Method for predicting a fault of a temperature sensor of an electric

By collecting temperature sensor data in the vehicle electric drive system, using fault characteristic model and additional filtering conditions for fault pre-identification, the problem of insufficient temperature sensor fault warning in the prior art is solved, and early prediction of temperature sensor faults and Safety response improves vehicle safety and user satisfaction.

CN118906815APending Publication Date: 2024-11-08MERCEDES BENZ GRP
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411055037.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-02
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The prior art has shortcomings in the early warning of temperature sensor failure of components of vehicle electrical drive system, which will only be repaired or replaced when the temperature sensor is completely invalid, affecting vehicle safety and user travel convenience.

Method used

By collecting temperature measurement data from the temperature sensor, pre-established fault characteristic model and additional filtering conditions for pre-identification of faults, and taking safety response measures when a fault is identified, such as notifying the user for repairs, to minimize the complete failure of the temperature sensor to the greatest extent possible due to complete failure of the temperature sensor. Caused by abnormal motor operation and vehicle failure to start.

Benefits of technology

Early prediction of temperature sensor failures is achieved, the accuracy and safety of fault predictions are improved, and the safety risks and maintenance costs are reduced due to temperature sensor failure.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118906815A_ABST
    Figure CN118906815A_ABST
Patent Text Reader

Abstract

The invention provides a method for pre-judging a fault of a temperature sensor of an electric driving system part of a vehicle. The method comprises the following steps: S1, collecting temperature measurement data of the temperature sensor of the electric driving system part of at least one vehicle; s2, pre-identifying a fault of the temperature sensor from the collected temperature measurement data by means of a pre-established fault feature model and at least one additional filtering condition; and S3, taking safety response measures for at least one vehicle under the condition that the temperature sensor fault is pre-recognized. The application also proposes a device for pre-determining a temperature sensor failure of an electric drive system component of a vehicle, a vehicle, a server and a computer program product. In the application, the fault of the motor temperature sensor can be pre-judged in advance, and the additional filtering condition is considered in the pre-recognition period, so that the screening accuracy is further improved, and the misjudgment probability is reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to a method for predicting temperature sensor failure of a vehicle's electric drive system component. The present application also relates to a device for predicting temperature sensor failure of a vehicle's electric drive system component, a vehicle, a server, and a computer program product. Background Art

[0002] The health of the vehicle's electric drive system components is crucial to the correct operation of the vehicle. Currently, there is no effective early warning for temperature sensors. Usually, the vehicle system will only know that the temperature sensor has failed when it fails completely. If the temperature sensor is repaired or replaced at this time, it is often too late, because the failure of the temperature sensor will cause the drive motor to work abnormally and the vehicle cannot start, which not only affects the safety of the vehicle, but also brings inconvenience to the user's normal travel and causes user dissatisfaction.

[0003] Therefore, the existing technology still has deficiencies in temperature sensor failure warning of electric drive system components. Summary of the invention

[0004] The purpose of the present application is to provide a method for predicting temperature sensor failure of a vehicle's electric drive system component, a device for predicting temperature sensor failure of a vehicle's electric drive system component, a vehicle, a server and a computer program product, so as to at least solve some of the problems in the prior art.

[0005] According to a first aspect of the present application, a method for predicting a temperature sensor failure of an electric drive system component of a vehicle is provided, the method comprising the following steps:

[0006] S1: Collecting temperature measurement data of a temperature sensor of at least one component of an electric drive system of a vehicle;

[0007] S2: pre-identifying a fault of the temperature sensor from the collected temperature measurement data by means of a pre-established fault characteristic model and at least one additional filtering condition; and

[0008] S3: In case of pre-identification of a temperature sensor failure, taking safety response measures for the at least one vehicle.

[0009] The present application particularly includes the following technical concepts: With the help of the method according to the present application, the abnormal state of the motor temperature sensor can be predicted in advance. In particular, for gradual fault types (such as poor contact caused by corrosion, etc.), a reliable monitoring list can be created very early, and the customer can be notified for timely repairs when necessary, so as to minimize the situation where the motor fails completely due to the complete failure of the temperature sensor and the vehicle cannot start. In addition, during the execution of pre-identification, additional filtering conditions are also considered to further improve the screening accuracy and reduce the probability of misjudgment.

