A method for identifying a cooling abnormality of a vehicle power module

Abstract

Embodiments of the present application provide a cooling abnormality identification method of a vehicle power module, and relate to the field of electric vehicles. The method comprises: obtaining buried point data of a plurality of vehicles; determining an actual temperature parameter of each vehicle and a data characteristic time sequence of each vehicle according to an electric drive signal of each vehicle and a driving state signal of each vehicle; obtaining an electric drive fitting curve characteristic table of each vehicle; fitting the temperature parameter by using the electric drive fitting curve characteristic table of each vehicle according to preset electric drive working condition parameters of each vehicle to obtain a reference temperature parameter of each vehicle; and performing cooling abnormality identification on a power module in each vehicle according to the reference temperature parameter, the actual temperature parameter and the data characteristic time sequence of each vehicle to obtain a cooling abnormality risk parameter. The present application can improve the cooling abnormality identification speed of the vehicle power module, and improve the anti-interference ability, the precision and the stability under a large number of vehicle data.

Description

Technical Field

[0001] This application relates to the field of electric vehicles, and more specifically, to a method for identifying cooling anomalies in automotive power modules. Background Technology

[0002] Automotive power modules are typically integrated into the inverters of new energy vehicles, functioning within the electric drive control module. They convert DC input to AC output to drive the motor. These modules operate in a high-voltage, high-current environment, using high-frequency switching of transistors to meet the speed and torque output requirements of the drive motor, enabling precise control and protection. However, due to the inherent characteristics of automotive power modules, abnormal cooling can lead to overheating, affecting their lifespan, increasing power losses, and reducing reliability. Simultaneously, the inverter's power output is limited, significantly reducing overall vehicle performance, such as acceleration. In severe cases, this can damage both the motor and the inverter, leading to unpredictable consequences.

[0003] In the existing technology, when automotive power modules experience cooling anomalies, most methods for identifying cooling anomalies rely on iterative processing of a single feature dimension based on the experience of human experts or the input of precise bench test data, in order to solve the problem of cooling anomalies in automotive power modules.

[0004] However, due to the lack of feature dimension data under different operating conditions of a large number of vehicles, this cooling anomaly identification method is easily affected by abnormal data values, resulting in poor anti-interference ability and poor stability of the vehicle power module. Furthermore, relying on the experience of human experts or the input of precise bench test data leads to poor adaptability and low reliability of the cooling anomaly identification method for the vehicle power module. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this application is to provide a method for identifying cooling anomalies in automotive power modules, which can monitor the cooling status of power modules in a large number of vehicles, while avoiding the influence of abnormal data values, thereby improving the anti-interference capability, accuracy and stability of identifying cooling anomalies in automotive power modules.

[0006] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows:

[0007] In a first aspect, embodiments of this application provide a method for identifying cooling anomalies in automotive power modules, the method comprising:

[0008] Acquire embedded data from multiple vehicles, whereby the embedded data for each vehicle includes: the electric drive signal of each vehicle and the driving status signal of each vehicle.

[0009] Based on the electric drive signal and driving status signal of each vehicle, the actual temperature parameter and data feature time series of each vehicle are determined, wherein the actual temperature parameter is used to characterize the actual temperature of the power module in the electric drive system of each vehicle.

[0010] Obtain the electric drive fitting curve characteristic table for each vehicle. The electric drive fitting curve characteristic table is used to characterize the relationship curve between the electric drive operating parameters and temperature parameters for each vehicle.

[0011] Based on the preset electric drive operating parameters of each vehicle, the temperature parameters are fitted using the electric drive fitting curve characteristic table of each vehicle to obtain the reference temperature parameters of each vehicle.

[0012] Based on the reference temperature parameters of each vehicle, the actual temperature parameters of each vehicle, and the data feature time series of each vehicle, cooling anomalies are identified in the power modules of each vehicle to obtain cooling anomaly risk parameters for each vehicle. The cooling anomaly risk parameters for each vehicle are used to characterize the risk of abnormal cooling in the power modules of each vehicle.

[0013] Optionally, determining the actual temperature parameters of each vehicle and the data characteristic time series of each vehicle based on the electric drive signal and the driving state signal of each vehicle includes:

[0014] The electric drive characteristic parameters of each vehicle at multiple time points are extracted from the electric drive signal of each vehicle. The electric drive characteristic parameters include: actual electric drive operating condition parameters and actual temperature parameters.

[0015] Extract the driving characteristic parameters of each vehicle at multiple time points from the driving status signal of each vehicle;

[0016] Based on the electric drive characteristic parameters and driving characteristic parameters at the multiple time points, a data feature time series for each vehicle is generated.

[0017] Optionally, the driving characteristic parameters include at least: a driving start / stop indicator;

[0018] The method further includes:

[0019] Based on the driving start-stop identifier in the driving characteristic parameters, the data feature time series of each vehicle are spliced ​​together to obtain a data segment of each vehicle under at least one driving cycle condition.

[0020] The step involves identifying cooling anomalies in the power modules of each vehicle based on reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, thereby obtaining cooling anomaly risk parameters for each vehicle, including:

[0021] Based on the reference temperature parameters of each vehicle, the actual temperature parameters of each vehicle, and the data segments of each vehicle under each driving cycle condition, the cooling anomaly of the power module in each vehicle under each driving cycle condition is identified, and the cooling anomaly risk parameters of each vehicle under each driving cycle condition are obtained.

[0022] Optionally, the preset electric drive operating condition parameters for each vehicle are: the preset electric drive operating condition parameter range for each vehicle;

[0023] The step of fitting temperature parameters to each vehicle using the electric drive fitting curve characteristic table of each vehicle based on the preset electric drive operating condition parameters of each vehicle to obtain reference temperature parameters for each vehicle includes:

[0024] The preset electric drive operating condition parameter range for each vehicle is segmented to obtain multiple electric drive operating condition sub-ranges for each vehicle.

[0025] Based on the multiple electric drive operating condition sub-intervals, the temperature parameters of each vehicle are fitted in the multiple electric drive operating condition sub-intervals using the electric drive fitting curve characteristic table of each vehicle, so as to obtain the reference temperature parameters of each vehicle in the multiple electric drive operating condition sub-intervals.

[0026] The step involves identifying cooling anomalies in the power modules of each vehicle based on reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, thereby obtaining cooling anomaly risk parameters for each vehicle, including:

[0027] Based on the actual electric drive operating condition parameters of each vehicle, the target reference temperature parameter of the target operating condition parameter sub-interval where the actual electric drive operating condition parameters are located is determined from the reference temperature parameters under each electric drive operating condition sub-interval.

[0028] Based on the target reference temperature parameter, the actual temperature parameter of each vehicle, and the data feature time series of each vehicle, cooling anomaly identification is performed on the power module in each vehicle to obtain the cooling anomaly risk parameter of each vehicle.

[0029] Optionally, the step of identifying cooling anomalies in the power modules of each vehicle based on the reference temperature parameters of each vehicle, the actual temperature parameters of each vehicle, and the data feature time series of each vehicle, to obtain cooling anomaly risk parameters for each vehicle, includes:

[0030] Based on the reference temperature parameters and the actual temperature parameters of each vehicle, abnormal data points of each vehicle are determined from the data feature time series of each vehicle.

[0031] Determine the proportion of outlier data in the time series of data features for each vehicle;

[0032] Based on the proportion of abnormal data, the cooling abnormality risk parameters for each vehicle are determined.

[0033] Optionally, the reference temperature parameter for each vehicle includes: each reference temperature parameter within the reference temperature range for each vehicle;

[0034] The step of determining abnormal data points for each vehicle from the data feature time series of each vehicle based on the reference temperature parameters and the actual temperature parameters of each vehicle includes:

[0035] The actual temperature parameter of each vehicle and the temperature parameter difference of each reference temperature parameter within the reference temperature range of each vehicle are determined respectively to obtain multiple temperature deviation values ​​for each vehicle;

[0036] Determine the target temperature deviation value with the largest absolute value among multiple temperature deviation values ​​for each vehicle;

[0037] If the target temperature deviation value is greater than the preset temperature deviation threshold, the data features of the target time point corresponding to the actual temperature parameter of each vehicle in the data feature time series of each vehicle are marked as the abnormal data points.

[0038] Optionally, determining the cooling anomaly risk parameter for each vehicle based on the proportion of abnormal data includes:

[0039] Based on the proportion of abnormal data, determine the current risk factor coefficient of the data feature time series of each vehicle;

[0040] Based on the current risk factor coefficients and the data feature time series of each vehicle, generate a current segment detail table for each vehicle;

[0041] The cooling anomaly risk parameters for each vehicle are determined based on the current segment details table and the historical risk factor coefficients for each vehicle.

[0042] Optionally, obtaining the electric drive fitting curve characteristic table for each vehicle includes:

[0043] If the electric drive fitting curve characteristic table of the first vehicle is stored in advance, then the electric drive fitting curve characteristic table of the first vehicle is obtained.

[0044] If the electric drive characteristic table of the second vehicle is not stored in advance, then based on the preset electric drive operating condition parameters of the second vehicle, the data characteristic time series of the second vehicle, and multiple historical data characteristic sequences of the second vehicle, the relationship curve between the electric drive operating condition parameters and temperature parameters of the second vehicle is fitted to obtain the electric drive fitting curve characteristic table of the second vehicle.

[0045] Optionally, the preset electric drive operating condition parameters of the second vehicle are: the preset electric drive operating condition parameter range of the second vehicle;

[0046] The step involves fitting a curve between the electric drive operating parameters and temperature parameters of the second vehicle based on preset electric drive operating parameters, the data feature time series of the second vehicle, and multiple historical data feature sequences of the second vehicle, to obtain an electric drive fitting curve characteristic table for the second vehicle, including:

[0047] The preset electric drive operating condition parameter range of the second vehicle is segmented to obtain multiple second electric drive operating condition sub-ranges of the second vehicle.

[0048] Based on the multiple second electric drive condition sub-intervals, the data feature time series of the second vehicle and the multiple historical data feature sequences of the second vehicle are grouped respectively to obtain the training dataset of the second vehicle under the multiple second electric drive condition sub-intervals.

