A new energy automobile inverter fault identification method, system and electronic equipment

Abstract

This invention discloses a method, system, and electronic device for identifying inverter faults in new energy vehicles. The method includes the following steps: S1: Acquire real-time vehicle network data via CAN bus; filter the data; S2: Select vehicles whose motor and motor controller fault indicator lights illuminate twice consecutively as vehicles to be tested, and extract the set of times the fault indicator lights illuminate; S3: For each selected vehicle, iteratively determine whether the real-time vehicle network data meets the scenario condition group and the electrical signal condition group. If both are met, an inverter fault is identified, and fault information is output; S4: Train a machine learning classification model based on historical labeled data to replace manual rule condition grouping for inverter fault identification. Compared with existing technologies, the technical solution of this invention can dynamically monitor vehicle information in real time, accurately identify inverter faults, and improve vehicle safety and user satisfaction.

Description

Technical Field

[0001] This invention belongs to the field of vehicle fault identification technology, and more specifically, it relates to a method for identifying faults in inverters of new energy vehicles; this invention also relates to a fault identification system and electronic equipment for inverters of new energy vehicles. Background Technology

[0002] Inverters are a core component of new energy vehicles, and their fault identification directly impacts vehicle safety and after-sales service quality. With the increasing popularity of new energy vehicles, developing efficient and accurate fault diagnosis technologies is of significant practical importance for ensuring driving safety, improving maintenance efficiency, and enhancing user experience. However, traditional methods rely on fault indicator lights and simple codes, lacking the fusion and analysis of multi-dimensional information such as real-time vehicle operating status and historical fault records, leading to incomplete diagnostic results. Currently, traditional diagnostic systems can only provide fault prompts, failing to accurately pinpoint fault types, dynamically monitor vehicle status, or promptly detect potential faults. Their real-time performance and early warning capabilities are insufficient, and the information provided is limited.

[0003] Therefore, how to provide a method for identifying inverter faults in new energy vehicles that can dynamically monitor vehicle information in real time, accurately identify inverter faults, and improve vehicle safety and user satisfaction has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method, system, and electronic device for identifying inverter faults in new energy vehicles. These devices can dynamically monitor vehicle information in real time, accurately identify inverter faults, and improve vehicle safety and user satisfaction.

[0005] The technical solution provided by this invention is as follows: This invention provides a method for identifying inverter faults in new energy vehicles. The method includes: S1: Data acquisition and preprocessing, including: S101: Real-time vehicle network data is acquired via the CAN bus, the real-time vehicle network data including motor and motor controller fault indicator status L(t), power system fault indicator status P(t), vehicle status S(t), DTC fault code masking signal D(t), motor controller DC bus current I(t), insulation resistance R(t), drive motor status M(t), and rechargeable energy storage device current C(t); S102: The data is filtered to remove abnormal data exceeding a threshold, the threshold being the maximum value of the CAN transmission data; S2: Filtering motor and motor controller fault indicator status data... Vehicles whose controller fault indicator lights up twice consecutively are designated as vehicles to be tested, and the set of times the fault indicator lights up {t1, t2, ..., tn} is extracted; S3: The selected vehicles are checked for inverter faults in real time, including: S301: For each time ti∈{t1, t2, ..., tn}, real-time vehicle network data within the time range from ti to ti+T seconds is extracted, where T is a preset time interval; S302: It is determined whether the real-time vehicle network data meets the scenario condition group; S303: It is determined whether the real-time vehicle network data meets the electrical signal condition group; S304: If both the scenario condition group and the electrical signal condition group are met, it is determined to be an inverter fault, and the fault information Fault(V, ti, Ft) is output, where V is the vehicle identifier and Ft is the fault type; S4: A machine learning classification model is trained based on historical labeled data to replace the manual rule condition group for inverter fault identification.

