehicle load, the total weight of the vehicle, such as obtained through suspension sensors; the temperature data, the temperature of the brake disc or brake pad, reflecting the thermal state of the braking system. The driver behavior data can include: the braking duration, the time from depressing the brake to releasing it; the braking frequency, the number of brakings per unit time; the hard braking flag, whether it belongs to hard braking (judged by the deceleration threshold). There are also other relevant data, such as GPS data: the position, speed, and heading of the vehicle, inertial measurement unit (IMU) data: including angular velocity, acceleration, etc., camera data: recording images or videos of the vehicle's surrounding environment.

[0039] When selecting the time-series data for the health monitoring method of vehicle brake pads, the actual use and working conditions of the vehicle can be considered to select appropriate time-series data. For example, for a certain type of vehicle, count the time-series data that affects the brake pads of this type of vehicle and establish a ranking list, and select the most influential time-series data from the top five or top ten of the ranking list as the time-series data to be collected when monitoring the health of this type of vehicle, and use the collected time-series data for subsequent calculations. The stronger the correlation between the selected time-series data and the brake pads, the more reference value the calculated brake pad wear score has. Therefore, both the quantity and type of the selected time-series data can be adjusted according to actual needs to adapt to various types of vehicles and driving conditions.

[0040] Continue to refer to Figure 1, in S102, according to the preset weight of the timing data in a single braking event, the wear score of the single braking event is calculated. By dynamically adjusting the weight of the timing data, different driving scenarios can be adapted, making the calculation of the wear score more flexible and adaptable.

[0041] In some embodiments, for a single braking event, the timing data is multiplied by a preset weight to obtain a weighted calculation result. If there are multiple pieces of timing data, different weights can be configured for the multiple pieces of timing data, and the different pieces of timing data are respectively multiplied by their configured preset weights. When configuring multiple weights, in order to balance the influence of multiple pieces of timing data on the wear fraction, it can be dynamically adjusted according to different working conditions. When the physical meanings, dimensions, or influence degrees of the respective pieces of timing data are different, different weights also need to be configured. Configuring different weights can more accurately reflect the importance of the data. For example, the brake pedal depth (0 - 1 dimensionless) and the vehicle speed (km / h) have different dimensions, and the influence needs to be standardized by weights. To solve this problem, dimensional consistency can also be achieved through normalization. Or when there are differences in the sensitivity of working conditions, for example, the pedal depth plays a decisive role in emergency braking, while the speed weight is more critical during high-speed driving, different weights can also be configured. When configuring appropriate weight values for each piece of timing data, the following can be considered: First, according to domain knowledge or expert experience, the relative importance of each piece of timing data to the brake pad wear can be determined. The weight should be positively correlated with the influence degree of the timing data on the braking behavior. The configured weight can be automatically adjusted according to the working conditions. For example, when the timing data is the brake pedal depth, vehicle speed, longitudinal acceleration, lateral acceleration, etc., the weight of the pedal depth is higher than that of the speed because the pedal depth directly represents the braking input, while the weight of the lateral acceleration is lower because the lateral acceleration mainly affects uneven wear rather than immediate wear. A dynamic association configuration method can be introduced. For example, the higher the speed, the greater the weight of the pedal depth (the depth is more sensitive to wear at high speeds), and the higher the brake temperature, the greater the speed weight (the risk of thermal fade intensifies). Second, historical data can be collected, including the timing data and the corresponding brake pad wear status, and a model can be trained using machine learning algorithms (such as linear regression, random forest, neural network) to learn the contribution of each piece of timing data to the wear status. According to the feature importance output by the model, the weight configuration is adjusted, that is, the weights are configured based on historical experience. For example, in a single braking event with an emergency brake, the weight of the braking force automatically increases, and in a single braking event with a long downhill, the weight of the braking duration automatically increases. Third, rules can be set according to the specific conditions of the braking event (such as vehicle speed, environment, load), and the weights can be dynamically adjusted according to the rules. For example, if the vehicle speed > 100 km / h, the weight of the braking force increases, and if the road surface humidity > 50%, the weight of the environmental conditions increases. In this weight configuration method, if there are multiple pieces of timing data with different weights configured, there will be at least two weighted calculation results. When there are at least two weighted calculation results, the weighted calculation results are added together to calculate the wear fraction of a single braking event. That is to say, if there are two types of timing data and two preset weights are respectively configured, after the two pieces of timing data are respectively multiplied by the preset weights, the two weighted calculation results are added together to obtain the final wear fraction.