[0049] Based on the training dataset under the second electric drive operating condition sub-interval, the relationship curves between electric drive operating condition parameters and temperature parameters are fitted under the multiple second electric drive operating condition sub-intervals respectively, and the characteristic table of electric drive fitting curves of the second vehicle under the multiple second electric drive operating condition sub-intervals is obtained.

[0050] Optionally, the step of fitting the relationship curves between the electric drive operating condition parameters and temperature parameters in the plurality of second electric drive operating condition sub-intervals based on the training dataset under the second electric drive operating condition sub-intervals, to obtain the electric drive fitting curve characteristic table of the second vehicle under the plurality of second electric drive operating condition sub-intervals, includes:

[0051] Monte Carlo simulations were performed on the training dataset for each second electric drive operating condition sub-interval to obtain the parameter fitting curve and fitting function for each second electric drive operating condition sub-interval.

[0052] Based on the parameter fitting curve and fitting function of each second electric drive operating condition sub-interval, a characteristic table of electric drive fitting curves of the second vehicle under each second electric drive operating condition sub-interval is generated.

[0053] Compared with the prior art, this application has the following beneficial effects:

[0054] The embodiments of this application provide a method for identifying cooling anomalies in automotive power modules. By acquiring embedded data from multiple vehicles and determining the actual temperature parameters and data feature time series of each vehicle based on the embedded data, the cloud server in the cloud can acquire data from a large number of vehicles, thereby avoiding the influence of abnormal data values ​​in the vehicle's embedded data and improving the stability and accuracy of vehicle cooling anomaly identification. The system acquires the electric drive fitting curve characteristic table for each vehicle. Based on this table, it fits the temperature parameters to the preset electric drive operating conditions for each vehicle to obtain reference temperature parameters. Data calculations and simulations are then performed based on different electric drive operating conditions for each vehicle, resulting in reference temperature parameters with good adaptability and high reliability. Simultaneously, based on the reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, cooling anomalies in the power modules of each vehicle are identified, yielding cooling anomaly risk parameters. This allows for the accumulation of cooling anomaly risk parameters over time, effectively enabling early warning based on the reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, resulting in high value density and utilization. Attached Figure Description

[0055] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 An example illustration of a cooling anomaly identification system for an automotive power module provided in this application embodiment. Figure 1 ;

[0057] Figure 2 An example illustration of a cooling anomaly identification system for an automotive power module provided in this application embodiment. Figure 2 ;

[0058] Figure 3 An example illustration of a cooling anomaly identification system for an automotive power module provided in this application embodiment. Figure 3 ;

[0059] Figure 4 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 1 ;

[0060] Figure 5 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 2 ;

[0061] Figure 6 A schematic diagram of the preprocessing of a cooling anomaly identification method provided in an embodiment of this application;

[0062] Figure 7 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 3 ;

[0063] Figure 8 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 4 ;

[0064] Figure 9 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 5 ;

[0065] Figure 10 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 6 ;

[0066] Figure 11 A schematic diagram of the theoretical temperature curve for a cooling anomaly identification method provided in this application embodiment;

[0067] Figure 12 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 7 ;

[0068] Figure 13 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 8 ;

[0069] Figure 14 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 9 ;

[0070] Figure 15 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 10 ;

[0071] Figure 16 A schematic diagram of a cooling anomaly identification device for an automotive power module provided in this application embodiment;

[0072] Figure 17 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0073] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can typically be arranged and designed in various different configurations.

[0074] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0075] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0076] In the description of this application, it should be noted that if terms such as "upper," "lower," "inner," or "outer" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of the invention is usually placed during use, they are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0077] Furthermore, the terms "first" and "second" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0078] It should be noted that, where there is no conflict, the features in the embodiments of this application can be combined with each other. Furthermore, it should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of a feature subsequently declared, but does not preclude the addition of other features.

[0079] Current methods for identifying cooling anomalies in automotive power modules suffer from a lack of feature data from a large number of vehicles, making them susceptible to abnormal data values. This results in poor anti-interference capability, low accuracy, and poor stability of the automotive power modules. Therefore, this application provides a method for identifying cooling anomalies in automotive power modules to avoid the influence of abnormal data values, thereby improving the speed, anti-interference capability, accuracy, and stability of cooling anomaly identification.

[0080] To clearly describe the cooling anomaly identification method for automotive power modules provided in this application embodiment, an exemplary cooling anomaly identification system for automotive power modules will be described in detail with reference to the accompanying drawings. Figure 1 An example illustration of a cooling anomaly identification system for an automotive power module provided in this application embodiment. Figure 1 .like Figure 1 As shown, the cooling anomaly identification system 100 for the vehicle power module may include: multiple vehicle terminals 110 and a cloud terminal 120.

[0081] Multiple vehicle terminals 110 can connect to the cloud 120 via a preset communication method to transmit data collected from the multiple vehicle terminals 110 to the cloud 120 for data storage and processing. The preset communication method can be selected according to actual conditions; for example, the preset communication method can be selected as the transmission of the vehicle terminal's wireless gateway T-box.

[0082] Alternatively, in one possible implementation, Figure 2 An example illustration of a cooling anomaly identification system for an automotive power module provided in this application embodiment. Figure 2 ,like Figure 2 As shown, each vehicle terminal 110 may include: an electric drive unit 111, a vehicle control unit 112, a driving status unit 113, a central gateway unit 114, and a communication unit 115. The electric drive unit 111 is communicatively connected to the vehicle control unit 112 to facilitate data interaction between them; the driving status unit 113 is also communicatively connected to the vehicle control unit 112 to facilitate data interaction between them; the electric drive unit 111 and the driving status unit 113 are also communicatively connected to the central gateway unit 114 to enable communication between the central gateway unit 114 and the electric drive unit 111 and the driving status unit 113; the central gateway unit 114 is connected to the cloud 120 via the communication unit 115 to upload data collected by the central gateway unit 114 to the cloud 120.

[0083] The electric drive unit 111 may include: a motor module 111-1, a cooling module 111-2, an automotive power module 111-3, and an electric drive control module 111-4; wherein, the motor module 111-1 can be used to provide the actual operating speed and torque of the motor; the cooling module 111-2 can be used to provide the cooling water flow rate and cooling water temperature; the automotive power module 111-3 can be used to provide the arm temperature, inverter temperature, etc. of the automotive power module; the electric drive control module 111-4 can be used to provide control signals or commands for the electric drive, such as cooling commands, torque commands, motor operating status, and cooling commands.

[0084] The automotive power module 111-3 is integrated into the inverter in the new energy vehicle, converting DC input into AC output to drive the motor. This automotive power module 111-3 can be selected according to actual needs; for example, it can be a power module composed of high-power semiconductor devices, such as IGBTs (Insulated Gate Bipolar Transistors). The electric drive control module 111-4 can also be selected according to actual needs; for example, it can be an MCU (Microcontroller Unit) control chip.

[0085] It should be noted that the electric drive unit 111 is communicatively connected to both the vehicle control unit 112 and the central gateway unit 114, using a high-speed CAN (Controller Area Network) bus for transmission; the driving status unit 113 is also communicatively connected to both the vehicle control unit 112 and the central gateway unit 114, using a low-speed CAN bus for transmission. Furthermore, it should be noted that in each vehicle's terminal 110, the sensor information and vehicle status characteristic information collected by each unit are transmitted to the central gateway unit 114 via the high- and low-speed CAN networks at different frequencies. The central gateway unit 114 aligns and packages the aggregated data according to the chronological order of timestamps, and then sends the packaged data packets for each vehicle to the cloud 120 via the communication unit 115 for storage and processing, thus completing the collection and transmission of raw data from the vehicle terminal 110.

[0086] The vehicle control unit 112 can be selected according to actual needs; for example, it can be an ECU (Electronic Control Unit) chip. The central gateway unit 114 can be selected according to actual needs; for example, it can be a CGW (Central Gateway) chip. The communication unit 115 can be selected according to actual needs; for example, it can be a wireless gateway T-box.

[0087] The cloud 120 may include: a cloud storage unit 121, a cloud server 122, and a cloud computing unit 123. The cloud storage unit 121 is communicatively connected to the cloud server 122 and the cloud computing unit 123, so that the cloud server 122 can obtain data from the cloud storage unit 121 and send it to the cloud computing unit 123 for corresponding calculations.

[0088] The cloud storage unit 121 contains multiple cloud server clusters, which can collect and organize the raw data packets of each vehicle uploaded by the communication unit 115, and decompress and structure the data uploaded daily before storing it in the data warehouse. The cloud server 122 can be understood as a cloud server cluster.

[0089] The cloud computing unit 123 is used to pull the structured vehicle feature data of the cloud storage unit 121 in the cloud 120 and perform calculations on each vehicle feature data, such as the full-cycle data processing and archiving from data cleaning preprocessing to curve fitting and theoretical value calculation, to risk factor generation.

[0090] It should be noted that, Figure 2 The description of data communication between vehicle terminal 110 and the cloud is based solely on a single vehicle terminal 110 and should not be construed as a limitation of this application.

[0091] Alternatively, in one possible implementation, Figure 3 An example illustration of a cooling anomaly identification system for an automotive power module provided in this application embodiment. Figure 3 .like Figure 3 As shown, the cloud computing unit 123 may include: a training component 123-1 and a driving component 123-2.

[0092] The training component 123-1 may include: all training set data segments divided into vehicle characteristic operating condition intervals for fitting curve parameters, and a table of fitted curve characteristic parameters generated after training. The driving component 123-2 may include: vehicle driving data segmented into vehicle characteristic operating condition intervals, after fitting and risk calculation; and, after each day's cooling anomaly detection, risk-free data segments will be cleared, while data segments with risk factors will be archived daily for future review. The driving component 123-2 may also include: a risk accumulation table and a detailed table of total risk factors accumulated daily.

[0093] This application proposes a cooling anomaly identification system for automotive power modules. The system utilizes communication modules in multiple vehicles, such as wireless gateways, to upload data to the cloud for storage and processing. By processing the data uploaded to the cloud, the system calculates cooling anomaly risk parameters for each vehicle, effectively identifying cooling anomalies in the automotive power modules and preventing damage to the vehicle's motor and inverter.