[0006] Furthermore, in a preferred embodiment of the present invention, the scene condition set specifically comprises: Condition 1:

[0007] The indicator light for the motor and motor controller being on is represented as 1. The number of data entries where the indicator light for the motor and motor controller is on exceeds n1 within the time period from ti to ti+T. Condition 2: t∈[ti,ti+T],L(t)=1 The fault indicator light for the motor and motor controller is lit when it is 1, and L(t) is continuously 1 during the time period from ti to ti+T. Condition 3: t∈[ti,ti+T],P(t)=1 The power system fault indicator light being on indicates 1, and P(t) is continuously 1 during the time period from ti to ti+T. Condition 4:

[0008] When the number of vehicle states S(t) that are either starting or have failed to start is greater than n², the vehicle state S(t) is represented as follows: ; Condition 5:

[0009] The number of DTC fault code masking signals that appear in the segment is the number of CAN masking signals that are enabled, and the number of DTC fault code masking signals D(t) that are enabled is greater than n3; Condition 6:

[0010] When ACC is activated, the number of illuminated fault indicator lights for the motor and motor controller is greater than n4. Condition 7:

[0011] The number of doors that open on the left front door is greater than n5; Condition 8:

[0012] The number of abnormal statuses of the drive motor is greater than n6.

[0013] Furthermore, in a preferred embodiment of the present invention, the electrical signal condition set specifically comprises: Condition 1:

[0014] That is, the number of motor controller DC bus currents I(t) of 0.1 Amperes exceeds n7; Condition 2:

[0015] That is, the number of bars with an insulation resistance R(t) of 20000 ohms exceeds n8; Condition 3:

[0016] The number of rechargeable energy storage devices with a current C(t) of 0 amperes exceeds n9.

[0017] Furthermore, in a preferred embodiment of the present invention, step S4 includes: S401: recording vehicle data for fault and non-fault scenarios and labeling them; S402: using the data to train a classification model for real-time fault classification.

[0018] Furthermore, in a preferred embodiment of the present invention, the classification model is trained using one of the following methods: random forest, support vector machine, or gradient boosting tree.

[0019] Furthermore, in a preferred embodiment of the present invention, the vehicle data in S401 includes: historical current signal information, alarm information, driving status information, and historical maintenance information.

[0020] Furthermore, in a preferred embodiment of the present invention, the historical current signal information includes the DC bus current of the motor controller, the insulation resistance, and the current of the rechargeable energy storage device; the driving status information includes the vehicle owner status signal and the door opening signal; the historical current signal information, alarm information, and driving status information are obtained through vehicle network data transmission; and the historical maintenance information is obtained through dealership maintenance records.

[0021] Furthermore, this invention also provides a fault identification system for new energy vehicle inverters. The system includes: a data acquisition module for acquiring real-time vehicle network data via a CAN bus, filtering the data to remove abnormal data exceeding a threshold, where the threshold is the maximum value of the CAN transmission data; a fault moment extraction module for selecting vehicles whose motor and motor controller fault indicator lights illuminate twice consecutively as vehicles to be detected; a judgment and identification module for sequentially judging whether the real-time vehicle network data of each selected vehicle meets the scenario condition group and the electrical signal condition group; if both are met, the system determines it to be an inverter fault and outputs fault information; a model optimization module for training a machine learning classification model based on historical labeled data to replace manual rule condition groups for inverter fault identification; and a database and optimized model storage module for storing vehicle data, fault records, and model parameters.

[0022] Furthermore, in a preferred embodiment of the present invention, the judgment and identification module includes: a condition judgment module, used to judge whether the real-time vehicle network data meets preset scenario conditions and current-related signal conditions; and a fault identification module, used to judge whether an inverter fault has occurred based on the scenario conditions and output fault vehicle information.

[0023] In addition, the present invention also provides an electronic device, comprising: a memory for storing a computer program for executing the new energy vehicle inverter fault identification method; and a processor for processing the computer program.