[0010] In an exemplary embodiment, in step S2, only when the at least one additional filtering condition is met, it is permitted to pre-identify the fault of the temperature sensor with the help of the fault characteristic model and / or it is permitted to draw a conclusion that the temperature sensor fault is pre-identified. By taking the additional filtering condition as a prerequisite for fault pre-identification, the data processing efficiency and the accuracy of fault prejudgment are improved.

[0011] In an exemplary embodiment, the at least one additional filtering condition is associated with temperature measurement data of another temperature sensor of another electric drive system component other than the electric drive system component, wherein the additional filtering condition includes: the temperature measurement data of the temperature sensor and the temperature measurement data of another temperature sensor do not meet the data consistency requirement. By verifying the data of multiple sensors, false alarms caused by accidental fluctuations in single sensor data can be reduced, and real faults can also be avoided from being missed.

[0012] In an exemplary embodiment, the at least one additional filtering condition includes: the driving environment temperature of the vehicle is lower than a predetermined environment temperature. By considering the influence of the environment temperature, the possible failure modes under low temperature conditions can be more accurately identified, thereby ensuring the reliability of the diagnosis result.

[0013] In an exemplary embodiment, the at least one additional filtering condition includes: a predetermined type of RTM level 2 alarm of the vehicle is triggered, and the predetermined type of RTM level 2 alarm reflects a driving environment condition that causes an abnormal temperature sensor of an electric drive system component. By incorporating the RTM level 2 alarm into the fault diagnosis process, false alarms can be reduced more accurately, which helps to improve the accuracy of fault prediction.

[0014] In an exemplary embodiment, the at least one additional filtering condition includes: when there is no thermal insulation between the high-voltage interlock plug of the vehicle and the low-voltage plug of the temperature sensor of the electric drive system component, the RTM high-voltage interlock alarm of the vehicle is triggered. Advantageously, for vehicles with a specific motor structure type, a more accurate fault pre-identification result can be obtained.

[0015] In an exemplary embodiment, the electric drive system components include a drive motor stator and / or a motor controller. Thus, key components in the electric drive system are covered, ensuring the comprehensiveness of fault warnings and enhancing system safety and reliability.

[0016] In an exemplary embodiment, the pre-established fault feature model is constructed to pre-identify the fault of the temperature sensor based on the following data features of the temperature measurement data: data credibility, the data credibility indicates whether the jump amplitude and / or data range of the temperature measurement data deviates from a predetermined level; and / or, data continuity, the data continuity indicates whether the temperature measurement data has frame loss, whether the frequency of frame loss exceeds a predetermined frequency, and / or whether the duration of frame loss exceeds a predetermined time threshold. By analyzing the data credibility, the model can distinguish between normal fluctuations and abnormal jumps, thereby more accurately predicting faults. By analyzing the data continuity, the model is allowed to respond quickly to potential faults.

[0017] In an exemplary embodiment, the method further includes the following steps: collecting known fault cases of temperature sensors of components of the electric drive system of the vehicle; for the known fault cases, extracting historical temperature measurement data collected by the temperature sensor within a predetermined time period before the failure node and performing data feature analysis on the historical temperature measurement data; and establishing a fault feature model based on the results of the data feature analysis. This ensures that the established fault feature model fits the actual usage and helps the model identify early signs of fault development in actual applications.

[0018] In an exemplary embodiment, in step S3, taking safety response measures includes: placing the at least one vehicle in a monitoring list; and / or, before the vehicle cannot be started due to a temperature sensor failure and / or the vehicle's RTM level 3 alarm is triggered due to a temperature sensor failure, notifying the vehicle user to repair the temperature sensor of the electric drive system component. By placing the vehicle in the monitoring list, vehicles that may have problems can be identified in advance, so that action can be taken before the failure occurs. By notifying the user in a timely manner, the safety risks caused by the failure of the temperature sensor can be reduced, and the maintenance costs can also be reduced.

[0019] In an exemplary embodiment, the method further includes the following steps: in the case of pre-identifying a temperature sensor failure of at least one vehicle, recording after-sales feedback for the at least one vehicle; and reversely verifying the pre-identification result of the temperature sensor failure with the help of the after-sales feedback, and updating and / or adjusting the fault feature model based on the reverse verification result. In this way, the fault feature model can be calibrated, false positives and false negatives can be reduced, and continuous optimization of temperature sensor failure warning can be achieved.

[0020] According to a second aspect of the present application, a device for predicting temperature sensor failure of components of an electric drive system of a vehicle is provided, wherein the device comprises a memory and a processor, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the processor is capable of executing the method described in the first aspect of the present application.

