System and method for failure detection of vehicle components, vehicle, medium and product

Abstract

The application discloses a vehicle part fault detection system, a vehicle part fault detection method, a vehicle, a computer readable storage medium and a computer program product. The vehicle part fault detection system comprises a vehicle end fault diagnosis module, a cloud end fault diagnosis module and a vehicle-mounted communication module; the vehicle end fault diagnosis module is configured to perform first fault diagnosis processing on the vehicle part according to first type data of the vehicle part, wherein the collection time length of the first type data is less than a first preset time length; the cloud end fault diagnosis module is configured to perform second fault diagnosis processing on the vehicle part according to second type data of the vehicle part, wherein the collection time length of the second type data is greater than the first preset time length. In this way, the vehicle end fault diagnosis module can realize rapid diagnosis of instant faults of the vehicle part, ensuring the timeliness of fault processing; the cloud end fault diagnosis module can realize potential fault analysis of the part, improving the accuracy of fault diagnosis.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a fault detection system for vehicle components, a fault detection method for vehicle components, a vehicle, a computer-readable storage medium, and a computer program product. Background Technology

[0002] Faulty vehicle components can lead to safety accidents, such as battery fires or explosions caused by overvoltage or thermal runaway, posing safety hazards. Therefore, there is an urgent need for a vehicle component fault detection system to monitor vehicle components in real time, thereby ensuring the safety of passengers and the vehicle itself. Summary of the Invention

[0003] This application provides a fault detection system for vehicle components, a fault detection method for vehicle components, a vehicle, a computer-readable storage medium, and a computer program product.

[0004] This application provides a fault detection system for vehicle components, the system including a vehicle-side fault diagnosis module, a cloud-based fault diagnosis module, and an in-vehicle communication module; the vehicle-side fault diagnosis module and the cloud-based fault diagnosis module establish a communication connection through the in-vehicle communication module; The vehicle-side fault diagnosis module is configured to perform a first fault diagnosis process on the vehicle components based on a first type of data of the vehicle components, wherein the collection time of the first type of data required for the first fault diagnosis process is less than a first preset time. The cloud-based fault diagnosis module is configured to perform a second fault diagnosis process on the vehicle components based on the second type of data of the vehicle components, wherein the collection time of the second type of data required for the second fault diagnosis process is longer than the first preset time.

[0005] Thus, by processing the first type of data through the vehicle-side fault diagnosis module, rapid identification and diagnosis of immediate faults in vehicle components can be achieved, reducing the diagnostic delay of real-time faults to a certain extent and ensuring the timeliness of fault handling. By processing the second type of data through the cloud-based fault diagnosis module, accurate analysis of potential faults and long-term operating trends of components can be achieved based on cloud computing resources and complex algorithms, improving the accuracy of fault diagnosis and realizing the rational allocation and utilization of resources. Compared with a single diagnostic processing path, the embodiment of this application establishes a bidirectional communication connection between the vehicle-side fault diagnosis module and the cloud-based fault diagnosis module through the vehicle communication module. Based on the first and second types of data, the vehicle-side and cloud-based modules can cooperate in a division of labor, balancing the timeliness and accuracy of fault diagnosis.

[0006] In some implementations, the system further includes an external module; The external module is configured to perform a third fault diagnosis process on the vehicle components based on the first type of data when the computing resources of the vehicle-side fault diagnosis module are insufficient.

[0007] Thus, the system also includes an external module; this external module is configured to perform third-level fault diagnosis processing on vehicle components based on the first type of data when the computing resources of the vehicle-side fault diagnosis module are insufficient. In this way, even when the computing resources of the vehicle-side fault diagnosis module are insufficient, the external module can perform third-level fault diagnosis processing on the first type of data, ensuring coverage of multiple fault detection scenarios and thereby improving the effectiveness of fault detection.

[0008] In some embodiments, the system includes an electric drive module connected to the vehicle-side fault diagnosis module and the vehicle-mounted communication module; The electric drive module is configured as follows: Acquire the first type of data and the second type of data of the vehicle components; The first type of data is transmitted to the vehicle-side fault diagnosis module; and / or The second type of data is transmitted to the cloud-based fault diagnosis module through the vehicle communication module.

[0009] Thus, the system also includes an electric drive module, which is connected to the vehicle-side fault diagnosis module and the vehicle-mounted communication module. The electric drive module is configured to acquire first-type and second-type data from vehicle components; transmit the first-type data to the vehicle-side fault diagnosis module; and transmit the second-type data to the cloud-based fault diagnosis module via the vehicle-mounted communication module. In this way, the vehicle-side fault diagnosis module can perform initial diagnostic processing on the first-type data collected and transmitted by the electric drive module, enabling rapid data flow from the acquisition end to the processing end. This ensures that the first-type data is processed efficiently and accurately at the vehicle end, thereby improving the system's reliability and stability to a certain extent.

[0010] In some embodiments, the cloud-based fault diagnosis module is configured to send fault diagnosis information to the electric drive module via the vehicle communication module in the event of a fault in the vehicle component; and / or The vehicle-side fault diagnosis module is configured to send fault diagnosis information to the electric drive module in the event of a fault in the vehicle component. The electric drive module is configured to control the vehicle components based on the received fault diagnosis information.

[0011] Thus, the cloud-based fault diagnosis module is configured to send fault diagnosis information to the electric drive module via the vehicle communication module in the event of a vehicle component failure; the vehicle-side fault diagnosis module is configured to send fault diagnosis information to the electric drive module in the event of a vehicle component failure; and the electric drive module is configured to control the vehicle components based on the received fault diagnosis information. In this way, both the cloud-based and vehicle-side fault diagnosis modules can send fault diagnosis information to the electric drive module in the event of a vehicle component failure, enabling the electric drive module to control the vehicle components based on the received fault diagnosis information, thereby preventing safety accidents and ensuring vehicle operational safety.

