Fault processing method and device for intelligent connected automobile cloud platform

Abstract

The invention discloses a fault processing method and device for an intelligent network connection automobile cloud platform, and the method comprises the steps: obtaining the construction operation and maintenance knowledge and platform architecture diagram of the intelligent network connection automobile cloud platform, and the terminal data of a service terminal, and carrying out the fine adjustment of a visual language model, and obtaining a platform customization model; using the platform customization model to output fault result information based on the monitoring data of the service terminal; determining target fault matching data including a repetition rate between the platform fault information and pre-configured fault result information; and determining a target fault processing rule based on the target fault matching data and a corresponding relationship between a preset fault processing rule and the fault matching data, thereby performing fault processing. According to the method, fault reasoning is carried out on the monitoring data of the service terminal of the intelligent network connection automobile cloud platform through the constructed platform customization model for the intelligent network connection automobile cloud platform, the maintenance and operation efficiency of the special vehicle is improved, and the safety and reliability of the special vehicle are ensured.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and particularly to a method and device for fault handling of an intelligent connected vehicle cloud platform. Background Art

[0002] Intelligent and Connected Vehicles (ICVs) refer to vehicles that integrate advanced on-vehicle sensors, controllers, actuators and other devices, and fuse communication and network technologies to achieve intelligent information exchange and sharing between vehicle and vehicle, vehicle and road, vehicle and person, vehicle and cloud, etc. The core technologies of intelligent and connected vehicles include autonomous driving technology and vehicle networking technology. Vehicle networking technology realizes real-time information exchange between vehicles and other vehicles, infrastructure, and cloud servers through advanced communication technologies.

[0003] With the continuous progress and application of technology, the fault maintenance of intelligent connected vehicle cloud platforms has become a field of great concern. Currently, existing third-party operation and maintenance systems are usually used to assist in the fault diagnosis of vehicle cloud platforms. However, the adaptability between general operation and maintenance systems and each vehicle cloud platform is not the same, resulting in unsatisfactory auxiliary diagnosis effects and high fees, leading to increased operation and maintenance costs. Summary of the Invention

[0004] To solve the above problems, embodiments of the present application provide a method, device, electronic device, computer-readable storage medium, and computer program product for fault handling of an intelligent connected vehicle cloud platform.

[0005] In a first aspect, to solve the above technical problems, the present application provides a method for fault handling of an intelligent connected vehicle cloud platform, including:

[0006] Obtain the construction operation and maintenance knowledge and platform architecture diagram of the intelligent connected vehicle cloud platform, as well as the terminal data of the service terminal of the intelligent connected vehicle cloud platform;

[0007] Fine-tune a vision-language model based on the construction operation and maintenance knowledge, the platform architecture diagram, and the terminal data to obtain a platform customization model of the intelligent connected vehicle cloud platform;

[0008] Obtain the monitoring data of the service terminal, and use the platform customization model to output the fault result information of the service terminal based on the monitoring data;

[0009] Based on pre-configured platform fault information, determine target fault matching data, where the target fault matching data includes the repetition rate between the platform fault information and the fault result information;

[0010] Based on the target fault matching data and the corresponding relationship between the preset fault handling rules and the fault matching data, determine the target fault handling rules, so as to perform fault handling on the intelligent connected vehicle cloud platform based on the target fault handling rules.

[0011] The beneficial effects are:

[0012] In the technical solution provided by the embodiments of the present application, by obtaining the construction and operation and maintenance knowledge and platform architecture diagram of the intelligent connected vehicle cloud platform, as well as the terminal data of the service terminals of the intelligent connected vehicle cloud platform; fine-tuning the vision language model based on the construction and operation and maintenance knowledge, platform architecture diagram and terminal data to obtain a platform customization model for the intelligent connected vehicle cloud platform. In this way, through the language vision model combined with the construction and operation and maintenance knowledge, platform architecture diagram and relevant terminal data of the intelligent connected vehicle cloud platform, a platform customization model customized for platform maintenance personnel is obtained, which is more suitable for platform requirements, and the customized platform customization model is convenient for secondary development, reducing the maintenance difficulty and cost of the platform customization model. When applying the platform customization model later, obtain the monitoring data of the service terminal, and use the platform customization model to output the fault result information of the service terminal based on the monitoring data; based on the pre-configured platform fault information, determine the target fault matching data, and the target fault matching data includes the repetition rate between the platform fault information and the fault result information; based on the target fault matching data and the corresponding relationship between the preset fault handling rules and the fault matching data, determine the target fault handling rules, so as to perform fault handling on the intelligent connected vehicle cloud platform based on the target fault handling rules. In this way, by using the large model as the platform customization model to infer fault problems, it can help maintenance personnel better troubleshoot faults, contribute to improving the maintenance and operation efficiency of special vehicles, and ensure their safety and reliability.