[0094] The cooling anomaly identification method for automotive power modules provided in the above embodiments of this application will be explained and described in detail below with reference to the accompanying drawings. Figure 4 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 1 This method can be derived from, for example... Figure 1 The cooling anomaly identification system 100 for the automotive power module shown is implemented by the cloud server 122 in the cloud 120. For example... Figure 4 As shown, the method for identifying cooling anomalies in automotive power modules includes the following steps:

[0095] S201. Obtain the data points of multiple vehicles.

[0096] In one possible implementation, edge-mounted data from multiple vehicles is transmitted back to cloud storage unit 121 in the cloud 120 via a communication module (such as a wireless gateway T-box). Cloud storage unit 121 then groups the edge-mounted data from multiple vehicles according to each vehicle's unique identifier (e.g., vehicle VIN). Data extraction is then performed on the grouped edge-mounted data for each vehicle to generate its own edge-mounted data. This data may include each vehicle's electric drive signals and driving status signals. The unique identifier for each vehicle can be selected based on specific circumstances; for example, it can be understood as a unique vehicle identification number.

[0097] For example, it can be based on the above Figure 1The electric drive unit 111 in the vehicle obtains data from the motor module 111-1, cooling module 111-2, vehicle power module 111-3, and electric drive control module 111-4. For example, the actual operating speed and torque of the motor can be obtained from the motor module 111-1 of each vehicle; the cooling water flow rate and cooling water temperature data can be obtained from the cooling module 111-2 of each vehicle; the bridge arm temperature of the vehicle power module can be obtained from the vehicle power module 111-3 of each vehicle; and the cooling command can be issued from the electric drive control module 111-4 of each vehicle when the vehicle needs to be cooled.

[0098] For example, it can be based on the above Figure 1 The driving status unit 113 in the middle can be used to obtain, such as driving start / stop identifier, driving mileage data, vehicle speed data and driving timestamp, based on the driving status unit 113 of each vehicle.

[0099] Additionally, it should be noted that the edge-mounted data of the vehicle in the above embodiments refers to all data collected by the vehicle uploaded by each vehicle, and should not be construed as a limitation on this application.

[0100] S202. Based on the electric drive signal and driving status signal of each vehicle, determine the actual temperature parameters of each vehicle and the data characteristic time series of each vehicle.

[0101] In one possible implementation, a data feature time series for each vehicle is determined based on its electric drive signal and driving status signal, facilitating subsequent processing based on the acquired data feature time series. The actual temperature parameter for each vehicle is then determined based on this data feature time series. This actual temperature parameter characterizes the actual temperature of the power module in each vehicle's electric drive system.

[0102] It should be noted that the actual temperature parameter can be represented, for example, as a quartile numerical record.

[0103] S203. Obtain the electric drive fitting curve characteristic table for each vehicle.

[0104] In one possible implementation, the electric drive fitting curve characteristic table for each vehicle in cloud storage unit 121 is extracted based on the unique identifier of each vehicle (e.g., vehicle VIN). This electric drive fitting curve characteristic table can be used to characterize the relationship between the electric drive operating parameters and the reference temperature parameter for each vehicle. This electric drive fitting curve characteristic table can generally be represented by a linear equation in one variable, as shown in the following formula (1):

[0105] T IGBT-Ref =λ*k*T tor +θ*b n +Tbase Formula (1)

[0106] Among them, T IGBT-Ref For each vehicle, the reference temperature parameter is T. tor Electric drive operating parameters for each vehicle; k n and b n To correspond to the linear parameters under different torque ranges, λ and θ are cooling water flow correction coefficients, and the above k n and b n Both λ and θ are related to the motor characteristics; T base This is the theoretical temperature error correction value.

[0107] As can be seen from the above formula (1), based on the electric drive operating parameters of each vehicle, there is a corresponding reference temperature parameter T for the vehicle power module. IGBT Furthermore, the reference temperature parameter for each vehicle varies with the electric drive operating parameters of each vehicle. Additionally, linear fitting calculations can be performed on the theoretical reference temperature of the vehicle power module via motor torque under different operating conditions for each vehicle using cloud data. This allows for the determination of the difference and degree of difference between the theoretical and actual temperatures of the vehicle power module under the corresponding load conditions.

[0108] S204. Based on the preset electric drive operating parameters of each vehicle, the temperature parameters are fitted using the electric drive fitting curve characteristic table of each vehicle to obtain the reference temperature parameters of each vehicle.

[0109] The preset electric drive operating parameters for each vehicle can be selected according to actual conditions. For example, the preset electric drive operating parameters for each vehicle can be selected as the actual operating speed and torque of the motor, coolant flow rate, and coolant temperature data.

[0110] In one possible implementation, temperature parameters are fitted based on the preset electric drive operating condition parameters of each vehicle and the electric drive fitting curve characteristic table of each vehicle. For example, the preset electric drive operating condition parameters T of each vehicle are... tor Substituting into the above formula (1), we obtain a parameter T that is consistent with the preset electric drive operating condition parameter. tor The corresponding reference temperature parameter T IGBT This allows us to obtain the reference temperature parameter T for each vehicle. IGBT .

[0111] It should be noted that the reference temperature parameter can be represented, for example, as a quartile numerical record.

[0112] S205. Based on the reference temperature parameters of each vehicle, the actual temperature parameters of each vehicle, and the data characteristic time series of each vehicle, the power module in each vehicle is identified for cooling anomalies, and the cooling anomaly risk parameters of each vehicle are obtained.

[0113] In one possible implementation, the actual temperature parameter of each vehicle and the reference temperature parameter T for a certain time period are obtained based on the time series data characteristics of each vehicle. IGBT And based on the actual temperature parameters of each vehicle and the reference temperature parameter T of each vehicle. IGBT The temperature difference between the two values ​​is used to determine whether the power module in each vehicle has reached a cooling anomaly point. This anomaly point is then used to identify cooling anomalies, thus obtaining the cooling anomaly risk parameter R for each vehicle at a given time period. P Among them, the cooling anomaly risk parameter R for each vehicle P R can be used to characterize the risk of abnormal cooling of the power module in each vehicle. P .

[0114] It should be noted that the cooling anomaly risk parameter R for each vehicle during a certain time period was obtained. P Subsequently, based on the cooling anomaly risk parameter R corresponding to the data feature time series of each vehicle mentioned above... P This involves generating a detailed data feature table for each vehicle based on its time series data characteristics, and then using the cooling anomaly risk parameter R for each vehicle at different times. P The parameters are accumulated to obtain the cooling anomaly risk parameter R for each vehicle. P .

[0115] Furthermore, for all previously grouped vehicles with unique identifiers, the risk R of abnormal cooling for multiple vehicles is... P The data features and time series are aggregated and calculated to generate a cooling anomaly warning table for vehicle power modules. Then, based on the cooling anomaly warning table for vehicle power modules, warnings are output to multiple vehicles, and the vehicles that need to be warned are those whose vehicle power modules have cooling anomalies.

[0116] The embodiments of this application provide a method for identifying cooling anomalies in automotive power modules. This method can acquire embedded data from multiple vehicles and determine the actual temperature parameters and data feature time series of each vehicle based on the embedded data. This ensures that the data acquired by the cloud server in the cloud is data from a large number of vehicles, thereby avoiding the influence of abnormal data values ​​in the vehicle's embedded data and improving the stability and accuracy of the vehicles. The system acquires the electric drive fitting curve characteristic table for each vehicle. Based on this table, it fits the temperature parameters to the preset electric drive operating conditions for each vehicle to obtain reference temperature parameters. Data calculations and simulations are then performed based on different electric drive operating conditions for each vehicle, resulting in reference temperature parameters with good adaptability and high reliability. Simultaneously, based on the reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, cooling anomalies in the power modules of each vehicle are identified, yielding cooling anomaly risk parameters. This allows for the accumulation of cooling anomaly risk parameters over time, effectively enabling early warning based on the reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, resulting in high value density and utilization.

[0117] exist Figure 4 Based on the above, the cooling anomaly identification method for automotive power modules provided in this application will be explained and described in detail below with reference to the accompanying drawings. Figure 5 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 2 .like Figure 5 As shown, the method described above, which determines the actual temperature parameters and data characteristic time series of each vehicle based on the electric drive signal and driving state signal of each vehicle, may include:

[0118] S301. Extract the electric drive characteristic parameters of each vehicle at multiple time points from the electric drive signal of each vehicle.

[0119] In one possible implementation, the cloud server 122 extracts electric drive characteristic parameters of each vehicle at multiple time points from the electric drive signals of each vehicle in the cloud storage unit 121. These electric drive characteristic parameters may include: actual electric drive operating condition parameters and actual temperature parameters.

[0120] Among them, the actual electric drive operating condition parameters are the actual electric drive operating condition data uploaded to the cloud 120 by the motor module 111-1 and cooling module 111-2 of each vehicle at the current moment, such as the actual operating speed and torque of the motor or the cooling water flow and cooling water temperature data.

[0121] The actual temperature parameter is the actual temperature data of the vehicle's power module at time 111-3, which is uploaded to the cloud at time 120.

[0122] S302. Extract driving characteristic parameters of each vehicle at multiple time points from the driving status signal of each vehicle.

[0123] In one possible implementation, the cloud server 122 extracts driving characteristic parameters of each vehicle at multiple time points from the driving signals of each vehicle in the cloud storage unit 121. These driving characteristic parameters can be selected according to actual conditions; for example, they can be driving start / stop indicators, driving mileage data, vehicle speed data, driving timestamps, etc.

[0124] S303. Generate a time series of data features for each vehicle based on electric drive characteristic parameters and driving characteristic parameters at multiple time points.

[0125] In one possible implementation, since both the electric drive characteristic parameters and driving characteristic parameters of the same vehicle have corresponding timestamps, and these timestamps are unique identifiers in the embedded data of a vehicle, the electric drive characteristic parameters and driving characteristic parameters of the same vehicle can be sorted according to the unique identifier timestamp of each vehicle's embedded data, so that the timestamps of the electric drive characteristic parameters and driving characteristic parameters of the same vehicle correspond. For example, if the timestamp of the electric drive characteristic parameter of the same vehicle is 9:00, then the timestamp of the driving characteristic parameter of the same vehicle is also 9:00. In this way, data with the same timestamp of the same vehicle can be combined together and arranged according to the order of the timestamps, thereby generating a data characteristic time series for each vehicle.