[0012] In some implementations, the cloud-based fault diagnosis module is configured as follows: If, in the case where data in the second type of data is continuously missing within a second preset time period, the data excluding the time period containing the missing data in the second type of data is determined as the target data; and / or In cases where data in the second type of data is in a non-continuous missing state, the data after imputation processing of the second type of data is determined as the target data; and / or If there are no missing data in the second type of data, then the second type of data is determined to be the target data.

[0013] Thus, the cloud-based fault diagnosis module is configured to, in cases where data in the second type of data is continuously missing within a second preset time period, identify data after the time period containing the missing data in the second type of data as the target data; in cases where data in the second type of data is not continuously missing, identify data after imputation processing of the second type of data as the target data; and in cases where there is no missing data in the second type of data, identify the second type of data as the target data. By classifying and preprocessing the second type of data according to its missing status, continuously missing data can be eliminated, and non-continuously missing values ​​can be accurately supplemented. This avoids diagnostic biases, misjudgments, and missed judgments caused by missing data, and to a certain extent improves the accuracy and reliability of the second fault diagnosis processing of the cloud-based fault diagnosis module, enabling precise identification of potential faults and performance degradation trends of vehicle components.

[0014] In some implementations, the cloud-based fault diagnosis module is configured as follows: The target data is subjected to feature extraction processing to determine the target feature data; The target feature data is clustered to obtain multiple clusters and the position and target distance of each cluster, wherein the target distance is the distance between the cluster and the target center position, and the target center position is the position of the cluster closest to the safety window among the multiple clusters; Based on the target distance, fault diagnosis information for the vehicle components is determined.

[0015] Thus, the cloud-based fault diagnosis module is configured to perform feature extraction on the target data to determine the target feature data; perform clustering on the target feature data to obtain multiple clusters and the position and target distance of each cluster, where the target distance is the distance between the cluster and the target center, and the target center is the position of the cluster closest to the safety window among the multiple clusters; and determine the fault diagnosis information of the vehicle parts based on the target distance. In this way, by performing feature extraction on the target data to obtain the target feature data, interference from invalid data can be eliminated, allowing subsequent fault diagnosis to be based on the feature data. By dividing the target feature data into multiple clusters through clustering, determining the position of each cluster and its target distance from the target center, and then accurately determining the fault diagnosis information of the vehicle parts based on the correlation between the target distance and the safety window, fault result deduction can be achieved, ensuring the accuracy of cloud-based fault diagnosis and improving the ability to identify unknown fault types to a certain extent.

[0016] This application provides a method for fault detection of vehicle components. The fault detection method is used in a vehicle component fault detection system, and the method includes: Based on the first type of data of the vehicle parts, a first fault diagnosis process is performed on the vehicle parts, wherein the collection time of the first type of data required for the first fault diagnosis process is less than a first preset time. Based on the second type of data of the vehicle components, a second fault diagnosis process is performed on the vehicle components, wherein the time required for collecting the second type of data for the second fault diagnosis process is longer than the first preset time.

[0017] Thus, based on the first type of data from vehicle components, a first fault diagnosis process is performed on the vehicle components, wherein the data acquisition time required for the first fault diagnosis process is less than a first preset time. Based on the second type of data from vehicle components, a second fault diagnosis process is performed on the vehicle components, wherein the data acquisition time required for the second fault diagnosis process is greater than the first preset time. In this way, by processing the first type of data through the vehicle-side fault diagnosis module, rapid identification and diagnosis of immediate faults in vehicle components can be achieved, reducing the diagnostic delay of real-time faults to a certain extent and ensuring the timeliness of fault handling. By processing the second type of data through the cloud-based fault diagnosis module, accurate analysis of potential faults and long-term operating trends of components can be achieved based on cloud computing resources and complex algorithms, improving the accuracy of fault diagnosis and achieving reasonable allocation and utilization of resources. Compared to a single diagnostic processing path, the embodiment of this application establishes a bidirectional communication connection between the vehicle-side fault diagnosis module and the cloud-based fault diagnosis module through an onboard communication module. Based on the first and second types of data, the vehicle-side and cloud-based modules can collaborate, balancing the timeliness and accuracy of fault diagnosis.

[0018] This application provides a vehicle including a memory and a processor. The memory stores a computer program, which, when executed by the processor, implements the steps of the above-described method.

[0019] This application provides a computer-readable storage medium storing a computer program that, when executed by one or more processors, implements the steps of the above-described method.

[0020] This application provides a computer program product, which includes a computer program / instruction that, when executed by a processor, implements the steps of the above-described method.

[0021] The vehicle, computer-readable storage medium, and computer program product provided in this application can perform a first fault diagnosis process on vehicle components based on a first type of data of the vehicle components, wherein the acquisition time of the first type of data required for the first fault diagnosis process is less than a first preset time; and perform a second fault diagnosis process on vehicle components based on a second type of data of the vehicle components, wherein the acquisition time of the second type of data required for the second fault diagnosis process is greater than the first preset time. In this way, by performing feature extraction processing on the target data to obtain target feature data, interference from invalid data can be eliminated, allowing subsequent fault diagnosis to be carried out based on the feature data. By using clustering processing to divide the target feature data into multiple clusters, determining the position of each cluster and the target distance from the target center position, and then accurately determining the fault diagnosis information of the vehicle components based on the correlation between the target distance and the safety window, fault result deduction can be achieved, ensuring the accuracy of cloud-based fault diagnosis and improving the ability to identify unknown fault types to a certain extent.