[0024] This invention provides a fault identification method for inverters in new energy vehicles, which, compared with existing technologies, includes: S1: Data acquisition and preprocessing, including: S101: Real-time vehicle network data is acquired via the CAN bus, the real-time vehicle network data including the status of the motor and motor controller fault indicator L(t), the status of the power system fault indicator P(t), the vehicle status S(t), the DTC fault code masking signal D(t), the DC bus current I(t) of the motor controller, the insulation resistance R(t), the status of the drive motor M(t), and the current C(t) of the rechargeable energy storage device; S102: The data is filtered to remove abnormal data exceeding a threshold, the threshold being the maximum value of the CAN transmission data; S2: Filtering the motor... Vehicles whose motor controller fault indicator lights up twice consecutively are selected as vehicles to be tested, and the set of times when the fault indicator lights up is {t1, t2, ..., tn} is extracted; S3: The inverter fault is judged one by one in real time for the selected vehicles, including: S301: For each time ti∈{t1, t2, ..., tn}, real-time vehicle network data within the time range of ti to ti+T seconds is extracted, where T is a preset time interval; S302: It is determined whether the real-time vehicle network data meets the scenario condition group; S303: It is determined whether the real-time vehicle network data meets the electrical signal condition group; S304: If both the scenario condition group and the electrical signal condition group are met, it is determined to be an inverter fault, and the fault information Fault(V, ti, Ft) is output, where V is the vehicle identifier and Ft is the fault type; S4: A machine learning classification model is trained based on historical labeled data to replace the manual rule condition group for inverter fault identification. In the execution of the technical solution involved in this invention, firstly, step S101 collects data through the CAN bus to achieve real-time monitoring of the vehicle's overall operating status. Then, step S102 filters out abnormal data exceeding the maximum value of CAN transmission data, eliminating noisy data and thus improving the overall data quality of the fault identification system. Step S2 selects only vehicles whose indicator lights illuminate twice consecutively, effectively reducing data redundancy. Step S3 captures inverter faults through multi-condition comprehensive judgment. Finally, step S4 continuously improves the accuracy of fault identification by establishing and optimizing a machine learning model, ensuring vehicle operational safety and providing strong support for subsequent fault statistics and maintenance services. Compared with existing technologies, the technical solution involved in this invention can dynamically monitor vehicle information in real time, accurately identify inverter faults, and improve vehicle safety and user satisfaction.

[0025] In addition, the present invention also provides a fault identification system and electronic device for new energy vehicle inverters, which also have the above-mentioned technical effects. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 is a schematic diagram of the new energy vehicle inverter fault identification method provided in an embodiment of the present invention. Detailed Implementation

[0028] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0029] It should be noted that when a component is referred to as being "fixed to" or "set on" another component, it can be directly on or indirectly set on the other component; when a component is referred to as being "connected to" another component, it can be directly connected to or indirectly connected to the other component.

[0030] It should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "first", "second", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the present invention 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. Therefore, they should not be construed as limitations on the present invention.

[0031] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" or "several" means two or more, unless otherwise explicitly specified.

[0032] It should be noted that the structures, proportions, sizes, etc., shown in the accompanying drawings of this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0033] Please Figure 1 As shown, the new energy vehicle inverter fault identification method provided in this embodiment of the invention includes: S1: Data acquisition and preprocessing, including: S101: Real-time vehicle network data is acquired in real time through the CAN bus, the real-time vehicle network data including motor and motor controller fault indicator status L(t), power system fault indicator status P(t), vehicle status S(t), DTC fault code masking signal D(t), motor controller DC bus current I(t), insulation resistance R(t), drive motor status M(t), and rechargeable energy storage device current C(t); S102: The data is filtered to remove abnormal data exceeding a threshold, the threshold being the maximum value of the CAN transmission data; S2: Filtering motor and... Vehicles whose motor controller fault indicator lights up twice consecutively are designated as vehicles to be tested, and the set of times the fault indicator lights up {t1, t2, ..., tn} is extracted; S3: The selected vehicles are checked for inverter faults in real time loop, including: S301: For each time ti∈{t1, t2, ..., tn}, real-time vehicle network data within the time range from ti to ti+T seconds is extracted, where T is a preset time interval; S302: It is determined whether the real-time vehicle network data meets the scenario condition group; S303: It is determined whether the real-time vehicle network data meets the electrical signal condition group; S304: If both the scenario condition group and the electrical signal condition group are met, it is determined to be an inverter fault, and the fault information Fault(V, ti, Ft) is output, where V is the vehicle identifier and Ft is the fault type; S4: A machine learning classification model is trained based on historical labeled data to replace the manual rule condition group for inverter fault identification.