[0042] When the influence degrees of the selected time-series data on a single braking event are the same or can be regarded as the same, the same preset weight can be configured for multiple time-series data. That is to say, in the calculation, multiple time-series data with the same configured preset weight are multiplied by the same preset weight, and then the product results are added together, which is the same as the calculation method of multiple time-series data with different weight values. Or because the weight values of the preset weights are the same, the time-series data can be normalized first and then added together, and then the added result is multiplied by the preset weight to obtain the final wear score.

[0043] In addition, in addition to directly considering setting weights for time-series data, when there are interacting time-series data in the time-series data, the time-series data that jointly affect the wear degree of the brake pad can be considered to be multiplied to optimize the calculation process. That is, when there are interacting time-series data in the time-series data, after multiplying the interacting time-series data, the product of the time-series data is obtained. If these time-series data are multiplied first, and then the product of the time-series data is multiplied by the preset weight to obtain the weighted calculation result, let the interacting time-series data be multiplied and bear the preset weight as a whole, which highlights the comprehensive influence of these time-series data, maintains the interaction between the time-series data, and more accurately reflects the comprehensive influence during the braking process. And let another part of the time-series data bear the remaining weight, which simplifies the calculation process and improves the calculation efficiency. By bearing the preset weight as a whole, the problem of uneven importance of each time-series data is avoided. The interacting time-series data refer to the time-series data with a preset interaction relationship according to historical experience. For example, the four time-series data of speed, braking depth, acceleration, and time have a close interaction during the braking process and cannot be completely decoupled. If the weight is directly decomposed into independent terms, the product of the normalized speed, braking depth, acceleration, and time can capture the non-linear relationship between these four time-series data and highlight their synergistic effect during the braking process. If the four time-series data of speed, braking depth, acceleration, and time are decomposed into independent terms and a weight is assigned to each time-series data separately, due to ignoring the coupling term, there will be a large error between the calculated wear score and the actual wear amount. Therefore, considering the interaction between time-series data when calculating the wear score, multiplying the time-series data and then jointly assigning a weight coefficient can retain the non-linear coupling relationship between the time-series data, reduce the number of weight configurations, simplify the calculation process, improve the calculation efficiency, and also separately retain the weight allocation of independent time-series data when there are independent time-series data, making the evaluation result more balanced and the calculated wear data have a higher matching degree with the actual wear amount.

[0044] Typically, time-series data is directly collected by each sensor. As raw data, time-series data contains a large amount of redundant information, and the sources of time-series data are different, resulting in differences in formats. Therefore, in some embodiments, to improve the uniformity of time-series data, remove irrelevant or redundant features, and reduce the data dimension, refer to Figure 2 , Figure 2 which is a schematic flowchart of a method for monitoring the health of vehicle brake pads provided by another exemplary embodiment of the present application. After S101, feature extraction can be performed on the time-series data in a single braking event to obtain feature parameters in the single braking event (refer to S201 in Figure 2 ); according to the preset weights of the feature parameters in the single braking event, the wear score of the single braking event can be calculated (refer to S202 in Figure 2 ). Feature parameters are the key information extracted from time-series data. These feature parameters can reflect the core information of time-series data and the important features of conceptual time-series data. Extracting feature parameters helps simplify the data and improve the calculation efficiency. For example, if the dimension of the raw data is N×4 (N is the number of sampling points), through feature extraction (such as taking extreme values, means, variances, etc.), the dimension is reduced to 1×n, and usually n << N. Therefore, feature extraction can significantly reduce the calculation complexity. By configuring the preset weights, the contribution of each feature parameter to the wear score can be quantified, avoiding a single feature dominating the calculation result of the wear score. Therefore, feature extraction and weighting can focus on the key information, and the configuration of preset weights can quantify the contribution of each feature parameter. The combination of the two can accurately evaluate the wear state of the brake pads and improve the fitting degree between the wear score and the actual wear state.