[0022] Additional aspects and advantages of embodiments of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of this application. Attached Figure Description

[0023] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, wherein: Figure 1 This is a flowchart illustrating a method for detecting faults in vehicle components according to certain embodiments of this application. Figure 2 This is one of the schematic diagrams of a vehicle component fault detection system according to certain embodiments of this application; Figure 3 This is a second schematic diagram of a vehicle component fault detection system according to certain embodiments of this application; Figure 4 This is a third schematic diagram of a vehicle component fault detection system according to certain embodiments of this application; Figure 5 This is a schematic diagram of the structure of a fault analysis and detection instrument according to certain embodiments of this application; Figure 6 This is a schematic diagram of a buck circuit according to certain embodiments of this application; Figure 7 This is a schematic diagram of the data acquisition and processing logic of some embodiments of this application; Figure 8 This is a schematic diagram of the data processing mechanism in some embodiments of this application; Figure 9 This is a schematic diagram of the fault diagnosis algorithm of some embodiments of this application; Figure 10 This is a schematic diagram of hierarchical clustering partitioning in some embodiments of this application. Detailed Implementation

[0024] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of this application, and should not be construed as limiting the embodiments of this application.

[0025] The proper functioning of vehicle components is fundamental to vehicle safety. Failures in any component can easily lead to various traffic accidents, threatening the lives of drivers and passengers. For example, the power battery, a core component of new energy vehicles, can easily cause serious accidents such as battery fires and explosions if it experiences overvoltage, overcharging, or thermal runaway. Furthermore, failures in critical components such as the vehicle's electric drive and power distribution systems can also lead to power interruptions and loss of control, greatly increasing the risk of collisions and rollovers during driving.

[0026] In related technologies, by real-time monitoring, fault diagnosis, and early warning of the operating status of vehicle components, potential faults can be detected in a timely manner and intervention measures can be taken, thereby avoiding safety risks from the source and effectively ensuring the travel safety of drivers and passengers and the stable operation of vehicles.

[0027] However, diagnostic systems for vehicle components typically employ a single processing path, performing offline analysis only in the cloud, which can easily lead to delays in the discovery of potential faults. Furthermore, the limited computing resources on the vehicle end make it unable to perform real-time calculations for complex diagnostic models such as deep learning and time-series prediction, thus reducing the accuracy and stability of fault identification.

[0028] Based on the above issues, please refer to Figure 1 This application provides a method for fault detection of vehicle components, the method comprising: 01: Based on the first type of data of the vehicle parts, perform the first fault diagnosis process on the vehicle parts, wherein the time required for collecting the first type of data for the first fault diagnosis process is less than the first preset time. 02: Based on the second type of data of the vehicle parts, perform a second fault diagnosis process on the vehicle parts, wherein the collection time of the second type of data required for the second fault diagnosis process is longer than the first preset time.

[0029] Please see Figure 2This application provides a vehicle component fault detection system 100. The vehicle component fault detection method of this application can be implemented by the vehicle component fault detection system 100 of this application. Specifically, the vehicle component fault detection system 100 includes a vehicle-side fault diagnosis module 110, a cloud-based fault diagnosis module 120, and an in-vehicle communication module 130; the vehicle-side fault diagnosis module 110 and the cloud-based fault diagnosis module 120 establish a communication connection through the in-vehicle communication module 130; the vehicle-side fault diagnosis module 110 is configured to perform a first fault diagnosis process on the vehicle component based on a first type of data of the vehicle component, wherein the acquisition time of the first type of data required for the first fault diagnosis process is less than a first preset time; the cloud-based fault diagnosis module 120 is configured to perform a second fault diagnosis process on the vehicle component based on a second type of data of the vehicle component, wherein the acquisition time of the second type of data required for the second fault diagnosis process is greater than the first preset time.

[0030] This application also provides a vehicle, which includes a memory and a processor. The vehicle component fault detection method of this application can be implemented by the vehicle of this application. Specifically, the memory stores a computer program, and the processor is used to perform a first fault diagnosis process on the vehicle components based on a first type of data, wherein the acquisition time of the first type of data required for the first fault diagnosis process is less than a first preset time. The processor is also used to perform a second fault diagnosis process on the vehicle components based on a second type of data, wherein the acquisition time of the second type of data required for the second fault diagnosis process is greater than the first preset time.

[0031] Specifically, vehicle parts refer to components involved in vehicle operation, such as power batteries, electric drive systems, and other related components.

[0032] A fault detection system is a system used to monitor the operating status of vehicle components, collect data, and diagnose faults in order to identify and determine the faults of vehicle components.

[0033] The vehicle component fault detection system 100 includes a vehicle-side fault diagnosis module 110, a cloud-based fault diagnosis module 120, and an in-vehicle communication module 130. The vehicle-side fault diagnosis module 110 is deployed on the vehicle and is used for real-time fault analysis and processing, adapting to the diagnostic needs of real-time acquired data (i.e., the first type of data). The cloud-based fault diagnosis module 120 is deployed on a cloud server and, compared to the vehicle-side fault diagnosis module 110, has stronger computing resources and data storage capabilities, adapting to the diagnostic needs of the second type of data. The in-vehicle communication module 130 connects the vehicle-side fault diagnosis module 110 and the cloud-based fault diagnosis module 120 to achieve data interaction and command transmission between the vehicle and the cloud, laying the communication foundation for subsequent first and second fault diagnosis processing.

[0034] The first preset duration is a pre-set threshold duration used to distinguish between the first type of data and the second type of data. It can be set according to the actual needs of vehicle component fault diagnosis, such as 12 hours, 24 hours, etc.

[0035] The first type of data is short-cycle data collected from vehicle parts, with a collection time shorter than the first preset time. This data can characterize the real-time operating status of vehicle parts, such as real-time data or short-cycle data collected within 5 hours.