[0034] Specifically, in the technical solution of the new energy vehicle inverter fault identification method provided by this embodiment of the invention, step S101 collects data through the CAN bus to achieve real-time monitoring of the vehicle's all-round operating status; step S102 filters abnormal data exceeding the maximum value of CAN transmission data and eliminates noisy data, thereby improving the data quality of the overall fault identification system; step S2 selects only vehicles whose indicator lights illuminate twice consecutively to effectively reduce data redundancy; step S3 captures inverter faults through comprehensive judgment of multiple conditions; and finally, step S4 continuously improves the accuracy of fault identification by establishing a machine learning model and optimizing the model, ensuring vehicle operating safety and providing strong support for subsequent fault statistics and maintenance services. The technical solution provided by this invention can dynamically monitor vehicle information in real time, accurately identify inverter faults, and improve vehicle safety and user satisfaction.

[0035] The S2 screening method identifies vehicles whose motor and motor controller fault indicator lights illuminate twice consecutively as vehicles to be inspected. This effectively reduces the number of vehicles requiring inverter fault detection, improves inspection efficiency, and ensures that high-risk vehicles receive focused attention.

[0036] It should be noted that in existing technologies, fault indicator lights provide limited information, only offering fault indications and failing to accurately pinpoint the fault type. This invention, upon determining an inverter fault, outputs fault information Fault(V, ti, Ft), where V is the vehicle identifier and Ft is the fault type, providing strong support for subsequent fault statistics and maintenance services. Furthermore, existing technologies cannot analyze vehicle status in real time, making it difficult to provide early warnings of potential faults. In this embodiment of the invention, real-time monitoring of the vehicle's all-around operating status is achieved based on vehicle network big data, preventing unexpected situations from occurring.

[0037] Specifically, in this embodiment of the invention, the scene condition group is as follows: Condition 1:

[0038] The indicator light for the motor and motor controller being on is represented as 1. The number of data entries where the indicator light for the motor and motor controller is on exceeds n1 within the time period from ti to ti+T. Condition 2: t∈[ti,ti+T],L(t)=1 The fault indicator light for the motor and motor controller is lit when it is 1, and L(t) is continuously 1 during the time period from ti to ti+T. Condition 3: t∈[ti,ti+T],P(t)=1 The power system fault indicator light being on indicates 1, and P(t) is continuously 1 during the time period from ti to ti+T. Condition 4:

[0039] When the number of vehicle states S(t) that are either starting or have failed to start is greater than n², the vehicle state S(t) is represented as follows: ; Condition 5:

[0040] The number of DTC fault code masking signals that appear in the segment is the number of CAN masking signals that are enabled, and the number of DTC fault code masking signals D(t) that are enabled is greater than n3; Condition 6:

[0041] When ACC is activated, the number of illuminated fault indicator lights for the motor and motor controller is greater than n4. Condition 7:

[0042] The number of doors that open on the left front door is greater than n5; Condition 8:

[0043] The number of abnormal statuses of the drive motor is greater than n6.

[0044] The conditions are as follows: Condition 1: A continuously illuminated fault indicator light suggests a possible serious fault in the motor, motor controller, or related components such as the inverter; Condition 2: As the fault indicator light serves as a fault indicator, a continuous illumination of the motor and motor controller fault indicator lights throughout the segment strongly suggests a fault; Condition 3: A continuously illuminated powertrain fault indicator light suggests a possible serious fault in the powertrain system, including the battery, motor, and inverter; Condition 4: Frequent vehicle starts within a short period suggest a possible problem with the starting system or powertrain system; Condition 5: Frequent diagnostic fault codes and CAN bus signal activation suggest a possible fault in the communication or control system; Condition 6: Frequent illumination of the fault indicator light when ACC is activated suggests a possible fault in the motor, motor controller, or related components such as the inverter in assisted driving mode; Condition 7: Frequent opening of the left front door may indicate a fault or abnormal operation of the door system; Condition 8: Frequent entry of the drive motor into abnormal states may indicate a fault in the motor or inverter. The above scenarios encompass the possible changes in real-time vehicle network data and the formulas that must be satisfied when an inverter malfunctions, which helps in determining the status of the vehicle inverter.