[0126] It should be noted that after obtaining the data feature time series of each vehicle, preprocessing is required. Since the raw edge-mounted data provided by the vehicle may contain abnormal sensor sampling points or noise points caused by electromagnetic interference, extreme points or noise points exceeding the signal physical value setting range in the data feature time series of each vehicle obtained in this application need to be cleared. The processing of extreme points or noise points within the signal physical value setting range can be adjusted according to the actual situation. Furthermore, linear interpolation is performed before and after the cleared extreme points or noise points in the data feature time series of each vehicle to remove outliers and filter signal noise.

[0127] After processing noise points, time series prediction based on a multivariate input preset algorithm is used to fill in potential continuous signal gaps in the signal transmission chain, thus completing and filling data features. This avoids the impact of outliers on cloud monitoring data processing, improving the stability and accuracy of early warnings while ensuring the reliability of large-scale early warnings, with wide coverage and high recall. Specifically, for single continuous outliers with multiple points in the data feature time series of each vehicle, or when the proportion of empty segments in the data feature time series of each vehicle is below a preset ratio, time series prediction based on a multivariate input preset algorithm is used to fill in continuous missing time series signal segments in order to restore the true data situation as much as possible. However, segments with a missing span greater than the preset ratio or that cannot be predicted by the corresponding multivariate input preset algorithm are directly discarded. The preset ratio and multivariate input preset algorithm can be selected according to the actual situation. For example, the preset ratio can be selected as 10%, and the multivariate input preset algorithm can be selected as LSTM (Long Short Term Memory Recurrent Neural Network).

[0128] For example, Figure 6 This diagram illustrates the preprocessing of a cooling anomaly identification method provided in this application. Taking an IGBT module as the automotive power module and an LSTM multivariate input preset algorithm as an example, the training set input to the LSTM consists of multiple normal historical driving data segments divided according to driving cycle conditions. The feature fields input to the LSTM network include the electric drive signal and the driving state signal of each vehicle, and the output is the predicted IGBT temperature feature value. Figure 6 As shown, the raw temperature data of the IGBT module has a data interruption problem. However, after preprocessing in the cooling anomaly identification method of the automotive power module of this application, the missing data can be clearly seen and the data is supplemented, thus ensuring the integrity of the data feature time series of each vehicle used in the processing of this application.

[0129] The embodiments of this application provide a method for identifying cooling anomalies in automotive power modules. This method determines the actual temperature parameters and data feature time series of each vehicle based on its electric drive signal and driving status signal. The method includes: extracting electric drive feature parameters for each vehicle at multiple time points from its electric drive signal; these feature parameters include actual electric drive operating condition parameters and actual temperature parameters; extracting driving feature parameters for each vehicle at multiple time points from its driving status signal; and generating a data feature time series for each vehicle based on these electric drive and driving feature parameters. Therefore, this application can combine the electric drive and driving feature parameters from the embedded data of each vehicle according to the same timestamp for the same vehicle, and then arrange the data according to the chronological order of the timestamps to generate a data feature time series for each vehicle, thus ensuring the logical consistency of the embedded data for each vehicle in subsequent processing.

[0130] In the above Figure 5 Based on the corresponding embodiment, in another embodiment, the driving characteristic parameters may include at least a driving start-stop indicator. This driving start-stop indicator is used to identify when each vehicle temporarily stops during driving, such as automatically shutting off while waiting at a red light, and automatically restarting when needed. The working principle of this automatic start-stop is as follows: when each vehicle is driving in a congested area, the driver presses the brake pedal, the electric drive unit 111 automatically detects that the engine is idling and not in gear, the wheels in the anti-lock electric drive unit 111 are locked, the speed sensor displays zero, and the battery unit indicates sufficient energy for the next start.

[0131] Based on the above embodiments, the cooling anomaly identification method for automotive power modules provided in this application may further include:

[0132] Based on the driving start-stop identifier in the driving characteristic parameters, the time series of data characteristics for each vehicle are spliced ​​together to obtain a data segment for each vehicle under at least one driving cycle condition.

[0133] In one possible implementation, based on the driving start-stop identifier in the driving feature parameters, the data feature time series of each vehicle is spliced ​​together according to the timestamp corresponding to the driving start-stop identifier, so as to avoid obtaining data feature time series of each vehicle that do not meet the preset training segment, thereby obtaining data segments of each vehicle under at least one driving cycle condition.

[0134] The preset training segment can be understood as the length of the preset data feature time series for each vehicle. The length of the preset data feature time series can be selected according to the actual situation. For example, the length of the preset data feature time series can be set to 60 minutes.

[0135] Optionally, in the above method, based on the reference temperature parameters of each vehicle, the actual temperature parameters of each vehicle, and the data feature time series of each vehicle, the cooling anomaly identification of the power module in each vehicle is performed to obtain the cooling anomaly risk parameters of each vehicle, which may include:

[0136] Based on the reference temperature parameters of each vehicle, the actual temperature parameters of each vehicle, and the data segments of each vehicle under each driving cycle condition, the cooling anomaly of the power module in each vehicle under each driving cycle condition is identified, and the cooling anomaly risk parameters of each vehicle under each driving cycle condition are obtained.

[0137] In one possible implementation, reference temperature parameters and actual temperature parameters for each vehicle are obtained from data segments under each driving cycle condition. Then, based on the temperature difference between the reference and actual temperature parameters, it is determined whether the power module in each vehicle has reached a cooling anomaly point. This anomaly point is then used for cooling anomaly identification, thereby obtaining a cooling anomaly risk parameter R for each vehicle over a certain time period. P This allows us to obtain the cooling anomaly risk parameter R for each vehicle under each driving cycle condition. P .

[0138] The embodiments of this application provide a method for identifying cooling anomalies in automotive power modules. This method involves splicing together the time series data features of each vehicle based on the driving start-stop identifier in the driving characteristic parameters to obtain data segments for each vehicle under at least one driving cycle condition. Based on the reference temperature parameters, actual temperature parameters, and data segments for each vehicle under each driving cycle condition, cooling anomalies in the power modules of each vehicle are identified under each driving cycle condition, resulting in cooling anomaly risk parameters for each vehicle under each driving cycle condition. Therefore, this application improves the accuracy of the cooling anomaly identification method for automotive power modules by identifying cooling anomalies in the power modules of each vehicle under each driving cycle condition based on the reference temperature parameters, actual temperature parameters, and data segments for each vehicle, thereby reducing the uncertainty and randomness of the data uploaded by each vehicle and optimizing the cooling anomaly identification method.

[0139] In the above Figure 5 Based on the corresponding embodiment, in another embodiment, the preset electric drive condition parameters for each vehicle can be: the preset electric drive condition parameter range for each vehicle. Figure 7 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 3 .like Figure 7 As shown, the method described above, which uses the electric drive fitting curve characteristic table of each vehicle to fit the temperature parameters based on the preset electric drive operating condition parameters of each vehicle, to obtain the reference temperature parameters for each vehicle, may include:

[0140] S401. Divide the preset electric drive operating condition parameter range for each vehicle into segments to obtain multiple electric drive operating condition sub-ranges for each vehicle.

[0141] In one possible implementation, the preset electric drive operating condition parameter range for each vehicle can be segmented based on preset electric drive operating condition parameters such as motor characteristics and the actual driving motor torque-speed of all users, resulting in multiple electric drive operating condition sub-ranges for each vehicle. To consider the actual preset electric drive operating condition parameter range for each vehicle and the balance between resource consumption and accuracy, the multiple electric drive operating condition sub-ranges for each vehicle can be divided into three torque range segments: high, medium, and low. However, this should not be construed as a limitation of this application.

[0142] S402. Based on multiple electric drive operating condition sub-ranges, the temperature parameters are fitted in multiple electric drive operating condition sub-ranges using the electric drive fitting curve characteristic table for each vehicle, so as to obtain the reference temperature parameters for each vehicle in multiple electric drive operating condition sub-ranges.

[0143] In one possible implementation, the temperature parameters of each vehicle under multiple electric drive operating condition sub-intervals are fitted based on formula (1) according to multiple electric drive operating condition sub-intervals and the electric drive fitting curve characteristic table of each vehicle, so as to obtain the reference temperature parameters of each vehicle under multiple electric drive operating condition sub-intervals.

[0144] The embodiment of this application provides a cooling anomaly identification method for automotive power modules. The method includes: a preset electric drive operating condition parameter range for each vehicle; and fitting temperature parameters using an electric drive fitting curve characteristic table for each vehicle based on the preset electric drive operating condition parameters to obtain reference temperature parameters for each vehicle. This includes: segmenting the preset electric drive operating condition parameter range for each vehicle to obtain multiple electric drive operating condition sub-ranges for each vehicle; and fitting temperature parameters using the electric drive fitting curve characteristic table for each vehicle under each of the multiple electric drive operating condition sub-ranges to obtain reference temperature parameters for each vehicle in the multiple electric drive operating condition sub-ranges. Therefore, this application can fit temperature parameters based on multiple electric drive operating condition sub-ranges to obtain reference temperature parameters for each vehicle in multiple electric drive operating condition sub-ranges. Since the reference temperature parameters for each vehicle in multiple electric drive operating condition sub-ranges are fitted based on multiple electric drive operating condition sub-ranges, the accuracy of the cooling anomaly identification method for automotive power modules is improved, thereby reducing the uncertainty and randomness of the data uploaded by each vehicle and optimizing the cooling anomaly identification method.

[0145] Optionally, Figure 8 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 4 .like Figure 8 As shown, the above method identifies cooling anomalies in the power modules of each vehicle based on the reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, obtaining cooling anomaly risk parameters for each vehicle. These parameters may include:

[0146] S501. Based on the actual electric drive operating condition parameters of each vehicle, determine the target reference temperature parameter of the target operating condition parameter sub-interval where the actual electric drive operating condition parameters are located from the reference temperature parameters under each electric drive operating condition sub-interval.

[0147] In one possible implementation, based on the actual electric drive operating parameters of each vehicle, a target reference temperature parameter is determined from the reference temperature parameters under each electric drive operating condition sub-interval, which is the target operating condition parameter sub-interval where the actual electric drive operating parameters are located. That is, for the same operating condition interval of the actual electric drive operating parameters of each vehicle, a target operating condition parameter sub-interval is further divided, and the target operating condition parameter sub-interval is an intersecting sub-interval. Then, a target reference temperature parameter is randomly selected in the target operating condition parameter sub-interval.