[0021] According to a third aspect of the present application, a vehicle is provided, comprising the device according to the second aspect of the present application.

[0022] According to a fourth aspect of the present application, a server is provided, wherein the server comprises the device described in the second aspect of the present application.

[0023] According to a fifth aspect of the present application, a computer program product having a program code unit is provided, wherein the program code unit is configured to cause the computer to execute the method according to the first aspect of the present application when the computer program product is run on a computer or stored on a computer-readable storage medium. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present application will be described in more detail below with reference to the accompanying drawings, so that the principles, features and advantages of the present application can be better understood. The accompanying drawings include:

[0025] Figure 1 A block diagram of a device for predicting a temperature sensor failure of a component of an electric drive system of a vehicle according to an exemplary embodiment of the present application is shown;

[0026] Figure 2 A flow chart showing a method for predicting a temperature sensor failure of an electric drive system component of a vehicle according to an exemplary embodiment of the present application is shown;

[0027] Figure 3 Shows Figure 2 a flow chart of additional steps of the illustrated method;

[0028] Figure 4 Shows Figure 2 a flow chart of further additional steps of the method shown; and

[0029] Figure 5 A graph showing temperature measurement data of temperature sensors of electric drive system components of a vehicle versus time. DETAILED DESCRIPTION

[0030] In order to make the technical problems, technical solutions and beneficial technical effects to be solved by the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and multiple exemplary embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the scope of protection of the present application.

[0031] Figure 1 A block diagram of a device for predicting temperature sensor failure of a component of an electric drive system of a vehicle according to an exemplary embodiment of the present application is shown.

[0032] like Figure 1 As shown, the device 10 is, for example, arranged in a server 2, which may also be referred to as a remote server or a vehicle backend server in the context. The device 10 includes a processor 101 and a memory 102. The memory 102 stores computer program instructions. When the computer program instructions are executed by the processor 101, the processor 101 can, for example, execute a method for predicting a temperature sensor failure of a component of an electric drive system of a vehicle (which will be described below with reference to Figure 2 Detailed description). The computer program instructions may be stored in a computer-readable storage medium. The computer-readable storage medium may include, for example, a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card, a secure digital card, a flash memory card, at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. The processor may be a central processing unit, or may be other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0033] exist Figure 1 1 and 2 , a plurality of vehicles 100, 200, 300 are also shown for interacting with the device 10. In the context, a "vehicle" includes, for example, a hybrid electric vehicle (HEV), a plug-in hybrid electric vehicle (PHEV), an extended range electric vehicle (EREV), and a battery electric vehicle (BEV).

[0034] The vehicles 100, 200, 300 include an electric drive system, two components of which are shown here as an example, namely, a drive motor 21 for pulling the vehicle and a motor controller 22 (such as an inverter) for controlling the motor 21. In addition, the electric drive system also includes temperature sensors 41, 42. Figure 12 shows a first temperature sensor 41 for the drive motor 21 and a second temperature sensor 42 for the motor controller 22. The first temperature sensor 41 is, for example, a drive motor stator temperature sensor, which is responsible for measuring the actual temperature of the stator winding of the drive motor 21. Generally, NTC (negative temperature coefficient) type thermistor sensors can be used as temperature sensors 41 and 42, and as the temperature changes, the resistance value of the thermistor sensor also changes accordingly.

[0035] Vehicles 100, 200, 300, for example, also have a real-time monitoring module 30 (RTM for short). The real-time monitoring module 30, the first temperature sensor 41, and the second temperature sensor 42 are respectively connected to the vehicle communication bus (e.g., CAN bus). Therefore, the real-time monitoring module 30 can collect the temperature measurement data detected by the first temperature sensor 41 and the second temperature sensor 42, and then report it to the server 2. The vehicle data collected and reported in this way can also be called real-time monitoring data or RTM data. In addition to the temperature measurement data of the temperature sensors 41 and 42 of the electric drive system components, the RTM data can also include insulation resistance data, vehicle position data, battery temperature data, battery voltage data, battery current data, battery state of charge data (SOC) and / or battery health status data (SOH), etc.

[0036] Vehicles 100, 200, 300 may be, for example, private cars, which report vehicle data for data sharing purposes or in accordance with national monitoring standards. In addition, vehicles 100, 200, 300 may also be test vehicles (teams) dedicated to testing purposes.