[0045] Specifically, in this embodiment of the invention, the electrical signal condition group is as follows: Condition 1:

[0046] That is, the number of motor controller DC bus currents I(t) of 0.1 Amperes exceeds n7; Condition 2:

[0047] That is, the number of bars with an insulation resistance R(t) of 20000 ohms exceeds n8; Condition 3:

[0048] The number of rechargeable energy storage devices with a current C(t) of 0 amperes exceeds n9.

[0049] It should be noted that in this embodiment, condition 1 is set because, to ensure the inverter operates under normal conditions, the DC bus current collected during the specified time period must always be at an extremely low level. The DC bus current of the motor controller refers to the current on the DC bus in the motor controller, which is the total current of the motor's input power supply. In the motor, the bus current includes all the current flowing through the three phases of the motor and is an important indicator of the motor controller's performance. The DC bus capacitor in the motor controller plays a role in smoothing the bus voltage, reducing peak voltage, absorbing pulse current, and preventing overcharging or transient voltage from affecting the controller. Condition 2 is set because if the insulation resistance is frequently 20,000 ohms, it may indicate a decline in the insulation performance of the high-voltage system and pose a safety hazard. Therefore, the information of an insulation resistance of 20,000 ohms cannot exceed a certain fixed value. In this embodiment of the invention, this fixed value is represented as n8. The current of the rechargeable energy storage device represents the current value of the energy storage device such as the battery pack or supercapacitor during the charging process. It is required that the DC bus of the motor controller has only a very small current, and that the number of 0 Ampere currents of the rechargeable energy storage device can be captured under non-charging conditions should be controlled at a fixed value. In this embodiment of the invention, this fixed value is represented as n9.

[0050] In one embodiment of the invention, when a new energy vehicle brand has 100,000 vehicles in use, n1=20, n2=10, n3=20, n4=15, n5=1, n6=10, n7=20, n8=10, n9=20.

[0051] Specifically, in this embodiment of the invention, step S4 includes: S401: recording vehicle data for fault and non-fault scenarios and labeling them; S402: using the data to train a classification model for real-time fault classification.

[0052] Specifically, in this embodiment of the invention, the classification model is trained using one of the following methods: random forest, support vector machine, or gradient boosting tree.

[0053] It should be noted that in S402, supervised learning modeling is performed on the aforementioned data. Machine learning algorithms such as random forest, support vector machine, and gradient boosting tree can be used to train a model with classification and discrimination capabilities. This model can quickly identify whether an inverter fault exists during real-time vehicle operation, improving the intelligence level of fault identification. The trained classification model can be embedded into the on-board diagnostic system or back-end server, automatically parsing and identifying vehicle status data without the need for additional sensors or manual intervention, enhancing the system's generalization ability and robustness under real-world conditions. Specifically, in this embodiment of the invention, the vehicle data in S401 includes: historical current signal information, alarm information, driving status information, and historical maintenance information.

[0054] Specifically, in this embodiment of the invention, the historical current signal information includes the DC bus current of the motor controller, the insulation resistance, and the current of the rechargeable energy storage device; the driving status information includes the vehicle owner status signal and the door opening signal; the historical current signal information, alarm information, and driving status information are obtained through vehicle network data transmission; and the historical maintenance information is obtained through dealership maintenance records.

[0055] It should be noted that historical current signal information can reflect the typical state characteristics of the vehicle's powertrain and high-voltage components during operation. Analyzing the changing trends of this data under different fault and non-fault scenarios helps to uncover the electrical characteristic patterns of potential inverter faults. Owner status signals and door opening signals can be used to assist in determining the context in which a fault occurs, effectively avoiding false alarms caused by abnormal operation or usage. Alarm information comes directly from fault codes or alarm signals triggered by the vehicle during operation, providing direct evidence of vehicle abnormalities. Cross-referencing these alarms with other data can further improve the reliability and accuracy of fault identification. In addition, historical maintenance information, through structured extraction of dealership maintenance records, forms a knowledge graph of vehicle history faults and maintenance behaviors, providing a real-world label reference for the classification model.