[0045] As a possible implementation, the collected time-series data can be selected to include: vehicle speed, longitudinal acceleration, lateral acceleration, and brake pedal depth, and then feature extraction is performed: according to the vehicle speed, the initial speed (v_initial) in a single braking event is extracted, that is, the vehicle speed at the start of the braking time, the speed at the start of braking, which may affect the kinetic energy and wear degree during braking. Brake wear is closely related to the conversion of kinetic energy, and the kinetic energy formula is E_k = 0.5mV0 2 . The initial speed V0 directly determines the total amount of energy that the braking system needs to dissipate. Extracting the initial speed can establish an explicit relationship between wear and energy dissipation and can predict the wear differences in different initial speed scenarios. According to the longitudinal acceleration (a_avg) of the vehicle, the average longitudinal acceleration in a single braking event is extracted, that is, the average value of the longitudinal acceleration during the braking event (taking the absolute value). When braking, the braking force F on the wheel by the ground = ma, and the longitudinal acceleration a xReflect the braking force intensity. The average longitudinal acceleration āx characterizes the average braking intensity during the entire braking process. The instantaneous acceleration is susceptible to sensor noise interference and cannot reflect the braking strategy (such as constant deceleration vs. strong first and then weak). Therefore, extracting the average longitudinal acceleration to quantify the impact of braking force on wear, the longitudinal acceleration reflects the braking force. The greater the acceleration, the greater the frictional force borne by the brake pads and the faster the wear. According to the lateral acceleration of the vehicle, extract the standard deviation of the lateral acceleration (lat_std) in a single braking event, that is, the standard deviation of the lateral acceleration during the braking event, which reflects the steering stability. The fluctuation of the lateral acceleration reflects the stability of the vehicle during braking. The greater the fluctuation, the more likely the brake pads will bear uneven frictional forces, resulting in local wear. During emergency lane-changing braking, the inner and outer wheels wear unevenly, which affects the wear distribution. The standard deviation of the lateral acceleration can measure the lateral stability of the vehicle. A large standard deviation indicates frequent direction corrections during braking (such as emergency obstacle avoidance). Therefore, extract the standard deviation of the lateral acceleration and introduce wear distribution evaluation to quantify the impact of vehicle stability during braking on wear. According to the brake pedal depth of the vehicle, extract the average brake depth (d_avg) and duration (t) in a single braking event. The brake depth reflects the intensity of the driver's use of the brake. The greater the depth, the faster the brake pads wear. The average brake depth represents the average value of the brake pedal depth during the braking event, and the duration represents the duration of a single braking event. Using only the maximum brake depth or total time alone cannot characterize dynamic braking behavior (such as intermittent braking vs. continuous heavy braking). Capture the "thermal-mechanical" coupling effect through the average brake depth and duration: long-term low-depth braking may cause changes in the wear mechanism due to temperature rise, quantifying the impact of driver behavior on wear and the impact of braking duration on wear. The initial speed, average longitudinal acceleration, standard deviation of the lateral acceleration, average brake depth, and duration extracted are the characteristic parameters in a single braking event, and during the use of this brake pad, the same characteristic parameters are extracted for any single braking event to maintain the unity of calculation. The selection of time series data can be adjusted according to actual needs. For example, adding or reducing features in this implementation manner, its necessity and reasonableness can be evaluated according to specific application scenarios and technical requirements. For example, features such as steering wheel angle or brake temperature can be added, or irrelevant features can be reduced to simplify the calculation.

[0046] After extracting features from the time series data, to avoid the problem of feature imbalance caused by dimensional differences, in some embodiments, refer to Figure 3 , Figure 3 is a schematic flowchart of a method for health monitoring of vehicle brake pads provided in an embodiment of the present application. After S201, the feature parameters can be further normalized to a preset range to obtain the normalized feature parameters (refer to Figure 3In step S301). The main purpose of normalization is to scale the values of different features to the same scale or range, thereby eliminating the differences in dimension and numerical range between features. For example, the preset range can be set to [0, 1] to eliminate the influence of dimension, ensure that each feature has a reasonable weight in the wear fraction calculation, and improve the accuracy of the wear fraction calculation. Normalization methods can include min-max normalization, Z-score standardization, etc. After normalization, when configuring preset weights for the non-normalized feature parameters, all features are in the same dimension, and the preset weights can directly represent the decision contribution degree of the feature. Normalization suppresses the numerical instability caused by dimension differences.