[0037] In an embodiment not shown, the device 10 may also be arranged locally in the vehicle 100 , 200 , 300 , and may perform a pre-diagnosis of a temperature sensor failure of the vehicle 100 , 200 , 300 itself.

[0038] Figure 2 A flow chart of a method for predicting a temperature sensor failure of a component of an electric drive system of a vehicle according to an exemplary embodiment of the present application is shown. The method exemplarily includes steps S1-S3, wherein step S2 is further shown to include sub-steps S21-S24. Figure 2 The method shown can be used, for example, in Figure 1 The device 10 is shown as being implemented in the context of the present invention.

[0039] In step S1 , temperature measurement data of a temperature sensor of at least one component of an electric drive system of a vehicle are collected.

[0040] As combined Figure 1As has been described, vehicle data reported by the real-time monitoring module can be received from one or more vehicles, wherein the vehicle data includes corresponding temperature measurement data. In addition, the vehicle data can also be actively requested from the vehicle.

[0041] The “electric drive system components” of a vehicle can be understood as components of the system that drives the vehicle through electricity. The components mainly involved in this article are, for example, the drive motor stator and the motor controller.

[0042] Next, in step S2, the fault of the temperature sensor is pre-identified from the collected temperature measurement data with the aid of a pre-established fault characteristic model and at least one additional filter condition. Figure 2 In the illustrated embodiment, step S2 is shown to include sub-steps S21 - S24 .

[0043] In sub-step S21, the temperature measurement data is analyzed with the aid of a pre-established fault feature model to determine whether a fault data feature is identified.

[0044] In this context, "pre-identification" can be understood as predicting the impending failure of the temperature sensor before it completely fails. It is understandable that the temperature sensor can still work at least partially normally for a period of time before it is completely damaged or fails, but during the operation, fault data characteristics may appear, or unobvious performance degradation may occur. By extracting these fault data characteristics or performance degradation and conducting corresponding analysis, the failure can be predicted as early as possible.

[0045] The fault characteristic model can be configured in different forms. In one embodiment, the fault characteristic model can be a physical model. For example, the physical characteristic model can be created and used with the help of empirical knowledge, the trend of changes in the collected historical temperature measurement data, and the data characteristic laws summarized or satisfied. The physical model can exist in the form of a table, in which one or more fault data characteristics are stored. When the data characteristics of the collected temperature measurement data match the fault data characteristics stored in the table, it is determined that the temperature sensor is about to fail. In another embodiment, the fault characteristic model can also be a machine learning model. The machine learning model can determine the fault information of the temperature sensor of the electric drive system components based on the training process of big data. The machine learning model can be based on, for example, a neural network, a support vector machine, and a Gaussian mixture model.

[0046] In one embodiment, the temperature measurement data may be analyzed in terms of data credibility and / or data continuity. Accordingly, a faulty data signature is a data signature that indicates untrustworthy and / or discontinuous data.

[0047] Data credibility is used to indicate whether the jump amplitude and / or data range of the temperature measurement data deviates from the predetermined level. For example, the temperature of the drive motor stator is generally in the range of 20°C to 40°C when it is at static room temperature and the coolant is working normally. If the temperature measurement data shows that the temperature of the drive motor stator is, for example, 200°C, the data is considered unreliable. For another example, the temperature of each component of the electric drive system parts generally does not jump significantly. If it is found that the instantaneous change amplitude of the drive motor stator temperature exceeds 150°C, the data is considered unreliable.

[0048] Data continuity is used to indicate whether frame loss occurs in the temperature measurement data, whether the frequency of frame loss exceeds a predetermined frequency, and / or whether the duration of frame loss exceeds a predetermined time threshold. Generally, the reading of the temperature sensor changes continuously, and even if the communication bus fails, it will not cause the temperature measurement data to be intermittent. Therefore, if "intermittent" or "frame loss" of temperature information is observed, it means that the data is unreliable. For example, under normal circumstances, the stator temperature sensor of the drive motor will send temperature readings at a fixed frequency (for example, once per second). Figure 1 The data stream received by the device 10 shown should be continuous. If the temperature sensor readings appear at irregular intervals, and such frame loss is not sporadic, but lasts for a certain period of time (for example, no temperature measurement data is recorded for 5 consecutive seconds) or has a certain pattern (for example, a set of data is lost every 10 seconds), it means that the data continuity requirement is not met.