[0056] Furthermore, this invention also provides a fault identification system for new energy vehicle inverters. The system includes: a data acquisition module for acquiring real-time vehicle network data via a CAN bus, filtering the data to remove abnormal data exceeding a threshold, where the threshold is the maximum value of the CAN transmission data; a fault moment extraction module for selecting vehicles whose motor and motor controller fault indicator lights illuminate twice consecutively as vehicles to be detected; a judgment and identification module for sequentially judging whether the real-time vehicle network data of each selected vehicle meets the scenario condition group and the electrical signal condition group; if both are met, the system determines it to be an inverter fault and outputs fault information; a model optimization module for training a machine learning classification model based on historical labeled data to replace manual rule condition groups for inverter fault identification; and a database and optimized model storage module for storing vehicle data, fault records, and model parameters.

[0057] Specifically, in this embodiment of the invention, the judgment and identification module includes: a condition judgment module, used to judge whether the real-time vehicle network data meets the preset scenario conditions and current-related signal conditions; and a fault identification module, used to judge whether an inverter fault has occurred based on the scenario conditions and output fault vehicle information.

[0058] The new energy vehicle inverter fault identification system described in the above embodiments also has the aforementioned technical effects.

[0059] In addition, the present invention also provides an electronic device, comprising: a memory for storing a computer program for executing the new energy vehicle inverter fault identification method; and a processor for processing the computer program, which also has the above-mentioned technical effects.

[0060] More specifically, this invention provides a fault identification solution for new energy vehicle inverters. Based on vehicle network big data, it monitors vehicle status in real time, provides early warnings of potential faults, and ensures safe vehicle operation. In terms of data analysis, it combines multi-dimensional data such as vehicle fault logs, vehicle status, and historical fault records, which better reflects actual engineering scenarios and improves the accuracy of fault identification. Finally, it utilizes machine learning optimization, using machine learning algorithms to refine the fault identification model. This allows for rule-based fault identification in the early stages and automatic identification through the machine learning model in the later stages, further improving the accuracy and recall rate of fault identification and ensuring the system's flexibility and adaptability. The technical solution provided by this invention does not require additional sensors; it achieves accurate fault identification solely through existing vehicle systems and vehicle network data, resulting in low implementation costs and greater customer acceptance.