[0148] S502. Based on the target reference temperature parameters, the actual temperature parameters of each vehicle, and the data characteristic time series of each vehicle, the power module in each vehicle is identified for cooling anomalies, and the cooling anomaly risk parameters of each vehicle are obtained.

[0149] In one possible implementation, data fitting is performed between the target reference temperature parameter and the data feature time series of each vehicle to obtain fitted data for the target reference temperature parameter of each vehicle. Similarly, data fitting is performed between the actual temperature parameter of each vehicle and the data feature time series of each vehicle to obtain fitted data for the actual temperature parameter of each vehicle. Based on both the fitted data for the target reference temperature parameter and the fitted data for the actual temperature parameter of each vehicle, cooling anomalies in the power modules of each vehicle are identified, resulting in a cooling anomaly risk parameter R for each vehicle. P .

[0150] The embodiments of this application provide a cooling anomaly identification method for automotive power modules. This method identifies cooling anomalies in the power modules of each vehicle based on reference temperature parameters, actual temperature parameters, and data feature time series for each vehicle, obtaining cooling anomaly risk parameters for each vehicle. The method includes: determining the target reference temperature parameter of the target operating condition parameter sub-interval where the actual electric drive operating condition parameters reside from the reference temperature parameters under each electric drive operating condition sub-interval; and identifying cooling anomalies in the power modules of each vehicle based on the target reference temperature parameter, the actual temperature parameters, and the data feature time series for each vehicle, obtaining cooling anomaly risk parameters for each vehicle. Therefore, this application can identify cooling anomalies in the power modules of each vehicle based on the target reference temperature parameter of the target operating condition parameter sub-interval for each vehicle under the same operating condition, as well as the actual temperature parameters and data feature time series for each vehicle, obtaining cooling anomaly risk parameters for each vehicle. This improves the accuracy of the cooling anomaly identification method for automotive power modules, thereby reducing the uncertainty and randomness of the data uploaded by each vehicle and optimizing the cooling anomaly identification method.

[0151] exist Figure 4 Based on the above, the cooling anomaly identification method for automotive power modules provided in this application will be explained and described in detail below with reference to the accompanying drawings. Figure 9 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 5 .like Figure 9 As shown, the above method identifies cooling anomalies in the power modules of each vehicle based on the reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, obtaining cooling anomaly risk parameters for each vehicle. These parameters may include:

[0152] S601. Based on the reference temperature parameters and actual temperature parameters of each vehicle, identify the abnormal data points of each vehicle from the data feature time series of each vehicle.

[0153] In one possible implementation, it is determined whether the temperature difference between the reference temperature parameter and the actual temperature parameter of each vehicle is greater than a set temperature threshold. If the temperature difference between the reference temperature parameter and the actual temperature parameter of each vehicle is greater than the set temperature threshold, then it is determined that there are abnormal data points in the data feature time series of each vehicle, and the abnormal data points are tagged and recorded to facilitate the timely processing of abnormal data points in the data feature time series of each vehicle.

[0154] S602. Determine the proportion of outlier data in the time series of data characteristics for each vehicle.

[0155] In one possible implementation, after identifying outlier data points in the data feature time series of each vehicle, the proportion of outlier data in the data feature time series of each vehicle is obtained based on the proportion of outlier data points in the data feature time series of each vehicle to the proportion of outlier data points in the data feature time series of each vehicle.

[0156] S603. Based on the proportion of abnormal data, determine the cooling abnormality risk parameters for each vehicle.

[0157] In one possible implementation, a preset value, such as 20%, is set for the percentage of abnormal data for each vehicle, and this preset value is assigned to the data feature time series of each vehicle. The assigned data feature time series is then used as the cooling anomaly risk parameter R for each vehicle. P .

[0158] The embodiments of this application provide a cooling anomaly identification method for automotive power modules. This method identifies cooling anomalies in the power modules of each vehicle based on the vehicle's reference temperature parameters, actual temperature parameters, and data feature time series, obtaining a cooling anomaly risk parameter for each vehicle. The method includes: determining abnormal data points for each vehicle from its data feature time series based on the reference and actual temperature parameters; determining the proportion of abnormal data points in the data feature time series of each vehicle; and determining the cooling anomaly risk parameter for each vehicle based on the proportion of abnormal data. Therefore, this application can assign the proportion of the cumulative length of abnormal data points in the data feature time series to the data feature time series as a risk factor coefficient, thereby ensuring the early warning accuracy of the cooling anomaly identification method.

[0159] exist Figure 9 Based on the above, the cooling anomaly identification method for automotive power modules provided in this application will be explained and described in detail below with reference to the accompanying drawings. Figure 10 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 6 .like Figure 10 As shown, the reference temperature parameter for each vehicle in the above method may include: each reference temperature parameter in the reference temperature range for each vehicle.

[0160] Optionally, the method described above, which determines outlier data points for each vehicle from its data feature time series based on the reference temperature parameters and actual temperature parameters of each vehicle, may include:

[0161] S701. Determine the actual temperature parameters of each vehicle and the temperature parameter differences of each reference temperature parameter within the reference temperature range for each vehicle, and obtain multiple temperature deviation values ​​for each vehicle.

[0162] The reference temperature range for each vehicle can be based on the reference temperature parameters of each vehicle, as well as the maximum and minimum deviation limits t of the cooling structure of the vehicle power module under the current season and environment. Ref-up and t Ref-down The reference temperature range [T] for each vehicle is obtained. IGBT参 -t Ref-up ,T IGBT参 +t Ref-down ].

[0163] In one possible implementation, the actual temperature parameter of each vehicle and the temperature parameter difference of each reference temperature parameter within the reference temperature range of each vehicle can be determined by the following formula (2), thereby obtaining multiple temperature deviation values ​​for each vehicle.

[0164] ΔT=T IGBT实 -T IGBT参 Formula (2)

[0165] Among them, T IGBT参 Reference temperature parameters within the reference temperature range for each vehicle.

[0166] It should be noted that the reference temperature parameter T for each vehicle... IGBT参 It is within the reference temperature range for each vehicle.

[0167] S702. Determine the target temperature deviation value with the largest absolute value among multiple temperature deviation values ​​for each vehicle.

[0168] In one possible implementation, after determining the actual temperature parameter of each vehicle and the temperature parameter difference of each reference temperature parameter within the reference temperature range of each vehicle according to the above formula (2), multiple temperature deviation values ​​can be obtained, and the target temperature deviation value with the largest absolute value among the multiple temperature deviation values ​​of each vehicle can be calculated by referring to the following formula (3).

[0169] MAX(ΔT)=|T IGBT实 -T IGBT参 | Formula (3)

[0170] S703. If the target temperature deviation value is greater than the preset temperature deviation threshold, mark the data features of the target time point corresponding to the actual temperature parameter of each vehicle in the data feature time series of each vehicle as abnormal data points.

[0171] The preset temperature deviation threshold IR can be selected according to the actual situation. For example, the preset temperature deviation threshold IR can be selected as 10.

[0172] In one possible implementation, if the target temperature deviation value MAX(ΔT) is greater than the preset temperature deviation threshold IR, then the data feature of the target time point corresponding to the actual temperature parameter of each vehicle in the data feature time series of each vehicle is marked as an abnormal data point.

[0173] For example, Figure 11 This is a schematic diagram of the theoretical temperature curve for a cooling anomaly identification method provided in an embodiment of this application. Figure 11 As shown, the theoretical temperature curve in the middle represents the theoretical temperature limit threshold (e.g., T). IGBT参 The theoretical temperature curve above represents the upper limit threshold of the theoretical temperature (e.g., T). IGBT参 -t Ref-up The theoretical temperature curve below represents the lower limit threshold of the theoretical temperature (e.g., T). IGBT参 +t Ref-down All the small dots represent the actual temperature parameters (such as T) for each vehicle in this application. IGBT实 As can be clearly seen, the small circles that exceed the upper and lower theoretical temperature thresholds are abnormal data points.

[0174] It should be noted that the theoretical upper temperature threshold and the theoretical lower temperature threshold can also be determined by the processing logic of the cloud server's method for identifying cooling anomalies in automotive power modules and the frequency of vehicle-side data being uploaded to the cloud.

[0175] The embodiments of this application provide a cooling anomaly identification method for automotive power modules. The method includes a reference temperature parameter for each vehicle, comprising: reference temperature parameters within a reference temperature range for each vehicle; determining abnormal data points for each vehicle from its data feature time series based on the reference temperature parameters and the actual temperature parameters of each vehicle, including: determining the temperature parameter difference between the actual temperature parameter and each reference temperature parameter within the reference temperature range for each vehicle, obtaining multiple temperature deviation values ​​for each vehicle; determining the target temperature deviation value with the largest absolute value among the multiple temperature deviation values ​​for each vehicle; and marking the data feature of the target time point corresponding to the actual temperature parameter of each vehicle in its data feature time series as an abnormal data point if the target temperature deviation value is greater than a preset temperature deviation threshold. Therefore, this application can determine that the target temperature deviation value obtained by the difference between the actual temperature parameter of each vehicle and the temperature parameter of each reference temperature parameter within the reference temperature range of each vehicle is greater than the preset temperature deviation threshold. This allows the data feature of the target time point corresponding to the actual temperature parameter of each vehicle in the data feature time series of each vehicle to be marked as an abnormal data point. Then, the reference temperature parameters within the reference temperature range of each vehicle are compared with the actual temperature parameters of each vehicle. This provides a reference baseline for the cooling anomaly identification method of the vehicle power module provided by this application and provides a standard for judging abnormal data points.

[0176] exist Figure 10 Based on the above, the cooling anomaly identification method for automotive power modules provided in this application will be explained and described in detail below with reference to the accompanying drawings. Figure 12 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 7 .like Figure 12 As shown, the method described above, which determines the cooling anomaly risk parameters for each vehicle based on the proportion of abnormal data, may include:

[0177] S801. Based on the proportion of abnormal data, determine the current risk factor coefficient of the data feature time series of each vehicle.

[0178] In one possible implementation, the proportion of outlier data points for each vehicle in the data feature time series of each vehicle is used as the current risk factor coefficient R for each vehicle's data feature time series. P For example, the current risk factor coefficient R of the data feature time series of each vehicle can be determined according to the following formula (4). P .