[0036] The first fault diagnosis process is the fault diagnosis operation performed by the vehicle-side fault diagnosis module 110 on the first type of data, which is used to achieve rapid identification of real-time faults in vehicle components.

[0037] The second type of data is longer-term data collected from vehicle parts and collected for a duration longer than the first preset duration. The second type of data has a large data volume and a long time span, which can characterize the long-term operating patterns and potential failure trends of vehicle parts, such as data collected for more than 24 hours.

[0038] The second fault diagnosis process is a fault diagnosis operation performed by the cloud-based fault diagnosis module 120 on the second type of data, which is used to accurately judge potential faults and long-term fault trends of vehicle parts.

[0039] Understandably, the vehicle-side fault diagnosis module 110 is limited by the vehicle's computing power and storage resources, making it unable to perform in-depth analysis of the second type of data, which may lead to a delay in the discovery of potential faults; while cloud processing has a delay, and the cloud-based fault diagnosis module 120 cannot achieve rapid identification of real-time faults in vehicle components.

[0040] By continuously collecting operational status data of vehicle components, and defining data with a collection duration less than a threshold as first-type data based on a first preset duration, the vehicle-side fault diagnosis module 110 can perform first-type fault diagnosis processing on the first-type data. By utilizing the localized computing power and lightweight diagnostic algorithms of the vehicle, the real-time operational status of components can be quickly analyzed, enabling rapid identification and diagnosis of real-time faults of vehicle components. This reduces the diagnostic delay of real-time faults to a certain extent and ensures the timeliness of fault handling. Furthermore, data with a collection duration exceeding the threshold is defined as the second type of data. This allows the cloud-based fault diagnosis module 120 to perform a second fault diagnosis process on the second type of data. Based on cloud computing resources and algorithms, the second type of data is deeply mined and analyzed to identify potential faults and long-term operating trends of components, thereby improving the accuracy of fault diagnosis. This enables the division of labor and collaboration between the vehicle and the cloud, ensuring the real-time performance and accuracy of fault diagnosis.

[0041] In one example, such as Figure 3 In the vehicle component fault detection system 100 shown, the vehicle-side fault diagnosis module 110 can detect the power battery in real time. The vehicle-mounted communication module 130, i.e., the whole vehicle communication system, includes a gateway, a communication module (Telematics Box, T-BOX), a Controller Area Network (CAN) communication network, etc. The vehicle-side fault diagnosis module 110 forwards the detected data to the gateway via the CAN communication network, such as the second type of data. Then, the whole vehicle communication system forwards the data on the CAN communication network to the T-BOX, which then uploads it to the cloud fault diagnosis module 120 or a mobile data communication module, such as a mobile terminal; or the T-BOX receives cloud data and forwards it to the vehicle-side communication network via the gateway.

[0042] In one example, such as Figure 3 In the vehicle component fault detection system 100 shown, the cloud monitoring platform, i.e. the cloud fault diagnosis module 120, may include functional modules such as a battery remaining useful life (RUL) prediction model, a battery state of charge (SOC) estimation model, a battery state of health (SOH) estimation model, a fault diagnosis and graded early warning model, a database management, and a service / device access module.

[0043] The cloud-based monitoring platform's service / device access module can communicate fault data, such as fault collection data, fault diagnosis data, and prediction result data, with the T-BOX in the vehicle communication module 130. The service / device access module is also responsible for standardizing the format and content of data from different sources to facilitate subsequent database management. The database management module uses the database to store and manage data uploaded to the cloud, improving data management and maintenance efficiency and facilitating the collection of relevant data for training machine learning algorithms. Models such as the Battery Lifetime RUL Prediction Model, Battery State of Charge (SOC) Prediction Model, and Battery State of Health (SOH) Prediction Model use battery voltage, current, temperature, and other information as inputs. They accurately predict various battery state indicators through time-series machine learning algorithms such as Long Short-Term Memory (LSTM) networks, and send the prediction results to the vehicle or mobile device via the service / device access module.

[0044] The vehicle-side fault diagnosis module 110 only processes the first type of data with a shorter cycle, which can avoid the limited computing and storage resources of the vehicle being occupied by the second type of data with a longer cycle, thus ensuring the efficiency of real-time diagnosis on the vehicle side. The cloud-based fault diagnosis module 120 can, for the second type of data with a longer cycle, realize the rational allocation and utilization of resources based on the massive computing resources in the cloud.

[0045] In addition, after obtaining the diagnostic results, the vehicle-side fault diagnosis module 110 can upload the diagnostic results or raw data to the cloud-side fault diagnosis module 120 through the vehicle communication module 130 to provide supplementary data for cloud-side data analysis; the cloud can feed back the diagnostic trend of the second type of data to the vehicle-side to provide a reference for the real-time diagnosis of the vehicle-side, and realize the diagnostic collaboration between the vehicle-side and the cloud.

[0046] In summary, the embodiments of this application process the first type of data through the vehicle-side fault diagnosis module 110, enabling rapid identification and diagnosis of real-time faults in vehicle components. This reduces the diagnostic delay of real-time faults to a certain extent and ensures the timeliness of fault handling. The second type of data is processed through the cloud-based fault diagnosis module 120, which, based on cloud computing resources and complex algorithms, enables precise analysis of potential component faults and long-term operational trends, improving the accuracy of fault diagnosis and achieving rational allocation and utilization of resources. Compared to a single diagnostic processing path, the embodiments of this application establish a bidirectional communication connection between the vehicle-side fault diagnosis module 110 and the cloud-based fault diagnosis module 120 through the vehicle-mounted communication module 130. This allows for collaborative work between the vehicle and cloud based on the first and second types of data, balancing the timeliness and accuracy of fault diagnosis.