[0061] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for fault identification in a new energy vehicle inverter, characterized in that, The method includes the following steps: S1: Data acquisition and preprocessing, including the following steps: S101: Real-time vehicle network data is acquired via CAN bus. The real-time vehicle network data includes: motor and motor controller fault indicator status L(t), power system fault indicator status P(t), vehicle status S(t), DTC fault code shielding signal D(t), motor controller DC bus current I(t), insulation resistance R(t), drive motor status M(t), and rechargeable energy storage device current C(t). S102: Filter the data to remove abnormal data that exceeds a threshold, where the threshold is the maximum value of the CAN transmission data; S2: Select vehicles whose motor and motor controller fault indicator lights illuminate twice consecutively as vehicles to be tested, and extract the set of times when the fault indicator lights illuminate: {t1, t2, ..., tn}; S3: Real-time, cyclical analysis of inverter faults for each selected vehicle, including the following steps: S301: For each time ti∈{t1, t2, ..., tn}, extract real-time vehicle network data within the time range from ti to ti+T seconds, where T is a preset time interval; S302: Determine whether the real-time vehicle network data meets the scenario condition group; S303: Determine whether the real-time vehicle network data meets the electrical signal condition group; S304: If both the scenario condition group and the electrical signal condition group are met, the inverter is determined to be faulty, and the fault information Fault(V, ti, Ft) is output, where V is the vehicle identifier and Ft is the fault type. S4: Train a machine learning classification model based on historical labeled data to replace manual rule condition groups for inverter fault identification; The specific scenario condition group is as follows: Condition 1: The fault indicator light for the motor and motor controller being on indicates 1, and in t i to t i Within the +T time period, the number of data entries with the fault indicator lights on the motor and motor controller exceeding n1; Condition 2: t∈[ti,ti+T],L(t)=1 The fault indicator light for the motor and motor controller being on indicates 1, and in t i to t i During the time interval +T, L(t) is continuously 1; Condition 3: t∈[ti,ti+T], P(t)=1 The power system fault indicator light being illuminated indicates 1, in t i to t i During the time interval +T, P(t) is continuously 1; Condition 4: When the number of vehicle states S(t) that are either starting or have failed to start is greater than n², the vehicle state S(t) is represented as follows: ; Condition 5: The number of DTC fault code masking signals that appear in the segment is the number of CAN masking signals that are enabled, and the number of DTC fault code masking signals D(t) that are enabled is greater than n3; Condition 6: When ACC is activated, the number of illuminated fault indicator lights for the motor and motor controller is greater than n4. Condition 7: The number of doors that open on the left front door is greater than n5; Condition 8: The number of abnormal statuses of the drive motor is greater than n6; The electrical signal condition group specifically includes: Condition 1: That is, the number of motor controller DC bus currents I(t) of 0.1 Amperes exceeds n7; Condition 2: That is, the number of bars with an insulation resistance R(t) of 20000 ohms exceeds n8; Condition 3: The number of rechargeable energy storage devices with a current C(t) of 0 amperes exceeds n9.

2. The method for fault identification of new energy vehicle inverters according to claim 1, characterized in that, Step S4 includes: S401: Record vehicle data for both faulty and non-faulty scenarios and label them; S402: Use the data to train a classification model for real-time fault classification.

3. The method for fault identification of new energy vehicle inverters according to claim 2, characterized in that, The classification model is trained using one of the following methods: random forest, support vector machine, or gradient boosting tree.

4. The method for fault identification of new energy vehicle inverters according to claim 3, characterized in that, The vehicle data mentioned in S401 includes: historical current signal information, alarm information, driving status information, and historical maintenance information.

5. The method for fault identification of new energy vehicle inverters according to claim 4, characterized in that, The historical current signal information includes the DC bus current of the motor controller, insulation resistance, and current of the rechargeable energy storage device. The driving status information includes the vehicle owner status signal and the door opening signal; The historical current signal information, alarm information, and driving status information are obtained through vehicle network data transmission. The historical repair information is obtained from the repair records of authorized dealerships.

6. A fault identification system for inverters in new energy vehicles, characterized in that, This system is used to implement the fault identification method for new energy vehicle inverters according to any one of claims 1 to 5, comprising: The data acquisition module is used to acquire real-time vehicle network data through the CAN bus, filter the data, and filter out abnormal data that exceeds a threshold, where the threshold is the maximum value of the CAN transmission data. The fault moment extraction module is used to filter vehicles whose motor and motor controller fault indicator lights illuminate twice consecutively as vehicles to be tested. The judgment and identification module is used to judge whether the real-time vehicle network data of each selected vehicle meets the scenario condition group and the electrical signal condition group in real time. If both are met, it is judged as an inverter fault and the fault information is output. The model optimization module is used to train a machine learning classification model based on historical labeled data, replacing manual rule condition sets for inverter fault identification. The database and optimization model storage module is used to store vehicle data, fault records, and model parameters.

7. A fault identification system for a new energy vehicle inverter according to claim 6, characterized in that, The judgment and recognition module includes: The condition judgment module is used to determine whether the real-time vehicle network data meets the preset scenario conditions and current-related signal conditions; The fault identification module is used to determine whether an inverter fault has occurred based on the scenario conditions and output fault vehicle information.

8. An electronic device, characterized in that, include: A computer program, wherein the computer program is used to execute the new energy vehicle inverter fault identification method according to any one of claims 1 to 5; A memory for storing the computer program; A processor for executing the computer program.

Images