[0047] As a possible implementation, when the extracted initial speed, average longitudinal acceleration, standard deviation of lateral acceleration, average braking depth, and duration are used as feature parameters in a single braking event, the feature parameters have different dimensions and units. Therefore, to eliminate the influence of dimension, the normalization method is as follows: Speed normalization: v_norm = v_initial / v_max, where v_norm represents the normalized initial speed, v_max is the maximum design vehicle speed, and v_initial is the initial speed; Braking depth normalization: d_norm = d_avg / 100, where d_norm represents the normalized average braking depth, d_avg is the average braking depth, and 100 is the maximum value of the braking pedal depth; Acceleration normalization: a_norm = a_avg / a_max, where a_norm is the normalized average longitudinal acceleration, a_avg is the average longitudinal acceleration, and a_max is the maximum design acceleration; Time normalization: t_norm = t / t_max, where t_max is the maximum design braking duration, t_norm is the normalized duration, and t is the duration; Standard deviation of lateral acceleration normalization: lat_std_norm = lat_std / lat_max, where lat_max is the maximum design lateral acceleration, lat_std_norm is the normalized standard deviation of lateral acceleration, and lat_std is the standard deviation of lateral acceleration. When the feature parameters are other parameters, normalization can also be performed to improve the accuracy, stability, and efficiency of the score calculation.

[0048] To further optimize the calculation process and improve the calculation accuracy, in some other embodiments, continue to refer to Figure 3 After S301, based on the normalized feature parameters and the preset weights of the normalized feature parameters in a single braking event, the wear fraction of the single braking event can be calculated (refer to Figure 3In step S302). That is to say, first, extract features from the time series data, then normalize the extracted features. On the basis of normalization, assign preset weights to each feature parameter to reflect its relative importance to the final evaluation result. The preset weights can be determined based on domain knowledge, historical data, or machine learning methods. Normalization eliminates the dimensional differences between feature parameters, and assigning preset weights further quantifies the contributions of each feature, ensuring the fairness and accuracy of the evaluation result. Normalization makes the data distribution more concentrated, and weight configuration enables the model to flexibly adjust the importance of each feature. The combination of the two improves the adaptability of the method for calculating the wear score to different scenarios. Moreover, normalization reduces the instability in numerical calculations, and weight configuration simplifies the computational complexity. The combination of the two enhances the computational efficiency and real-time performance. Finally, the normalized feature parameters are comparable, and weight configuration synthesizes the influence of multi-dimensional features. The combination of the two provides a more comprehensive and reliable evaluation result.

[0049] As a possible implementation, when the initial speed, average longitudinal acceleration, standard deviation of lateral acceleration, average braking depth, and duration extracted are used as feature parameters in a single braking event, and the feature parameters are normalized to obtain the normalized initial speed v_norm, normalized average braking depth d_norm, normalized average longitudinal acceleration a_norm, and normalized duration t_norm, set the overall weights of speed, depth, acceleration, and time to 60%, and set the normalized standard deviation of lateral acceleration lat_std_norm to 40%. Then the calculation formula for the wear score can be: (0.6 × v_norm × d_norm × a_norm × t_norm) + (0.4 × lat_std_norm). When designing this calculation formula, the interaction of the four factors of speed, braking depth, acceleration, and time during braking is considered. They jointly affect the wear degree of the brake pads. If these factors are multiplied first and then multiplied by 0.6, it can comprehensively reflect the combined effect of these four factors, thus more accurately evaluating the wear contribution of the braking event to the brake pads.

[0050] Specifically, this design makes the product term bear 60% of the weight as a whole, while the standard deviation of lateral acceleration alone bears 40% of the weight. This not only highlights the combined influence of speed, braking depth, deceleration, and time, but also simplifies the calculation process. 60% and 40% are empirical values and can be modified according to the actual situation of the vehicle in actual calculations.