[0049] If the fault data feature of the temperature sensor is not identified in sub-step S21, the identification is continued in this step. Otherwise, it is further checked in the next sub-step S22 whether the additional filtering condition is satisfied. In another embodiment not shown, it is also possible to check whether at least one additional filtering condition is satisfied before analyzing the temperature measurement data with the help of the fault feature model, and only when the at least one additional filtering condition is satisfied is it allowed to pre-identify the fault of the temperature sensor with the help of the fault feature model. In sub-step S22, for example, the temperature measurement data of the stator temperature sensor of the drive motor can be compared with the temperature measurement data of the motor controller temperature sensor, and it is determined whether the two meet the data consistency requirements.

[0050] Since there is a direct physical connection between the drive motor and the motor controller and / or cooling water circulates through them in sequence, there should be a certain correlation between the temperature readings of the drive motor stator temperature sensor and the motor controller temperature sensor. Therefore, the measurement results of the two temperature sensors can be used for cross-checking. For example, if it is found that the temperature measurement result of the drive motor stator is interrupted for more than 1 minute and the temperature measurement result of the motor controller remains continuous at the same moment, it is considered that the data consistency requirement is not met. For another example, if the deviation between the temperature measurement result of the drive motor stator and the temperature measurement result of the motor controller exceeds a certain degree at the same time or in the same time period, it can also be considered that the data consistency requirement is not met.

[0051] For example, if it is found in sub-step S22 that the temperature measurement data of the drive motor stator temperature sensor and the temperature measurement data of the motor controller temperature sensor are significantly different, it indicates that the temperature sensor is likely to be abnormal, and thus the additional filtering condition is met. With this additional filtering condition, false positive errors obtained only by model analysis of the motor stator temperature sensor data can be filtered out, reducing the misjudgment rate.

[0052] If it is determined in sub-step S22 that the data consistency requirement is met (i.e., the additional filtering condition is not met), then although the fault data characteristics of the temperature sensor have been identified with the help of the fault characteristic model, it is not possible to draw the conclusion that the temperature sensor is about to fail. Only when it is determined in sub-step S22 that the additional filtering condition is met, is it allowed to draw the conclusion that the temperature sensor fault is pre-identified.

[0053] Optionally, the pre-identification result can also be filtered based on another additional filtering condition in another sub-step S23. Specifically, it can be checked in this step whether the driving environment temperature of the vehicle is lower than the predetermined ambient temperature. If this is the case, it means that the vehicle is traveling in a relatively cold area, where the geographical conditions are more likely to cause the temperature sensors of the electric drive system components to fail. Therefore, the monitoring target range can be further narrowed with the help of this additional filtering condition to improve the recognition accuracy. For example, it can be determined in step S23 that the driving environment temperature of the vehicle is lower than the predetermined ambient temperature (for example, 0°C) before the conclusion that the temperature sensor fault has been pre-identified is allowed. On the contrary, even if the fault data characteristics of the temperature sensor have been identified with the help of the fault characteristic model, the confidence of such a result is considered to be low, and therefore the relevant sensor fault is not considered to be pre-identified.

[0054] Optionally, in another sub-step S24, the pre-identification result can be filtered based on another additional filtering condition, which is related to the triggering of the RTM secondary alarm of a predetermined type of vehicle. Specifically, for a specific type of vehicle (for example, if there is no thermal insulation between the high-voltage interlock plug of the vehicle and the low-voltage plug of the temperature sensor of the electric drive system component), it can be checked in this sub-step whether the RTM high-voltage interlock alarm of the vehicle is triggered. Only when the RTM high-voltage interlock alarm of the vehicle is triggered is it allowed to draw a conclusion that the temperature sensor fault is pre-identified.

[0055] In step S3 , in the event of a pre-identified temperature sensor failure, a safety response measure is taken for at least one vehicle.

[0056] In one embodiment, for example, at least one vehicle can be placed in a monitoring list to focus on the temperature sensor performance of the electric drive system components of the vehicle. For example, after-sales feedback can be recorded for all vehicles in the monitoring list, and the pre-identification results of the temperature sensors of these sensors can be reversely verified with the help of the after-sales feedback. Figure 3 The embodiment shown is described in detail. For another example, the production batch and model of the vehicle in the monitoring list can also be obtained to analyze the cause of the failure. The monitoring list can also be updated regularly (for example, monthly).