[0047] Please see Figure 4In some implementations, the system further includes an external module 140; The external module 140 is configured to perform a third fault diagnosis process on vehicle components based on the first type of data when the computing resources of the vehicle-side fault diagnosis module 110 are insufficient.

[0048] Specifically, the external module 140 is a fault diagnosis module independent of the vehicle-side fault diagnosis module 110 and the cloud-based fault diagnosis module 120. It has independent computing, data processing and fault diagnosis capabilities and is used to supplement computing power when the computing resources of the vehicle-side fault diagnosis module 110 are insufficient. It can establish communication connections with the vehicle-side fault diagnosis module 110 and the cloud-based fault diagnosis module 120 through the communication module.

[0049] The third fault diagnosis process is the fault diagnosis operation performed by the external module 140 on the first type of data of vehicle parts.

[0050] When the computing resources of the vehicle-side fault diagnosis module 110 are insufficient, it can be assumed that when processing the first type of data, the computing power, storage, data processing bandwidth and other resources of the vehicle-side fault diagnosis module 110 cannot meet the real-time diagnosis requirements of the current first type of data, and there may be a state of excessive computing power, such as diagnosis delay and algorithm execution interruption.

[0051] By enabling the external module 140 to perform third-party fault diagnosis processing on vehicle components based on the first type of data when the computing resources of the vehicle-side fault diagnosis module 110 are insufficient, the fault detection coverage can be ensured for multiple scenarios, thereby improving the effectiveness of fault detection.

[0052] Compared to the vehicle-side fault diagnosis module 110, the external module 140 has more computing resources and can provide more convenient supporting services and user interface.

[0053] In one example, external module 140 can be as follows: Figure 3 The fault acquisition and analysis instrument shown has the following structure: Figure 5 As shown, it includes a controller, indicator lights, power supply, Buck circuit, wireless module, etc. The fault acquisition and analysis instrument interacts with the vehicle-side fault diagnosis module 110 via a CAN network. Figure 6 In the Buck circuit diagram shown, the circuit is based on the principle of inductor energy storage. It controls the switching state of the switch S by controlling the PWM wave with a variable duty cycle of the input to convert the DC voltage provided by the input power supply into an adjustable low voltage output, thereby meeting the power supply requirements of different circuits.

[0054] In one example, such as Figure 7In the data acquisition and processing flowchart shown, for the second type of data, such as long-cycle data, which has low real-time requirements and high computing resource requirements, the vehicle can directly upload the data through the T-BOX of the vehicle communication module 130, and the cloud vehicle monitoring platform, i.e. the cloud fault diagnosis module 120, will analyze and diagnose the second type of data. For the first type of data, such as real-time short-cycle fault data, the rapid diagnostic system, i.e., the vehicle-side fault diagnosis module 110, can perform preliminary analysis and processing. If the computing resources of the vehicle-side fault diagnosis module 110 are insufficient to diagnose the first type of data, an external fault diagnosis device, i.e., a fault acquisition and analysis instrument, can be connected for analysis and status upload.

[0055] In addition, the diagnostic results obtained by the external module 140 can also be sent to the cloud-based fault diagnosis module 120 for storage and analysis.

[0056] Thus, the system also includes an external module 140; the external module 140 is configured to perform third fault diagnosis processing on vehicle components based on the first type of data when the computing resources of the vehicle-side fault diagnosis module 110 are insufficient. In this way, the external module 140 can be used to perform third fault diagnosis processing on the first type of data when the computing resources of the vehicle-side fault diagnosis module 110 are insufficient, which can ensure the coverage of multiple fault detection scenarios and thus improve the effectiveness of fault detection.

[0057] Please refer to it again. Figure 4 In some embodiments, the system further includes an electric drive module 150, which is connected to the vehicle-side fault diagnosis module 110 and the vehicle communication module 130. The electric drive module 150 is configured as follows: Acquire first- and second-class data for vehicle components; The first type of data is transmitted to the vehicle-side fault diagnosis module 110; and / or The second type of data is transmitted to the cloud fault diagnosis module 120 via the vehicle communication module 130.

[0058] Specifically, the electric drive module 150 refers to the hardware components in the vehicle's powertrain system used for driving, energy conversion, and transmission, such as... Figure 3 The all-in-one electric drive system shown integrates functions such as motor and motor controller. It can undertake the power output task for vehicle driving and also serve as the source of vehicle status data acquisition, responsible for capturing the operating parameters of vehicle components such as battery, motor, and electronic control in real time.

[0059] The vehicle-side fault diagnosis module 110 can establish a connection with the electric drive module 150 through the vehicle communication network or internal bus to perform first diagnostic processing on the first type of data collected and transmitted by the electric drive module 150, so as to realize the rapid flow of data from the acquisition end to the processing end and ensure that the first type of data is processed efficiently and accurately at the vehicle end.

[0060] The second type of data collected and transmitted by the electric drive module 150 can also be transmitted to the cloud fault diagnosis module 120 through the vehicle communication module 130, providing a data basis for the second fault diagnosis and processing.

[0061] By integrating the electric drive module 150, which is responsible for data acquisition, with the vehicle-side fault diagnosis module 110, which is responsible for the first fault diagnosis process, the hardware layout and wiring inside the vehicle can be simplified, the overall system construction cost and electromagnetic interference risk can be reduced, and the reliability and stability of the system can be improved.

[0062] Thus, the system also includes an electric drive module 150, which is connected to the vehicle-side fault diagnosis module 110 and the vehicle-mounted communication module 130. The electric drive module 150 is configured to acquire first-type and second-type data from vehicle components; transmit the first-type data to the vehicle-side fault diagnosis module 110; and transmit the second-type data to the cloud-based fault diagnosis module 120 via the vehicle-mounted communication module 130. In this way, the vehicle-side fault diagnosis module 110 can perform first-type diagnostic processing on the first-type data collected and transmitted by the electric drive module 150, achieving rapid data flow from the acquisition end to the processing end, ensuring that the first-type data is processed efficiently and accurately at the vehicle end, and improving the reliability and stability of the system to a certain extent.