[0179]

[0180] Where n is the total number of outlier data points in the time series data features of each vehicle, and L is the number of outlier data points. n The length of outlier data points in the time series data features for each vehicle.

[0181] S802. Based on the current risk factor coefficients and the data characteristic time series of each vehicle, generate a current segment detail table for each vehicle.

[0182] In one possible implementation, based on the current risk factor coefficient R P Together with the data characteristics of each vehicle in the time series, such as each vehicle's unique identifier, timestamp, driving mileage, actual temperature parameters, and reference temperature parameters, a current segment detail table for each vehicle is generated.

[0183] S803. Based on the current segment details table for each vehicle and the historical risk factor coefficients for each vehicle, determine the cooling anomaly risk parameters for each vehicle.

[0184] In one possible implementation, cooling anomaly risk parameters for each vehicle are formed by summing the current segment details for each vehicle and the historical risk factor coefficients for each vehicle.

[0185] The embodiments of this application provide a cooling anomaly identification method for automotive power modules. This method determines the cooling anomaly risk parameters for each vehicle based on the proportion of abnormal data. The method includes: determining the current risk factor coefficient of the data feature time series for each vehicle based on the proportion of abnormal data; generating a current segment detail table for each vehicle based on the current risk factor coefficient and the data feature time series; and determining the cooling anomaly risk parameters for each vehicle based on the current segment detail table and the historical risk factor coefficients. Therefore, this application can determine the current risk factor coefficient of the data feature time series for each vehicle based on the proportion of abnormal data. Abnormal data points in the data feature time series of each vehicle form a current segment detail table, and the cumulative risk factor coefficients over time are statistically analyzed. This effectively combines and utilizes historical risk factor coefficient data, and the process and data are traceable, thereby ensuring the accuracy of cooling anomaly warnings for automotive power modules and providing advance warnings for abnormal vehicles.

[0186] exist Figure 4 Based on the above, the cooling anomaly identification method for automotive power modules provided in this application will be explained and described in detail below with reference to the accompanying drawings. Figure 13 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 8 .like Figure 13 As shown, the method described above for obtaining the electric drive fitting curve characteristic table for each vehicle may include:

[0187] S901. If the electric drive fitting curve characteristic table of the first vehicle is stored in advance, then obtain the electric drive fitting curve characteristic table of the first vehicle.

[0188] In one possible implementation, if the electric drive fitting curve characteristic table of the first vehicle is stored in advance, the electric drive fitting curve characteristic table of the first vehicle is extracted by the cloud server 122 according to the electric drive fitting curve characteristic table of the first vehicle stored in the cloud storage unit 121, and then the electric drive fitting curve characteristic table of the first vehicle is obtained.

[0189] S902. If the electric drive characteristic table of the second vehicle is not stored in advance, then based on the preset electric drive operating condition parameters of the second vehicle, the data characteristic time series of the second vehicle, and multiple historical data characteristic sequences of the second vehicle, the relationship curve between the electric drive operating condition parameters and temperature parameters of the second vehicle is fitted to obtain the electric drive fitting curve characteristic table of the second vehicle.

[0190] In one possible implementation, if the electric drive characteristic table of the second vehicle is not stored in advance, the second vehicle is trained according to the data characteristic time series of the second vehicle and multiple historical data characteristic sequences of the second vehicle, according to the preset electric drive operating condition parameters of the second vehicle, and the relationship curve between the electric drive operating condition parameters and temperature parameters of the second vehicle is fitted based on formula (1), thereby obtaining the electric drive fitting curve characteristic table of the second vehicle.

[0191] The embodiment of this application provides a cooling anomaly identification method for automotive power modules. The method for obtaining the electric drive fitting curve characteristic table for each vehicle includes: if the electric drive fitting curve characteristic table for a first vehicle is pre-stored, then obtaining the electric drive fitting curve characteristic table for the first vehicle; if the electric drive characteristic table for a second vehicle is not pre-stored, then fitting the relationship curve between the electric drive operating parameters and temperature parameters of the second vehicle based on the preset electric drive operating condition parameters of the second vehicle, the data characteristic time series of the second vehicle, and multiple historical data characteristic sequences of the second vehicle, to obtain the electric drive fitting curve characteristic table for the second vehicle. Therefore, this application analyzes the cases where the electric drive fitting curve characteristic table of the vehicle is stored in advance or not. In the case where the electric drive fitting curve characteristic table of the vehicle is not stored, the relationship curve between the electric drive operating parameters and temperature parameters of the second vehicle is fitted according to the preset electric drive operating parameters of the second vehicle, the data feature time series of the second vehicle, and multiple historical data feature sequences of the second vehicle to obtain the electric drive fitting curve characteristic table of the second vehicle. Based on the electric drive fitting curve characteristic table, the theoretical reference temperature of the vehicle power module under different operating conditions is fitted to cloud data.

[0192] exist Figure 13Based on the above, the cooling anomaly identification method for automotive power modules provided in this application will be explained and described in detail below with reference to the accompanying drawings. Figure 14 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 9 .like Figure 14 As shown, the preset electric drive operating condition parameters of the second vehicle in the above method can be: the preset electric drive operating condition parameter range of the second vehicle.

[0193] Optionally, the above method, based on the preset electric drive operating condition parameters of the second vehicle, the data feature time series of the second vehicle, and multiple historical data feature sequences of the second vehicle, performs relationship curve fitting on the electric drive operating condition parameters and temperature parameters of the second vehicle to obtain the electric drive fitting curve characteristic table of the second vehicle, which may include:

[0194] S1001. The preset electric drive operating condition parameter range of the second vehicle is segmented to obtain multiple second electric drive operating condition sub-ranges of the second vehicle.

[0195] In one possible implementation, the same operating condition parameter interval within the preset electric drive operating condition parameter interval of the second vehicle is segmented, thus dividing the same operating condition interval into n non-overlapping electric drive operating condition sub-intervals Ω. i ,Right now This leads to multiple sub-ranges of the second electric drive operating conditions for the second vehicle.

[0196] It should be noted that when training in the second electric drive condition sub-interval, the more segments there are in the interval, the higher the accuracy of the fitting calculation will be, but the corresponding algorithm complexity and calculation time will increase. Considering the actual vehicle electric drive condition parameter interval and the balance between resource consumption and accuracy, the interval can be divided into three torque interval segments: high, medium and low.

[0197] S1002. Based on multiple second electric drive condition sub-intervals, the data feature time series of the second vehicle and multiple historical data feature sequences of the second vehicle are grouped to obtain the training dataset of the second vehicle under multiple second electric drive condition sub-intervals.

[0198] In one possible implementation, based on multiple second electric drive condition sub-intervals... The data feature time series of the second vehicle and multiple historical data feature sequences of the second vehicle are grouped, such as from multiple second electric drive condition sub-intervals. Randomly selected electric drive operating condition sub-interval Ω i Furthermore, the same number of data entries are extracted at each electric drive condition sub-interval level to form a new dataset, thereby obtaining the training dataset of the second vehicle under multiple second electric drive condition sub-intervals.

[0199] S1003. Based on the training dataset under the second electric drive condition sub-interval, the relationship curves between the electric drive condition parameters and temperature parameters are fitted under multiple second electric drive condition sub-intervals to obtain the characteristic table of the electric drive fitting curves of the second vehicle under multiple second electric drive condition sub-intervals.

[0200] In one possible implementation, based on the training dataset under the second electric drive condition sub-interval, and based on the above formula (1), the relationship curves between the electric drive condition parameters and temperature parameters are fitted under multiple second electric drive condition sub-intervals, thereby obtaining multiple parameter fitting curves f(T). tor ,j), where j is the number of curves fitting the relationship between electric drive operating condition parameters and temperature parameters under multiple second electric drive operating condition sub-intervals. The multiple parameter fitting curves f(T) tor ,j), the electric drive fitting function of the second vehicle under multiple second electric drive operating condition sub-intervals can be obtained as follows: According to the electric drive fitting function, as follows The characteristic table of electric drive fitting curves for the second vehicle under multiple second electric drive operating condition sub-intervals is obtained.

[0201] For example, the same operating condition interval is divided into n non-overlapping electric drive operating condition sub-intervals Ω. i ,Right now Randomly select electric drive operating condition sub-interval Ω i And extract the same number of data entries at each electric drive operating condition sub-interval level to combine them into a new dataset. For each newly formed dataset, such as a total of p operating condition interval number datasets, where p is the minimum number of simulations that converges to the threshold range based on the amount of data. Each dataset is fitted and trained using the above formula (1) to form a parameter fitting curve f(T) tor If j), then the fitting function for the electric drive operating condition sub-interval is: Finally, the characteristic fitting curve parameter table for each working condition range of the vehicle is compiled.

[0202] The embodiment of this application provides a method for identifying cooling anomalies in automotive power modules. The method uses preset electric drive operating condition parameters for the second vehicle as follows: a preset electric drive operating condition parameter range for the second vehicle; based on the preset electric drive operating condition parameters, the data feature time series of the second vehicle, and multiple historical data feature sequences of the second vehicle, a curve fitting of the relationship between the electric drive operating condition parameters and temperature parameters is performed on the second vehicle to obtain a characteristic table of the electric drive fitting curve for the second vehicle. This includes: segmenting the preset electric drive operating condition parameter range of the second vehicle to obtain multiple second electric drive operating condition sub-ranges; grouping the data feature time series of the second vehicle and multiple historical data feature sequences of the second vehicle according to the multiple second electric drive operating condition sub-ranges to obtain a training dataset for the second vehicle under the multiple second electric drive operating condition sub-ranges; and performing curve fitting of the relationship between the electric drive operating condition parameters and temperature parameters under the multiple second electric drive operating condition sub-ranges according to the training dataset under the multiple second electric drive operating condition sub-ranges to obtain a characteristic table of the electric drive fitting curve for the second vehicle under the multiple second electric drive operating condition sub-ranges. Therefore, this application can perform curve fitting of the relationship between electric drive operating condition parameters and temperature parameters for the same operating condition interval within the preset electric drive operating condition parameter interval of the second vehicle, thereby obtaining a characteristic table of electric drive fitting curves for the second vehicle under multiple second electric drive operating condition sub-intervals. During the training process, curve fitting of the relationship between electric drive operating condition parameters and temperature parameters can be performed on the operating condition data group after the intervals are divided, so as to reduce the uncertainty of electric drive fitting parameters, avoid the randomness of feature data, optimize the hyperparameters of electric drive fitting parameters, and improve the accuracy of electric drive fitting parameters.