[0051] Continue to refer to Figure 1, in S103, the wear scores of single braking events in the preset time period for brake pad health monitoring are added together to obtain the total wear score. Taking each single braking event as a unit, the wear score of each single braking event is calculated separately. For a brake pad, from the time when the brake pad is newly replaced until any moment, the accumulated wear scores of all braking times in the preset time period for brake pad health monitoring can be used as the total wear score of the brake pad. That is to say, taking the preset time period for brake pad health monitoring as a whole and each single braking event as multiple units within this preset time period, when calculating the total wear score, multiple braking events need to be integrated to jointly evaluate the wear degree of the brake pad. After replacing the brake pad, the wear scores of new braking events are accumulated again as the total wear score of the new brake pad. That is to say, every time the brake pad is replaced, the total wear score needs to be reset once. However, the historical wear scores and the actual wear conditions of the brake pads can be statistically recorded as historical experience, which is used to screen time series data and adjust the values of preset weights. The braking process is affected by various factors such as speed, road conditions, and driver reaction. These factors may be different in each braking event, resulting in the uniqueness of each event. Time series data can record the changes of various parameters during the braking process over time and provide detailed information on the evolution of events. Therefore, collecting the time series data of each braking event separately can ensure the independence and accuracy of the time series data in different events, avoid data interference between different braking events, reduce the error between the wear score and the actual wear situation of each braking event, and thus further reduce the error between the total wear score and the actual model situation of the brake pad when calculating the total wear score, improving the accuracy of wear calculation. After replacing the brake pad, recalculating the total wear score of the brake pad can provide a more accurate life prediction based on the initial state and actual usage conditions of the new brake pad, avoid prediction deviation caused by using old data, and ensure the reliability of the total wear score evaluation. If a model is used to calculate the total wear score, after replacing the brake pad, the system can recalibrate the wear model to adapt to the characteristics of the new brake pad. In addition, the wear situation of each brake pad may be different due to factors such as installation position, usage frequency, and driving conditions. Performing independent calculations for each brake pad can accurately capture these differences. The total wear score of each brake pad has individual differences. Performing calculations on the time series data of each new brake pad can accurately reflect individual differences, and thus can improve the consistency between the total wear score and the actual wear situation of the brake pad.

[0052] Continue to refer to Figure 1 , in S104, according to the total wear score, calculate the health degree of the brake pad.

[0053] Among them, the health degree reflects the health condition of the brake pads. The total wear score is calculated and is an indicator reflecting the wear degree of the brake pads. However, for different brake pads with the same total wear score, their wear degrees are not necessarily the same. There is a lack of a quantitative evaluation standard for only the total wear score, and it cannot respond to the wear state in real time. Whether the brake pads need to be repaired or replaced cannot be directly determined by the total wear score. Therefore, to solve the problem of the lack of a clear judgment standard, a preset score value is set as a reference benchmark value, usually a preset threshold or limit value, to convert the subjective evaluation into an objective threshold judgment. The preset score value can set a fixed standard according to the designed life, usage conditions and safety standards of the brake pads. The preset score value can also be dynamically adjusted according to the working conditions. Static thresholds cannot meet the requirements of multiple scenarios. For example, the service life of the same brake pad in mountainous areas and cities can vary by up to 30%. When driving on a race track, due to high temperatures, the attenuation of the friction material accelerates, and the preset score value is reduced by 15% based on the original design. On winter ice and snow roads, due to the increased braking frequency, the preset score value can be increased by 10% based on the original design.

[0054] After completing the design of the preset score value, in some embodiments, the total wear score is compared with the preset score value to obtain the health degree; among them, the smaller the difference between the total wear score and the preset score value, the lower the health degree. The preset score value can represent the designed life of the preset brake pads. The preset score value is a clear reference point that can help quantify the gap between the current state and the ideal state of the equipment. Moreover, although the total wear scores of brake pads of different vehicles are the same, the actual wear degrees they represent may be different. Therefore, the total wear scores of different brake pads can be uniformly expressed as a percentage or a score by comparing with their respective preset score values, which is convenient for horizontal comparison and intuitively determining the health degree of the brake pads.