[0063] In some implementations, the cloud-based fault diagnosis module 120 is configured to send fault diagnosis information to the electric drive module 150 via the onboard communication module 130 in the event of a vehicle component failure; and / or The vehicle-side fault diagnosis module 110 is configured to send fault diagnosis information to the electric drive module 150 in the event of a fault in a vehicle component. The electric drive module 150 is configured to control vehicle components based on received fault diagnosis information.

[0064] Specifically, the fault diagnosis information is the instruction information generated by the cloud-based fault diagnosis module 120 and the vehicle-side fault diagnosis module 110 after determining that there is a fault in the vehicle parts. This information includes the fault type, fault level, fault location, and corresponding handling strategy, so that the vehicle-side electric drive module 150 can perform corresponding fault handling based on the fault diagnosis information.

[0065] The electric drive module 150 can actively adjust, limit, or intervene in the operating status of vehicle components based on the received fault diagnosis information, such as controlling battery charging and discharging, limiting motor power, cutting off fault circuits, and controlling the vehicle's power on and off, in order to prevent the occurrence of safety accidents.

[0066] Understandably, the vehicle-side fault diagnosis module 110 performs the first fault diagnosis process on the first type of data, and the cloud-side fault diagnosis module 120 performs the second fault diagnosis process on the second type of data. Both the vehicle-side and cloud-side independently complete the identification and judgment of vehicle component faults. When either end determines that a component is faulty, it immediately sends fault diagnosis information to the electric drive module 150.

[0067] In one example, after completing the first fault diagnosis process, the vehicle-side fault diagnosis module 110 can synchronize fault diagnosis information back to the electric drive module 150, such as fault level, fault code, and suggested measures, and after completing the second fault diagnosis process, the cloud-based fault diagnosis module 120 can synchronize the fault diagnosis information back to the electric drive module 150. This allows the electric drive module 150 to execute corresponding vehicle control strategies based on the diagnostic results. The electric drive module 150 can coordinate with the all-in-one electric drive system to execute corresponding fault handling commands based on the fault type and risk level. For example, it can link the power distribution box to control the charging and discharging process of the power battery, and link the vehicle energy management intelligent control system to control the energy management and distribution of the electric drive system.

[0068] In addition, after the electric drive module 150 completes the fault control operation, it will feed back the execution status and execution result of the operation to the corresponding fault diagnosis sending end in real time. At the same time, it will upload the fault diagnosis information, execution operation, execution result and other data to the cloud fault diagnosis module 120 through the vehicle communication module 130 for data filing and storage, so as to provide data support for subsequent fault analysis and algorithm optimization.

[0069] Thus, the cloud-based fault diagnosis module 120 is configured to send fault diagnosis information to the electric drive module 150 via the vehicle communication module 130 in the event of a vehicle component failure; the vehicle-side fault diagnosis module 110 is configured to send fault diagnosis information to the electric drive module 150 in the event of a vehicle component failure; and the electric drive module 150 is configured to control the vehicle components based on the received fault diagnosis information. In this way, both the cloud-based fault diagnosis module 120 and the vehicle-side fault diagnosis module 110 can send fault diagnosis information to the electric drive module 150 in the event of a vehicle component failure, enabling the electric drive module 150 to control the vehicle components based on the received fault diagnosis information, thereby preventing safety accidents and ensuring vehicle operational safety.

[0070] In some implementations, the cloud-based fault diagnosis module 120 is configured as follows: If, in the case where data in the second category exists in a state of continuous missing data within a second preset time period, the data excluding the time period in which the missing data in the second category is located is determined as the target data; and / or In cases where data in the second category is discontinuously missing, the data after imputation of the second category is identified as the target data; and / or If there are no missing data in the second type of data, then the second type of data is determined as the target data.

[0071] Specifically, the second preset duration is a pre-set time threshold range.

[0072] The second preset time period of continuous missing state refers to the continuous blank state in which the second type of data of vehicle parts has no collected values ​​and no transmission records, which may be caused by continuous network interruption or long-term sensor failure.

[0073] Discontinuous missing states refer to random missing states, that is, the second type of data missing states are characterized by scattered, irregular single points or small amounts of missing data, without forming a continuous missing period of time, which may be caused by sudden factors such as network signal fluctuations and data transmission packet loss.

[0074] The target data is the data that the cloud-based fault diagnosis module 120 uses to perform targeted missing data processing on the second type of data, and is ultimately used for subsequent feature extraction, cluster analysis, and fault determination.

[0075] By performing integrity verification and missing type determination on the second type of data, it is possible to distinguish three states of the second type of data within a second preset time period: continuous missing, non-continuous missing, and no missing. Different processing methods are used to obtain the target data based on different states, ensuring the accuracy of cloud-based second fault diagnosis and processing from the data source.

[0076] like Figure 8 As shown, for the second type of data that is continuously missing within the second preset time period, all original data within the time period containing the missing data can be directly removed, and the remaining valid and continuous data can be determined as the target data, i.e., the data processed in the first stage. This avoids the invalid time periods of continuous missing data interfering with subsequent diagnostic analysis. For example, if there is continuous missing data within the time period from 1t to 6t, all individual voltage data within the time period from t1 to t6 will be deleted, and the data in this time period will not be used. For the second type of data that is not continuously missing, an interpolation processing method adapted to the operating data characteristics of vehicle components can be used to reasonably estimate and supplement the missing values. For example, the average value can be used for interpolation processing to restore the temporal continuity and integrity of the data. The data after interpolation is determined as the target data. For the second type of data without missing data, the original second type of data can be directly determined as the target data without additional processing and can directly enter the subsequent diagnostic stage.