[0203] exist Figure 14 Based on the above, the cooling anomaly identification method for automotive power modules provided in this application will be explained and described in detail below with reference to the accompanying drawings. Figure 15 A flowchart illustrating a cooling anomaly identification method for an automotive power module provided in this application embodiment. Figure 10 .like Figure 15 As shown, the above method, based on the training dataset under the second electric drive condition sub-interval, performs curve fitting of the relationship between electric drive condition parameters and temperature parameters under multiple second electric drive condition sub-intervals to obtain a characteristic table of electric drive fitting curves for the second vehicle under multiple second electric drive condition sub-intervals, which may include:

[0204] S1101. Perform Monte Carlo simulation on the training dataset under each second electric drive operating condition sub-interval to obtain the parameter fitting curve and fitting function for each second electric drive operating condition sub-interval.

[0205] In one possible implementation, Monte Carlo simulations are performed on the training dataset of the same operating condition sub-interval under each second electric drive operating condition sub-interval to obtain the parameter fitting curve f(T) for each second electric drive operating condition sub-interval. tor,j) and fitting function

[0206] S1102. Based on the parameter fitting curve and fitting function of each second electric drive operating condition sub-interval, generate a characteristic table of electric drive fitting curves for the second vehicle in each second electric drive operating condition sub-interval.

[0207] In one possible implementation, a fitting function for each second electric drive sub-interval is generated based on the parameter fitting curve of each second electric drive sub-interval, and a characteristic table of the electric drive fitting curve of the second vehicle under each second electric drive sub-interval is generated based on the fitting function.

[0208] The embodiment of this application provides a cooling anomaly identification method for automotive power modules. This method involves fitting the relationship curves between electric drive parameters and temperature parameters in multiple second electric drive condition sub-intervals based on a training dataset under each sub-interval, thereby obtaining an electric drive fitting curve characteristic table for the second vehicle in each sub-interval. The method includes: performing Monte Carlo simulations on the training dataset for each sub-interval to obtain the parameter fitting curve and fitting function for each sub-interval; and generating an electric drive fitting curve characteristic table for the second vehicle in each sub-interval based on the parameter fitting curve and fitting function. Therefore, this application can generate an electric drive fitting curve characteristic table for the second vehicle in each sub-interval based on the parameter fitting curve and fitting function, thereby reducing the uncertainty of electric drive fitting parameters, avoiding the randomness of feature data, optimizing the hyperparameters of electric drive fitting parameters, and improving the accuracy of electric drive fitting parameters.

[0209] Based on the same inventive concept, this application also provides a cooling anomaly identification device for automotive power modules. Since the principle of the device in this application is similar to the cooling anomaly identification method for automotive power modules described above in this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0210] Figure 16 This is a schematic diagram of a cooling anomaly detection device for an automotive power module provided in an embodiment of this application. Figure 16 As shown, the cooling anomaly handling device 1200 may include:

[0211] The first acquisition module 1201 is used to acquire embedded data of multiple vehicles. The embedded data of each vehicle includes: the electric drive signal of each vehicle and the driving status signal of each vehicle.

[0212] The determination module 1202 is used to determine the actual temperature parameters of each vehicle and the data characteristic time series of each vehicle based on the electric drive signal and the driving status signal of each vehicle. The actual temperature parameters are used to characterize the actual temperature of the power module in the electric drive system of each vehicle.

[0213] The second acquisition module 1203 is used to acquire the electric drive fitting curve characteristic table for each vehicle. The electric drive fitting curve characteristic table is used to characterize the relationship curve between the electric drive operating parameters and the reference temperature parameters for each vehicle.

[0214] The fitting module 1204 is used to fit the temperature parameters according to the preset electric drive operating condition parameters of each vehicle and the electric drive fitting curve characteristic table of each vehicle to obtain the reference temperature parameters of each vehicle.

[0215] The identification module 1205 is used to identify cooling anomalies in the power modules of each vehicle based on the reference temperature parameters of each vehicle, the actual temperature parameters of each vehicle, and the data feature time series of each vehicle, and to obtain the cooling anomaly risk parameters of each vehicle. The cooling anomaly risk parameters of each vehicle are used to characterize the risk of abnormal cooling of the power modules in each vehicle.

[0216] Optionally, in one optional implementation, the determining module 1202 is specifically used to: extract electric drive characteristic parameters of each vehicle at multiple time points from the electric drive signal of each vehicle, the electric drive characteristic parameters including: actual electric drive operating condition parameters and actual temperature parameters; extract driving characteristic parameters of each vehicle at multiple time points from the driving state signal of each vehicle; and generate a data characteristic time series of each vehicle based on the electric drive characteristic parameters and driving characteristic parameters at multiple time points.

[0217] Driving characteristic parameters include at least: driving start / stop indicator.

[0218] Optionally, in one alternative implementation, the determining module 1202 is further configured to: splice the data feature time series of each vehicle according to the driving start-stop identifier in the driving feature parameters to obtain a data segment of each vehicle under at least one driving cycle condition.

[0219] Optionally, in one optional implementation, the identification module 1205 is specifically used to: identify the cooling anomaly of the power module in each vehicle under each driving cycle condition based on the reference temperature parameter of each vehicle, the actual temperature parameter of each vehicle, and the data segment under each driving cycle condition of each vehicle, and obtain the cooling anomaly risk parameter of each vehicle under each driving cycle condition.

[0220] The preset electric drive operating parameters for each vehicle are: the preset electric drive operating parameter range for each vehicle.

[0221] Optionally, in one optional implementation, the determining module 1202 is specifically used to: segment the preset electric drive operating condition parameter range of each vehicle to obtain multiple electric drive operating condition sub-ranges for each vehicle; and, based on the multiple electric drive operating condition sub-ranges, use the electric drive fitting curve characteristic table of each vehicle to fit the temperature parameters in the multiple electric drive operating condition sub-ranges to obtain the reference temperature parameters of each vehicle in the multiple electric drive operating condition sub-ranges.

[0222] Optionally, in one optional implementation, the identification module 1205 is specifically used to: determine the target reference temperature parameter of the target operating condition parameter sub-interval where the actual electric drive operating condition parameter is located from the reference temperature parameters under each electric drive operating condition sub-interval, based on the actual electric drive operating condition parameters of each vehicle; and identify cooling anomalies in the power module of each vehicle based on the target reference temperature parameter, the actual temperature parameter of each vehicle, and the data feature time series of each vehicle, to obtain the cooling anomaly risk parameter of each vehicle.

[0223] Optionally, in one implementation, the identification module 1205 is specifically used to: determine the abnormal data points of each vehicle from the data feature time series of each vehicle based on the reference temperature parameters and the actual temperature parameters of each vehicle; determine the proportion of abnormal data in the data feature time series of each vehicle; and determine the cooling anomaly risk parameter of each vehicle based on the proportion of abnormal data.

[0224] The reference temperature parameters for each vehicle include: each reference temperature parameter within the reference temperature range for each vehicle.

[0225] Optionally, in one optional implementation, the identification module 1205 is specifically used to: determine the actual temperature parameter of each vehicle and the temperature parameter difference of each reference temperature parameter within the reference temperature range of each vehicle, to obtain multiple temperature deviation values ​​for each vehicle; determine the target temperature deviation value with the largest absolute value among the multiple temperature deviation values ​​for each vehicle; if the target temperature deviation value is greater than a preset temperature deviation threshold, mark the data feature of the target time point corresponding to the actual temperature parameter of each vehicle in the data feature time series of each vehicle as an abnormal data point.

[0226] Optionally, in one implementation, the identification module 1205 is specifically used to: determine the current risk factor coefficient of the data feature time series of each vehicle based on the proportion of abnormal data; generate a current segment detail table for each vehicle based on the current risk factor coefficient and the data feature time series of each vehicle; and determine the cooling anomaly risk parameter for each vehicle based on the current segment detail table of each vehicle and the historical risk factor coefficient of each vehicle.

[0227] Optionally, in one optional implementation, the second acquisition module 1203 is specifically used to: if the electric drive fitting curve characteristic table of the first vehicle is stored in advance, then acquire the electric drive fitting curve characteristic table of the first vehicle; if the electric drive characteristic table of the second vehicle is not stored in advance, then according to the preset electric drive operating condition parameters of the second vehicle, the data characteristic time series of the second vehicle, and multiple historical data characteristic sequences of the second vehicle, fit the relationship curve of the electric drive operating condition parameters and temperature parameters of the second vehicle to obtain the electric drive fitting curve characteristic table of the second vehicle.

[0228] The preset electric drive operating parameters of the second vehicle are: the preset electric drive operating parameter range of the second vehicle.

[0229] Optionally, in one implementation, the second acquisition module 1203 is specifically used to: segment the preset electric drive operating condition parameter range of the second vehicle to obtain multiple second electric drive operating condition sub-ranges of the second vehicle; group the data feature time series of the second vehicle and multiple historical data feature sequences of the second vehicle according to the multiple second electric drive operating condition sub-ranges to obtain a training dataset of the second vehicle under the multiple second electric drive operating condition sub-ranges; and perform relationship curve fitting between electric drive operating condition parameters and temperature parameters under the multiple second electric drive operating condition sub-ranges according to the training dataset under the multiple second electric drive operating condition sub-ranges to obtain an electric drive fitting curve characteristic table of the second vehicle under the multiple second electric drive operating condition sub-ranges.

[0230] Optionally, in one optional implementation, the second acquisition module 1203 is specifically used to: perform Monte Carlo simulation on the training dataset under each second electric drive condition sub-interval to obtain the parameter fitting curve and fitting function of each second electric drive condition sub-interval; and generate a characteristic table of electric drive fitting curves of the second vehicle under each second electric drive condition sub-interval based on the parameter fitting curve and fitting function of each second electric drive condition sub-interval.