[0057] In another embodiment, before the vehicle cannot be started due to a temperature sensor failure and / or before the vehicle's RTM level 3 alarm is triggered due to a temperature sensor failure, the vehicle user is notified to repair the temperature sensor of the electric drive system component. This allows the user to make corresponding preparations in advance and minimize the impact of a complete failure of the temperature sensor.

[0058] Figure 3 Shows Figure 2 A flow chart of additional steps of the method shown. Figure 3 middle, Figure 2 The method shown also includes additional steps S4 and S5, wherein the additional step S4 is shown to include sub-steps S41-S45. These additional steps S4 and S5 may be, for example, Figure 2 Step S3 of the method shown is then performed.

[0059] exist Figure 3 In the illustrated embodiment, it is assumed that a vehicle with a pre-identified temperature sensor failure has been placed in a monitoring list.

[0060] Therefore, in the additional step S41, the subsequent after-sales feedback can be obtained for the vehicles listed in the monitoring list. For example, the maintenance records of vehicle users or the user complaints / inquiries received by the 4S store about vehicle abnormalities can be obtained regularly from the 4S store. The maintenance records collected by the 4S store are usually presented in the form of a quick inspection report, which records the corresponding fault code. This fault code may point to a fault related to the temperature sensor, or it may point to other types of faults.

[0061] In an additional step S42, the pre-identification result of the temperature sensor failure is reversely verified with the help of after-sales feedback to check the accuracy of the previous pre-identification result.

[0062] exist Figure 3 In the example, three reverse verification results are listed by means of steps S43-S45. For example, 16 vehicles are recorded in the monitoring list during the monitoring period. Figure 2 The process shown is selected from all the models sold and has been used earlier Figure 2 The process shown predicts a fault in a temperature sensor of an electric drive system component.

[0063] In step S43, it is found that 7 of the 16 vehicles have subsequently been repaired by 4S stores or the 4S stores have received telephone complaints / inquiries from customers. Through some conventional diagnostic methods, it has been confirmed that the abnormal vehicle condition is caused by the failure of the temperature sensor. Therefore, this reverse verification result shows that the pre-identification results for these 7 vehicles are accurate.

[0064] In step S44, it is found that 4 of the 16 vehicles also had a repair experience at the 4S store or the 4S store had received telephone complaints / inquiries from customers. However, these 4 vehicles were not classified as temperature sensor failures according to the detected fault codes / with conventional diagnostic methods at the time, but were mistakenly classified as other fault categories. However, in essence, the abnormal vehicle condition was still caused by the degradation of the temperature sensor. Therefore, this reverse verification result also shows that the pre-identification results for these 4 vehicles are also accurate.

[0065] In step S45, it is found that 5 of the 16 vehicles did not receive any after-sales feedback (i.e., there is no repair record or telephone complaint). In one embodiment, it can be considered that this reverse verification result indicates that it is temporarily uncertain whether the pre-identification results for these 5 vehicles are accurate, so these 5 vehicles can continue to be retained in the monitoring list. In another embodiment, it can also be considered that this reverse verification result indicates that the pre-identification results are wrong, so these 5 vehicles are removed from the monitoring list.

[0066] In step S5, the pre-established fault feature model can be updated and / or adjusted based on the reverse verification results obtained above. For example, if the reverse verification result shows that the accuracy of pre-identification is above 90%, the current fault feature model may not be changed for the time being. If the reverse verification result shows that the accuracy of pre-identification is lower than 90%, the corresponding parameters or built-in conditions of the model may be adjusted. For example, it was originally stipulated that when the temperature jump exceeds 10°C, it is considered that there is a risk of sensor failure, but due to the high false recognition rate, the amplitude threshold of the temperature jump can be increased to 20°C accordingly. For another example, originally the potential fault was identified only based on the temperature measurement results of the drive motor. Later, when making a judgment, the temperature measurement data of the power electronic device may be additionally introduced to improve the accuracy of pre-identification by using the cross-validation results.

[0067] Figure 4 Shows Figure 2 A flow chart of additional additional steps of the method shown. Figure 4 middle, Figure 2 The method shown also includes additional steps S01-S03, and these additional steps are intended to introduce how the fault characteristic model mentioned above is established.

[0068] In the additional step S01, known fault cases of the temperature sensors of the electric drive system components of the vehicle are collected. For example, the abnormal conditions of the vehicle reported by the user can be obtained from the 4S store, and then the known fault cases caused by the failure of the temperature sensors of the electric drive system components can be screened out from these abnormal conditions. For another example, the RTM level 3 alarm triggering situation of the vehicle can also be obtained from the vehicle backend server, and then the RTM level 3 alarms caused by the failure of the drive motor temperature sensor can be screened out according to the alarm flag bit in the RTM alarm data.