[0077] In one example, if the second type of data contains a continuous blank state where no valid value is found within a second preset time period, such as a continuous missing minute of data, it can be determined as continuous missing within the second preset time period; if the missing data is scattered, irregular, single-point, or a small amount of blank data, and does not reach the second preset time period of continuous missing, it is determined as non-continuous missing; if the timestamp, collection frequency, and value of the data all meet the preset standards and there is no blank state, it is determined as no missing data.

[0078] Thus, the cloud-based fault diagnosis module 120 is configured to, when data in the second type of data is continuously missing within a second preset time period, determine the data after the time period containing the missing data in the second type of data as the target data; when data in the second type of data is not continuously missing, determine the data after interpolation processing of the second type of data as the target data; and when there is no missing data in the second type of data, determine the second type of data as the target data. In this way, by classifying and preprocessing the second type of data according to its missing status, continuously missing data can be eliminated, and non-continuously missing values ​​can be accurately supplemented, avoiding diagnostic biases, misjudgments, and missed judgments caused by missing data. This improves the accuracy and reliability of the second fault diagnosis processing of the cloud-based fault diagnosis module 120 to a certain extent, enabling precise identification of potential faults and performance degradation trends of vehicle components.

[0079] In some implementations, the cloud-based fault diagnosis module 120 is configured as follows: Perform feature extraction processing on the target data to determine the target feature data; Clustering is performed on the target feature data to obtain multiple clusters and the location and target distance of each cluster. The target distance is the distance between the cluster and the target center, and the target center is the location of the cluster that is closest to the safety window among the multiple clusters. Based on the target distance, determine the fault diagnosis information of vehicle components.

[0080] Specifically, feature extraction processing is used to extract feature information that can characterize the operating status and fault characteristics of vehicle parts from the preprocessed target data, so as to transform the original data into feature data that can be used for fault identification, i.e. target feature data, and provide a data foundation for subsequent clustering processing.

[0081] Clustering is used to group target feature data with similar characteristics into one category to form multiple data clusters, which can be used to classify and identify data of unknown fault types.

[0082] A cluster refers to a set of target feature data with similar characteristics formed after clustering. Each cluster represents an operating state of a vehicle component, including normal operating state and different types of fault states.

[0083] The target center location is the center coordinate of the cluster closest to the safety window among multiple clusters. The cluster at the target center location represents the normal operating status of vehicle components and provides a benchmark reference for judging whether other clusters are in a fault state.

[0084] The safety window is a pre-defined range of characteristic data for the normal operation of vehicle components. Characteristic data within this range indicates that the component is fault-free, while data outside this range indicates a risk of failure.

[0085] Fault diagnosis information is obtained through cluster analysis and distance determination. It includes information related to faults, such as whether vehicle parts are faulty, the fault level, and the fault type. In other words, it is the result of cloud-based fault diagnosis, which provides a clear basis for the vehicle to execute targeted fault handling strategies, for technicians to analyze the causes of faults, and for optimizing diagnostic algorithms.

[0086] By performing feature extraction on the target data, target feature data can be obtained, eliminating interference from invalid data. This allows subsequent fault diagnosis to be based on the feature data. Through clustering, the target feature data can be divided into multiple clusters, determining the location of each cluster and its target distance from the target center. Then, based on the correlation between the target distance and the safety window, the fault diagnosis information of vehicle parts can be accurately determined, enabling fault result deduction, ensuring the accuracy of cloud-based fault diagnosis, and improving the ability to identify unknown fault types to a certain extent.

[0087] In one example, the intrinsic difference feature, average deviation feature, and entropy feature of the target data can be extracted separately and combined into a difference-entropy feature set, i.e., the target feature data. Among them, the intrinsic difference feature can be used to compare the extreme changes of the battery's external characteristics during a certain charge and discharge process longitudinally; the average deviation feature can be used to compare the differences between each individual cell and the average state of the system laterally to reflect the non-uniformity of the system.

[0088] In one example, such as Figure 9 In the fault diagnosis algorithm flow shown, the cloud fault diagnosis module 120 can judge the missing status and process the missing data of the received and stored second type of data, obtain the target data, perform feature extraction processing on the target data, obtain the target feature data, and then perform hierarchical clustering on the feature space of the target feature data through clustering processing.

[0089] exist Figure 10 In the hierarchical clustering diagram shown, the high-dimensional space formed by all target feature data extracted from the vehicle component operation data is the complete feature space, which is used to provide the original data input for subsequent clustering processing.

[0090] By grouping similar feature data into several large clusters, such as cluster c11 and cluster c1n in the first layer, and then further subdividing each cluster to form more specific sub-clusters, such as cluster cK1 and cluster cKm in the subsequent Kth layer, until the distance from each sample data to its cluster center is less than a pre-set distance threshold, the data can be quickly divided into normal and abnormal categories. Then, the abnormal categories are further subdivided to finally locate the specific fault type and achieve fault level classification, thus avoiding indiscriminate analysis of all data and improving diagnostic efficiency.

[0091] In this hierarchical clustering, each layer can use K-means clustering as the clustering algorithm, and the distance calculation formula is as follows:

[0092] in, It can take any value; it can be negative, positive, or infinite.

[0093] In one example, scores can be assigned based on the center position of each cluster obtained by hierarchical clustering and the target distance of each cluster. The target distance is the distance between the cluster and the cluster closest to the safety window among multiple clusters. The farther the cluster is from the safety window, the lower the score, and the closer the cluster is, the higher the score. Finally, the fault level can be classified according to the preset score range.