[0231] It should be noted that for details not disclosed in the cooling anomaly identification device for vehicle power modules in this application embodiment, please refer to the details disclosed in the cooling anomaly identification method for vehicle power modules in this application embodiment, which will not be repeated here.

[0232] These modules can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more microprocessors, or one or more Field Programmable Gate Arrays (FPGAs). Alternatively, when a module is implemented using processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. Furthermore, these modules can be integrated together as a system-on-a-chip (SoC).

[0233] Figure 17 This is a schematic diagram of the structure of a cloud server provided in an embodiment of this application, such as... Figure 17 As shown, the cloud server 1300 may include a processor 1301, a memory 1302, and a bus. The memory 1302 stores machine-readable instructions executable by the processor 1301. When the electronic device is running, the machine-readable instructions are executed. The processor 1301 and the memory 1302 communicate via the bus. The processor 1301 is used to execute the steps of the cooling anomaly identification method for the vehicle power module in the above embodiments. It should be noted that the cloud server 1300 in this application embodiment and the cloud server 122 in the above embodiments are the same server and should not be construed as limiting this application.

[0234] The memory 1302, processor 1301, and bus components are electrically connected directly or indirectly to enable data transmission or interaction. For example, these components can be electrically connected to each other via one or more communication buses or signal lines. The mobile storage device includes at least one software function module that can be stored in the memory 1302 as software or firmware or embedded in the operating system (OS) of the electronic device. The processor 1301 is used to execute executable modules stored in the memory 1302, such as the software function modules and computer programs included in the cooling anomaly identification method of the automotive power module of the mobile storage medium.

[0235] The memory 1302 may be, but is not limited to, random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable read-only memory (EPROM), electrically erasable read-only memory (EEPROM), etc.

[0236] Optionally, embodiments of this application also provide a computer-readable storage medium storing a computer program. When the computer program is run by a processor, the processor executes the steps of the cooling anomaly identification method for the vehicle power module of the mobile storage medium described in the above embodiments. The specific implementation and technical effects are similar and will not be repeated here.

[0237] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0238] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0239] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units.

[0240] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0241] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for identifying cooling anomalies in automotive power modules, characterized in that, The cooling anomaly identification method includes: Acquire embedded data from multiple vehicles, and the embedded data for each vehicle includes: the electric drive signal of each vehicle and the driving status signal of each vehicle; Based on the electric drive signal and driving status signal of each vehicle, the actual temperature parameter and data feature time series of each vehicle are determined, wherein the actual temperature parameter is used to characterize the actual temperature of the power module in the electric drive system of each vehicle. Obtain the electric drive fitting curve characteristic table for each vehicle. The electric drive fitting curve characteristic table is used to characterize the relationship curve between the electric drive operating parameters and the reference temperature parameters for each vehicle. Based on the preset electric drive operating parameters of each vehicle, the temperature parameters are fitted using the electric drive fitting curve characteristic table of each vehicle to obtain the reference temperature parameters of each vehicle. Based on the reference temperature parameters of each vehicle, the actual temperature parameters of each vehicle, and the data feature time series of each vehicle, cooling anomalies are identified in the power modules of each vehicle to obtain cooling anomaly risk parameters for each vehicle. The cooling anomaly risk parameters for each vehicle are used to characterize the risk of abnormal cooling in the power modules of each vehicle. The step involves identifying cooling anomalies in the power modules of each vehicle based on reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, thereby obtaining cooling anomaly risk parameters for each vehicle, including: Based on the reference temperature parameters and the actual temperature parameters of each vehicle, abnormal data points of each vehicle are determined from the data feature time series of each vehicle. Determine the proportion of outlier data in the time series of data features for each vehicle; Based on the proportion of abnormal data, determine the current risk factor coefficient of the data feature time series of each vehicle; Based on the current risk factor coefficients and the data feature time series of each vehicle, generate a current segment detail table for each vehicle; The cooling anomaly risk parameters for each vehicle are determined based on the current segment details table and the historical risk factor coefficients for each vehicle.

2. The method for identifying cooling anomalies in automotive power modules according to claim 1, characterized in that, The step of determining the actual temperature parameters of each vehicle and the data characteristic time series of each vehicle based on the electric drive signal and the driving status signal of each vehicle includes: The electric drive characteristic parameters of each vehicle at multiple time points are extracted from the electric drive signal of each vehicle. The electric drive characteristic parameters include: actual electric drive operating condition parameters and actual temperature parameters. Extract the driving characteristic parameters of each vehicle at multiple time points from the driving status signal of each vehicle; Based on the electric drive characteristic parameters and driving characteristic parameters at the multiple time points, a data feature time series for each vehicle is generated.

3. The method for identifying cooling anomalies in automotive power modules according to claim 2, characterized in that, The driving characteristic parameters include at least: driving start / stop indicator; The method further includes: Based on the driving start-stop identifier in the driving characteristic parameters, the data feature time series of each vehicle are spliced ​​together to obtain a data segment of each vehicle under at least one driving cycle condition. The step involves identifying cooling anomalies in the power modules of each vehicle based on reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, thereby obtaining cooling anomaly risk parameters for each vehicle, including: Based on the reference temperature parameters of each vehicle, the actual temperature parameters of each vehicle, and the data segments of each vehicle under each driving cycle condition, the cooling anomaly of the power module in each vehicle under each driving cycle condition is identified, and the cooling anomaly risk parameters of each vehicle under each driving cycle condition are obtained.

4. The method for identifying cooling anomalies in automotive power modules according to claim 2, characterized in that, The preset electric drive operating condition parameters for each vehicle are: the preset electric drive operating condition parameter range for each vehicle; The step of fitting temperature parameters to each vehicle using the electric drive fitting curve characteristic table of each vehicle based on the preset electric drive operating condition parameters of each vehicle to obtain reference temperature parameters for each vehicle includes: The preset electric drive operating condition parameter range for each vehicle is segmented to obtain multiple electric drive operating condition sub-ranges for each vehicle. Based on the multiple electric drive operating condition sub-intervals, the temperature parameters of each vehicle are fitted in the multiple electric drive operating condition sub-intervals using the electric drive fitting curve characteristic table of each vehicle, so as to obtain the reference temperature parameters of each vehicle in the multiple electric drive operating condition sub-intervals. The step involves identifying cooling anomalies in the power modules of each vehicle based on reference temperature parameters, actual temperature parameters, and data feature time series of each vehicle, thereby obtaining cooling anomaly risk parameters for each vehicle, including: Based on the actual electric drive operating condition parameters of each vehicle, the target reference temperature parameter of the target operating condition parameter sub-interval where the actual electric drive operating condition parameters are located is determined from the reference temperature parameters under the electric drive operating condition sub-interval of each vehicle. Based on the target reference temperature parameter, the actual temperature parameter of each vehicle, and the data feature time series of each vehicle, cooling anomaly identification is performed on the power module in each vehicle to obtain the cooling anomaly risk parameter of each vehicle.

5. The method for identifying cooling anomalies in automotive power modules according to claim 1, characterized in that, The reference temperature parameters for each vehicle include: each reference temperature parameter within the reference temperature range for each vehicle; The step of determining abnormal data points for each vehicle from the data feature time series of each vehicle based on the reference temperature parameters and the actual temperature parameters of each vehicle includes: The actual temperature parameter of each vehicle and the temperature parameter difference of each reference temperature parameter within the reference temperature range of each vehicle are determined respectively to obtain multiple temperature deviation values ​​for each vehicle; Determine the target temperature deviation value with the largest absolute value among multiple temperature deviation values ​​for each vehicle; If the target temperature deviation value is greater than the preset temperature deviation threshold, the data features of the target time point corresponding to the actual temperature parameter of each vehicle in the data feature time series of each vehicle are marked as the abnormal data points.

6. The method for identifying cooling anomalies in automotive power modules according to claim 1, characterized in that, The step of obtaining the electric drive fitting curve characteristic table for each vehicle includes: If the electric drive fitting curve characteristic table of the first vehicle is stored in advance, then the electric drive fitting curve characteristic table of the first vehicle is obtained. If the electric drive characteristic table of the second vehicle is not stored in advance, then based on the preset electric drive operating condition parameters of the second vehicle, the data characteristic time series of the second vehicle, and multiple historical data characteristic sequences of the second vehicle, the relationship curve between the electric drive operating condition parameters and temperature parameters of the second vehicle is fitted to obtain the electric drive fitting curve characteristic table of the second vehicle.

7. The method for identifying cooling anomalies in automotive power modules according to claim 6, characterized in that, The preset electric drive operating condition parameters of the second vehicle are: the preset electric drive operating condition parameter range of the second vehicle; The step involves fitting a curve between the electric drive operating parameters and temperature parameters of the second vehicle based on preset electric drive operating parameters, the data feature time series of the second vehicle, and multiple historical data feature sequences of the second vehicle, to obtain an electric drive fitting curve characteristic table for the second vehicle, including: The preset electric drive operating condition parameter range of the second vehicle is segmented to obtain multiple second electric drive operating condition sub-ranges of the second vehicle. Based on the multiple second electric drive condition sub-intervals, the data feature time series of the second vehicle and the multiple historical data feature sequences of the second vehicle are grouped respectively to obtain the training dataset of the second vehicle under the multiple second electric drive condition sub-intervals. Based on the training dataset under the second electric drive operating condition sub-interval, the relationship curves between electric drive operating condition parameters and temperature parameters are fitted under the multiple second electric drive operating condition sub-intervals respectively, and the characteristic table of electric drive fitting curves of the second vehicle under the multiple second electric drive operating condition sub-intervals is obtained.

8. The method for identifying cooling anomalies in automotive power modules according to claim 7, characterized in that, The step involves fitting the relationship curves between the electric drive parameters and temperature parameters in the plurality of second electric drive condition sub-intervals based on the training dataset under the second electric drive condition sub-intervals, respectively, to obtain a characteristic table of the electric drive fitting curves of the second vehicle under the plurality of second electric drive condition sub-intervals, including: Monte Carlo simulations were performed on the training dataset for each second electric drive operating condition sub-interval to obtain the parameter fitting curve and fitting function for each second electric drive operating condition sub-interval. Based on the parameter fitting curve and fitting function of each second electric drive operating condition sub-interval, a characteristic table of electric drive fitting curves of the second vehicle under each second electric drive operating condition sub-interval is generated.

Images