[0069] In an additional step S02, for a known fault case, historical temperature measurement data collected by the temperature sensor within a predetermined time period before the failed node is extracted and data feature analysis is performed on the historical temperature measurement data.

[0070] Here, the failure node can be defined as at least one of the following, for example: the moment when the RTM level 3 alarm is triggered due to a temperature sensor failure, the moment when the vehicle cannot be started due to a temperature sensor failure, and / or the moment when the user reports the failure. Due to different ways of collecting known failure cases, the definition of failure nodes may also be different. It should be noted that the failure node does not necessarily correspond directly to the failure moment of the sensor itself, but may also refer to the moment around the moment when the sensor completely fails (especially with a lag of some time). According to actual needs, data from 3-5 days before the failure node can be extracted, and this predetermined time period can also be set longer or shorter.

[0071] In one embodiment, the historical temperature measurement data within the predetermined time period can be extracted from the RTM data collected by the vehicle backend server. In another embodiment, the corresponding historical temperature measurement data can also be extracted from a data recorder (EDR) stored locally in the vehicle.

[0072] In the additional step S03, a fault feature model is established based on the results of the data feature analysis. Specifically, the historical temperature measurement data can be analyzed in terms of data continuity and / or data credibility, and the data feature commonalities or data feature laws in these historical temperature measurement data can be summarized, and then a physical model can be established accordingly. For another example, the historical temperature measurement data can also be used as data samples to train a machine learning model, thereby obtaining a trained fault feature model. If the fault feature model is based on machine learning, then its output results can include not only the fault pre-identification results (i.e., whether the sensor fault is pre-identified), but also the confidence of the pre-identification results and / or the remaining service life and expected failure time of the temperature sensor.

[0073] Figure 5 A graph showing temperature measurement data of temperature sensors of electric drive system components of a vehicle versus time.

[0074] exist Figure 5 , a first temperature curve 51 of the drive motor stator and a second temperature curve 52 of the motor controller are shown over time, which are measured by means of a first temperature sensor of the drive motor stator and a second temperature sensor of the motor controller, respectively. It can be seen that before time t0, the change trends and change amplitudes of the first temperature curve 51 and the second temperature curve 52 are basically consistent, and the numerical ranges are also roughly overlapped. This shows that the temperature of the drive motor and the temperature of the power electronic device meet the consistency requirements, and their respective temperature sensors perform well without any abnormal performance.

[0075] From the moment t0, the instantaneous temperature jump 511 of the first temperature curve 51 is gradually observed, and it is also observed that the first temperature curve 51 has a frame loss 512 in the time period t1 to t2, while in the time period t1 to t2, the second temperature curve 52 of the motor controller has a continuous value. In addition, at the moment t2, it is also detected that the RTM L2-9 alarm of the vehicle is triggered. This alarm does not directly indicate that the stator temperature of the drive motor or the temperature sensor of the motor controller is abnormal, but according to experience, this alarm has driving environment conditions that may cause the motor stator temperature sensor to be abnormal. Exemplarily, at the moment t2 or before the moment t2, the temperature sensor fault of the drive motor can be pre-identified by means of the fault feature model combined with at least one additional filtering condition. At this time, the temperature sensor of the drive motor stator has not completely failed, and the vehicle has not been started or the RTM level 3 alarm has been triggered due to severe performance degradation. At the moment t3 after the moment t2, the corresponding RTM level 3 alarm (such as RTM L3-8 alarm) is triggered due to the detection of excessively high temperature of the drive motor.

[0076] By comparison, it can be seen that the moment of pre-identification of the motor stator temperature sensor fault is significantly earlier than the actual failure moment t3 of the sensor. Through this kind of fault prediction, the vehicle user can be notified in advance to go to the 4S shop to repair or replace the temperature sensor in time. This improves the safety of the whole vehicle and the user satisfaction.

[0077] Although the specific embodiments of the present application are described in detail herein, they are provided for the purpose of explanation only and should not be considered to limit the scope of the present application. Various substitutions, changes and modifications may be conceived without departing from the spirit and scope of the present application.