[0094] In one example, if the second type of data uploaded to the cloud fault diagnosis module 120 is processed for the second fault diagnosis, and the current level is determined to be high risk, then the vehicle communication module 130 needs to send information to the mobile terminal to notify the vehicle owner of the fault status. At the same time, the vehicle communication module 130 sends control commands to the vehicle through the T-BOX module, such as limiting the vehicle status through the multi-in-one electric drive, controlling the vehicle's power on / off, and controlling the charging status, to reduce vehicle safety risks. If the second type of data uploaded to the cloud fault diagnosis module 120 is processed for a second fault diagnosis, and the current level is determined to be low risk, then a warning message can be sent to the mobile terminal through the vehicle communication module 130, and a warning message can be sent to the vehicle through the T-BOX module of the vehicle communication module 130, reminding the vehicle owner to send the vehicle for inspection and repair within a certain period of time.

[0095] Thus, the cloud-based fault diagnosis module 120 is configured to perform feature extraction processing on the target data to determine the target feature data; perform clustering processing on the target feature data to obtain multiple clusters and the position and target distance of each cluster, where the target distance is the distance between the cluster and the target center position, and the target center position is the position of the cluster closest to the safety window among the multiple clusters; and determine the fault diagnosis information of the vehicle parts based on the target distance. In this way, by performing feature extraction processing on the target data to obtain the target feature data, interference from invalid data can be eliminated, allowing subsequent fault diagnosis to be carried out based on the feature data. By dividing the target feature data into multiple clusters through clustering processing, determining the position of each cluster and its target distance from the target center position, and then accurately determining the fault diagnosis information of the vehicle parts based on the correlation between the target distance and the safety window, fault result deduction can be achieved, ensuring the accuracy of cloud-based fault diagnosis and improving the ability to identify unknown fault types to a certain extent.

[0096] This application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it implements the steps of the vehicle component fault detection method described above.

[0097] This application also provides a computer program product, including a computer program / instructions. When the computer program / instructions are executed by a processor, the steps of the fault detection method for vehicle components described above can be implemented.

[0098] It is understood that a computer program includes computer program code. Computer program code can be in the form of source code, object code, executable files, or some intermediate form. Computer-readable storage media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media, etc.

[0099] In this specification, the terms "specifically," "furthermore," "particularly," "understandably," etc., refer to specific features, structures, materials, or characteristics described in connection with embodiments or examples that are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0100] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of executable request code comprising one or more steps for implementing a particular logical function or process, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0101] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A fault detection system for vehicle components, characterized in that, The system includes a vehicle-side fault diagnosis module, a cloud-based fault diagnosis module, and an in-vehicle communication module; the vehicle-side fault diagnosis module and the cloud-based fault diagnosis module establish a communication connection through the in-vehicle communication module. The vehicle-side fault diagnosis module is configured to perform a first fault diagnosis process on the vehicle components based on a first type of data of the vehicle components, wherein the collection time of the first type of data required for the first fault diagnosis process is less than a first preset time. The cloud-based fault diagnosis module is configured to perform a second fault diagnosis process on the vehicle components based on the second type of data of the vehicle components, wherein the collection time of the second type of data required for the second fault diagnosis process is longer than the first preset time.

2. The system according to claim 1, characterized in that, The system also includes an external module; The external module is configured to perform a third fault diagnosis process on the vehicle components based on the first type of data when the computing resources of the vehicle-side fault diagnosis module are insufficient.

3. The system according to claim 1, characterized in that, The system also includes an electric drive module, which is connected to the vehicle-side fault diagnosis module and the vehicle-mounted communication module. The electric drive module is configured as follows: Acquire the first type of data and the second type of data of the vehicle components; The first type of data is transmitted to the vehicle-side fault diagnosis module; and / or The second type of data is transmitted to the cloud-based fault diagnosis module via the vehicle communication module. The vehicle-side fault diagnosis module is configured to perform the first fault diagnosis process on the vehicle components based on the first type of data.

4. The system according to claim 3, characterized in that, The cloud-based fault diagnosis module is configured to send fault diagnosis information to the electric drive module via the vehicle communication module in the event of a fault in the vehicle component; and / or The vehicle-side fault diagnosis module is configured to send fault diagnosis information to the electric drive module in the event of a fault in the vehicle component. The electric drive module is configured to control the vehicle components based on the received fault diagnosis information.

5. The system according to claim 3, characterized in that, The cloud-based fault diagnosis module is configured as follows: If, in the case where data in the second type of data is continuously missing within a second preset time period, the data excluding the time period containing the missing data in the second type of data is determined as the target data; and / or In the case where there are non-continuous missing data in the second type of data, the data after imputation of the second type of data is determined to be the target data. and / or If there are no missing data in the second type of data, then the second type of data is determined to be the target data.

6. The system according to claim 5, characterized in that, The cloud-based fault diagnosis module is configured as follows: The target data is subjected to feature extraction processing to determine the target feature data; The target feature data is clustered to obtain multiple clusters and the position and target distance of each cluster, wherein the target distance is the distance between the cluster and the target center position, and the target center position is the position of the cluster closest to the safety window among the multiple clusters; Based on the target distance, fault diagnosis information for the vehicle components is determined.

7. A method for fault detection of vehicle components, characterized in that, The fault detection method is used in a fault detection system for vehicle components, and the method includes: Based on the first type of data of the vehicle parts, a first fault diagnosis process is performed on the vehicle parts, wherein the collection time of the first type of data required for the first fault diagnosis process is less than a first preset time. Based on the second type of data of the vehicle components, a second fault diagnosis process is performed on the vehicle components, wherein the time required for collecting the second type of data for the second fault diagnosis process is longer than the first preset time.

8. A vehicle, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, which, when executed by the processor, implements the method of claim 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by one or more processors, implements the method of claim 7.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method of claim 7.

Images