[0055] As a possible implementation method, the calculation formula for the health degree can be designed as: Health degree = max(0, 100 - (total wear score / preset score value) × 100); where, (total wear score / preset score value) × 100 converts this ratio into a percentage form, and 100 - (total wear score / preset score value) × 100 represents subtracting the ratio of the total wear score to the preset score value from 100%, reflecting the deviation degree between the total wear score and the preset score value. max(0, (total wear score / preset score value) × 100) ensures that the score of the health degree will not be negative, and the minimum value is 0. Finally, the value range of the health degree is between [0, 100]. The larger the value of the health degree, the better the health degree or state, and the smaller the value, the worse the health degree or state.

[0056] In addition, after obtaining the health degree of the brake pads, the health degree score can be displayed in real time on the vehicle-mounted system or the mobile terminal, so that the maintenance personnel or the drivers can intuitively view the wear degree of the brake pads and timely maintain and replace the brake pads.

[0057] If the health score is only displayed in real time on the in-vehicle system or mobile terminal, the driver may not be able to view the screen in a timely manner during vehicle operation. Therefore, an alarm threshold can also be designed. When the health level is lower than the alarm threshold (e.g., 20%), a warning prompt is triggered to remind the driver to replace the brake pads in a timely manner.

[0058] The alarm threshold can also be divided into multiple levels in detail. For example, the alarm threshold value of the first level is the highest, which is used to remind the driver or operation and maintenance personnel to pay attention to the brake pads and perform maintenance on the brake pads. There may be potential risks, and it is expected that the brake pads need to be inspected or replaced, so that the driver or operation and maintenance personnel can reasonably plan the vehicle usage time and replace the brake pads in a timely manner when they need to be replaced, improving the vehicle operation safety. The alarm threshold value of the second level is lower than that of the first level, further reminding the driver or operation and maintenance personnel of the need to inspect or replace the brake pads, so that the driver or operation and maintenance personnel can inspect or replace the brake pads in a timely manner. The alarm threshold of the third level is lower than that of the second level, indicating the most urgent situation, and the driver or operation and maintenance personnel need to immediately replace the brake pads to ensure the normal and safe operation of the vehicle. The division of the alarm threshold can be designed according to the actual needs and usage scenarios of the vehicle and the driver. Providing a real-time warning function can effectively prevent safety accidents caused by excessive wear of the brake pads.

[0059] Figure 4 is a schematic structural diagram of a health monitoring device for vehicle brake pads provided by an exemplary embodiment of the present application, as Figure 4 shown, the health monitoring device 4 of the vehicle brake pads includes: an acquisition module 41, configured to acquire the timing data of any single braking event during a preset time period for brake pad health monitoring; wherein, the timing data represents multi-dimensional data in the braking behavior, and the preset time period for brake pad health monitoring includes from the start of replacing the brake pads to any time point; a first calculation module 42, configured to calculate the wear score of a single braking event according to the preset weight of the timing data in the single braking event; a second calculation module 43, configured to add up the wear scores of single braking events during the preset time period for brake pad health monitoring to obtain the total wear score; a third calculation module 44, configured to calculate the health level of the brake pads according to the total wear score; wherein, the health level reflects the health degree of the brake pads.

[0060] As a possible implementation manner, the first calculation module 42 can also be configured to: in a single braking event, multiply the timing data by a preset weight to obtain a weighted calculation result; when there are at least two weighted calculation results, add up the weighted calculation results to calculate the wear score of the single braking event.

[0061] As a possible implementation, the first calculation module 42 can also be configured to: when there is interacting timing data in the timing data, multiply the interacting timing data to obtain the product of the timing data; multiply the product of the timing data by a preset weight to obtain a calculation result with the weight assigned; wherein, the interacting timing data represents the timing data that has a preset interaction relationship according to historical experience.

[0062] As a possible implementation, the first calculation module 42 can be configured to: extract feature parameters in the timing data of a single braking event to obtain the feature parameters in the single braking event; calculate the wear score of the single braking event according to the preset weight of the feature parameters in the single braking event.