Claims

1. A method for predicting a fault of a temperature sensor (41, 42) of an electric drive system component of a vehicle (100), the method comprising the following steps: S1: collecting temperature measurement data of a temperature sensor (41, 42) of an electric drive system component of at least one vehicle (100); S2: pre-identifying a fault of a temperature sensor (41, 42) from the collected temperature measurement data by means of a pre-established fault characteristic model and at least one additional filtering condition; as well as S3: In the event of a pre-identified failure of the temperature sensor (41, 42), taking safety response measures for the at least one vehicle (100).

2. The method according to claim 1, wherein: In step S2, only when the at least one additional filter condition is met is it permitted to pre-identify a fault of the temperature sensor (41, 42) by means of a fault characteristic model and / or is it permitted to draw a conclusion that a fault of the temperature sensor (41, 42) is pre-identified.

3. The method according to claim 1 or 2, wherein: The at least one additional filtering condition is associated with temperature measurement data of another temperature sensor of another electric drive system component other than the electric drive system component, wherein the additional filtering condition includes: the temperature measurement data of the temperature sensor and the temperature measurement data of another temperature sensor do not meet data consistency requirements.

4. The method according to any one of claims 1 to 3, wherein: The at least one additional filtering condition includes: the driving environment temperature of the vehicle (100) is lower than a predetermined environment temperature.

5. The method according to any one of claims 1 to 4, wherein: The at least one additional filtering condition includes: a predetermined type of RTM secondary alarm of the vehicle is triggered, and the predetermined type of RTM secondary alarm reflects a driving environment condition that causes an abnormality in a temperature sensor (41, 42) of an electric drive system component.

6. The method according to claim 5, wherein: The at least one additional filtering condition includes: when there is no thermal isolation between the high-voltage interlock plug of the vehicle (100) and the low-voltage plug of the temperature sensor of the electric drive system component, the RTM high-voltage interlock alarm of the vehicle is triggered.

7. The method according to any one of claims 1 to 6, wherein: The electric drive system components include a drive motor stator and / or a motor controller.

8. The method according to any one of claims 1 to 7, wherein: The pre-established fault characteristic model is constructed to pre-identify faults of the temperature sensor (41, 42) based on the following data characteristics of the temperature measurement data: Data credibility, the data credibility indicating whether the jump amplitude and / or data range of the temperature measurement data deviates from a predetermined level; and / or Data continuity, where the data continuity indicates whether frame loss occurs in the temperature measurement data, whether a frequency of frame loss exceeds a predetermined frequency, and / or whether a duration of frame loss exceeds a predetermined time threshold.

9. The method according to any one of claims 1 to 8, wherein: The method further comprises the following steps: Collecting known failure cases of temperature sensors (41, 42) of electric drive system components of a vehicle (100); For known fault cases, extracting historical temperature measurement data collected by temperature sensors (41, 42) within a predetermined time period before the failed node and performing data feature analysis on the historical temperature measurement data; and A fault feature model is established based on the results of data feature analysis.

10. The method according to any one of claims 1 to 9, wherein: In step S3, taking security response measures includes: placing the at least one vehicle (100) on a monitoring list; and / or Before the vehicle (100) cannot be started and / or the RTM level 3 alarm of the vehicle (100) is triggered due to a fault in the temperature sensor (41, 42), the vehicle user is notified to repair the temperature sensor (41, 42) of the electric drive system component.

11. The method according to any one of claims 1 to 10, wherein: The method further comprises the following steps: In the event of a pre-identified failure of a temperature sensor (41, 42) of at least one vehicle (100), recording after-sales feedback for the at least one vehicle (100); and The pre-identification result of the temperature sensor (41, 42) fault is reversely verified with the help of after-sales feedback, and the fault characteristic model is updated and / or adjusted based on the reverse verification result.

12. A device (10) for predicting failure of a temperature sensor (41, 42) of an electric drive system component of a vehicle (100), wherein: The device (10) comprises a processor (101) and a memory (102), wherein the memory (102) stores computer program instructions, and when the computer program instructions are executed by the processor (101), the processor (101) is capable of performing the method according to any one of claims 1 to 11.

13. A vehicle (100) comprising a device (10) according to claim 12.

14. A server (2) comprising the device (10) according to claim 12. 15 . A computer program product having program code means, which are configured to cause a computer to execute the method according to claim 1 , when the computer program product is run on a computer or is stored on a computer-readable storage medium.

Citation Information

Cited By

  • High-temperature sensor control method for severe environment monitoring

    CN119124403A