[0063] As a possible implementation, the device 4 for health monitoring of vehicle brake pads can be configured to: normalize the feature parameters to a preset range to obtain the normalized feature parameters; wherein, the first calculation module 42 can also be configured to: calculate the wear score of the single braking event according to the normalized feature parameters and the preset weight of the normalized feature parameters in the single braking event.

[0064] As a possible implementation, the timing data includes vehicle speed, longitudinal acceleration, lateral acceleration, and brake pedal depth; the first calculation module 42 can also be configured to: extract the initial speed in a single braking event according to the vehicle speed; extract the average longitudinal acceleration in the single braking event according to the longitudinal acceleration of the vehicle; extract the standard deviation of the lateral acceleration in the single braking event according to the lateral acceleration of the vehicle; extract the average brake depth and duration in the single braking event according to the brake pedal depth of the vehicle; use the initial speed, the average longitudinal acceleration, the standard deviation of the lateral acceleration, the average brake depth, and the duration as the feature parameters in the single braking event.

[0065] As a possible implementation, the third calculation module 44 can be configured to: compare the total wear score with a preset score value to obtain the health degree; wherein, the smaller the difference between the total wear score and the preset score value, the lower the health degree.

[0066] As a possible implementation, the device 4 for health monitoring of vehicle brake pads can be configured to: when the depth of the brake pedal is greater than or equal to a first preset threshold, determine that a single braking event starts; when the depth of the brake pedal is less than a second preset threshold, determine that the single braking event ends.

[0067] Regarding the device in the above embodiments, the specific manner in which each unit performs operations has been described in detail in the embodiments related to the method, and will not be elaborated in detail here.

[0068] An apparatus for health monitoring of vehicle brake pads, the electronic device includes: one or more processors; a memory for storing processor-executable instructions; one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the vehicle brake pad health monitoring method provided by this application.

[0069] Next, refer to Figure 5 to describe the electronic device according to an embodiment of this application. The electronic device can be either the first device or the second device, or both, or a stand-alone device independent of them, and the stand-alone device can communicate with the first device and the second device to receive the input signals collected from them.

[0070] Figure 5 The block diagram of the electronic device according to an embodiment of this application is illustrated.

[0071] As Figure 5 shown, the electronic device 50 includes one or more processors 51 and a memory 52.

[0072] The processor 51 can be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and can control other components in the electronic device 50 to perform desired functions.

[0073] The memory 52 can include one or more computer program products, and the computer program products can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory can include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory can include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions can be stored on the computer-readable storage media, and the processor 51 can run the program instructions to implement the vehicle brake pad health monitoring methods of various embodiments of this application described above and / or other desired functions. Various contents such as input signals, signal components, noise components, etc. can also be stored in the computer-readable storage media.

[0074] In one example, the electronic device 50 can further include: an input device 53 and an output device 54, and these components are interconnected through a bus system and / or other forms of connection mechanisms (not shown).

[0075] When the electronic device is a stand-alone device, the input device 53 can be a communication network connector for receiving the input signals collected from the first device and the second device.

[0076] In addition, the input device 53 may further include, for example, a keyboard, a mouse, and the like.

[0077] The output device 54 can output various information to the outside, including the determined distance information, direction information, etc. The output device 54 may include, for example, a display, a speaker, a printer, a communication network, and remote output devices connected thereto, and the like.

[0078] Of course, for simplicity, Figure 5 only some of the components related to the present application in the electronic device 50 are shown, and components such as a bus, an input / output interface, and the like are omitted. In addition, according to specific application scenarios, the electronic device 50 may further include any other appropriate components.

[0079] The computer program product may be written in any combination of one or more programming languages for programming code to perform the operations of the embodiments of the present application. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, executed as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0080] A computer-readable storage medium stores a computer program for executing the method for health monitoring of vehicle brake pads provided by the embodiments of the present application.

[0081] The computer-readable storage medium may employ any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, include but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0082] The above description has been given for purposes of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions, and sub-combinations thereof.

Claims

1. A method for monitoring the health of a vehicle brake pad, characterized in that: The vehicle brake pad health monitoring method comprises: In a preset time period for brake pad health monitoring, time series data of any single braking event is collected; wherein the time series data represents data reflecting the vehicle state and behavior in the braking behavior, and the preset time period for brake pad health monitoring includes from the start of brake pad replacement to any time point; Calculate and obtain a wear score of a single braking event according to a preset weight of the time series data in a single braking event; Add the wear scores of a single braking event in the preset time period of brake pad health monitoring to obtain a total wear score; The health of the brake pad is calculated according to the total wear score; wherein the health reflects the health degree of the brake pad.

2. The vehicle brake pad health monitoring method according to claim 1, characterized in that: According to the preset weight of the time series data in a single braking event, the wear score of the single braking event is calculated, including: In a single braking event, the time series data is multiplied by the preset weight to obtain a weighted calculation result; When there are at least two weighted calculation results, the weighted calculation results are added together to calculate the wear score of a single braking event.

3. The vehicle brake pad health monitoring method according to claim 2, characterized in that: In a single braking event, the time series data is multiplied by the preset weight to obtain the weighted calculation results, including: When there are interacting time series data in the time series data, the product of the time series data is obtained by multiplying the interacting time series data; The product of the time series data is multiplied by a preset weight to obtain a weighted calculation result; wherein the interactive time series data represents time series data with an interactive relationship preset based on historical experience.

4. The vehicle brake pad health monitoring method according to claim 1, characterized in that: According to the preset weight of the time series data in a single braking event, the wear score of the single braking event is calculated, including: Extracting features from the time series data in a single braking event to obtain feature parameters in the single braking event; The wear score of the single braking event is calculated according to the preset weight of the characteristic parameter in the single braking event.

5. The vehicle brake pad health monitoring method according to claim 4, characterized in that: The vehicle brake pad health monitoring method further includes: Normalizing the characteristic parameters to a preset range to obtain normalized characteristic parameters; The wear score of a single braking event is calculated based on the preset weight of the characteristic parameter in a single braking event, including: The wear score of the single braking event is calculated based on the normalized characteristic parameters and the preset weights of the normalized characteristic parameters in the single braking event.

6. The vehicle brake pad health monitoring method according to claim 4, characterized in that: The time series data includes vehicle speed, longitudinal acceleration, lateral acceleration and brake pedal depth; feature extraction is performed on the time series data in a single braking event to obtain feature parameters in the single braking event, including: According to the vehicle speed, the initial speed in a single braking event is extracted; According to the longitudinal acceleration of the vehicle, the average longitudinal acceleration in a single braking event is extracted; According to the lateral acceleration of the vehicle, the standard deviation of the lateral acceleration in a single braking event is extracted; According to the brake pedal depth of the vehicle, the average braking depth and duration in a single braking event are extracted; The initial speed, the average longitudinal acceleration, the standard deviation of the lateral acceleration, the average braking depth and the duration are used as characteristic parameters in a single braking event.

7. The vehicle brake pad health monitoring method according to claim 1, characterized in that: Based on the total wear score, the health of the brake pad is calculated, including: The total wear score is compared with a preset score value to obtain a health degree; wherein, the smaller the difference between the total wear score and the preset score value, the lower the health degree.

8. A device for monitoring the health of a vehicle brake pad, characterized in that: The device for monitoring the health of the vehicle brake pad comprises: A collection module, used to collect time series data of any single braking event in a preset time period for brake pad health monitoring; wherein the time series data represents data reflecting the vehicle state and behavior in the braking behavior, and the preset time period for brake pad health monitoring includes from the start of brake pad replacement to any time point; A first calculation module, configured to calculate a wear score of a single braking event according to a preset weight of the time series data in the single braking event; The second calculation module is used to add the wear scores of single braking events in a preset time period of brake pad health monitoring to obtain a total wear score; The third calculation module is used to calculate the health of the brake pad according to the total wear score; wherein the health reflects the health level of the brake pad.

9. A device for monitoring the health of a vehicle brake pad, characterized in that: include: one or more processors; Memory; as well as One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the vehicle brake pad health monitoring method according to any one of claims 1 to 8.

10. A vehicle, characterized in that: include: Brake pads; The vehicle brake pad health monitoring device as claimed in claim 9, wherein the vehicle brake pad health monitoring device is communicatively connected to the brake pad.

Images