[0013] In a second aspect, the present invention provides a fault handling device for an intelligent connected vehicle cloud platform, including a collection unit, a model construction unit, a processing unit, a determination unit and an execution unit;

[0014] The collection unit is used to obtain the construction and operation and maintenance knowledge and platform architecture diagram of the intelligent connected vehicle cloud platform, as well as the terminal data of the service terminals of the intelligent connected vehicle cloud platform;

[0015] The model construction unit is used to fine-tune the vision language model based on the construction and operation and maintenance knowledge, the platform architecture diagram and the terminal data to obtain a platform customization model for the intelligent connected vehicle cloud platform;

[0016] The processing unit is used to obtain the monitoring data of the service terminal and use the platform customization model to output the fault result information of the service terminal based on the monitoring data;

[0017] A determination unit, configured to determine target fault matching data based on pre-configured platform fault information, where the target fault matching data includes the repetition rate between the platform fault information and the fault result information;

[0018] An execution unit, configured to determine a target fault handling rule based on the target fault matching data and the correspondence between a preset fault handling rule and the fault matching data, so as to perform fault handling on the intelligent connected vehicle cloud platform based on the target fault handling rule.

[0019] In a third aspect, the present application further provides an electronic device, including: one or more processors; a storage device, configured to store one or more programs, and when the one or more programs are executed by the one or more processors, enable the electronic device to implement the above-mentioned fault handling method of the intelligent connected vehicle cloud platform.

[0020] In a fourth aspect, the present application further provides a computer-readable storage medium, on which computer-readable instructions are stored, and when the computer-readable instructions are executed by a processor of a computer, enable the computer to execute the above-mentioned fault handling method of the intelligent connected vehicle cloud platform.

[0021] In a fifth aspect, the present application further provides a computer program product or a computer program, where the computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the fault handling method of the intelligent connected vehicle cloud platform provided in the above various alternative embodiments.

[0022] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application. Obviously, the accompanying drawings in the following description are only some embodiments of the present application, and those of ordinary skill in the art can obtain other drawings according to these drawings without creative efforts. In the drawings:

[0024] Figure 1 is a flowchart of a fault handling method for an intelligent connected vehicle cloud platform shown in an exemplary embodiment of the present application;

[0025] Figure 2 is a block diagram of a fault handling device for an intelligent connected vehicle cloud platform shown in an exemplary embodiment of the present application;

[0026] Figure 3 It is a schematic structural diagram of a computer system of an electronic device suitable for implementing the embodiments of the present application. Detailed implementation manners

[0027] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0028] The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0029] The flowcharts shown in the drawings are only exemplary descriptions and do not necessarily include all contents and operations / steps, nor do they necessarily need to be executed in the described order. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined. Therefore, the actual execution order may change according to the actual situation.

[0030] In the present application, "a plurality of" refers to two or more. "And / or" describes the association relationship of associated objects and indicates that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0031] In the present application, all actions of obtaining signals, information, or data are carried out on the basis of strictly following the relevant data protection regulations and policies of the country where it is located and with the authorization of the owner of the corresponding device.

[0032] The owner refers to an individual or entity that owns or controls the relevant device (which may be a device, a system, or other tools that can collect data).

[0033] In the field of intelligent connected vehicles, the "owner" mainly includes:

[0034] (1) Automobile manufacturers: As developers of vehicle hardware and systems, they control the vehicle's underlying hardware and software platforms and have management and control rights over the data generated during vehicle operation, such as driving and fault data.

[0035] (2) Component suppliers: Provide key components for automobiles and have certain ownership of the data collected and processed from the components, which are used for product optimization and after-sales, such as data generated by sensors and chips.

[0036] (3) Vehicle owners or users: The actual users of the vehicle, who have the right to decide on the usage method and scope of vehicle data, such as whether to share data on driving trajectories, driving habits, etc., and have the need and right to protect the privacy of their own relevant data.

[0037] (4) Service providers: Provide services such as software and data analysis, and have the right to use and manage the data obtained and processed under the framework of the agreement, but the ownership usually belongs to other entities.

[0038] To solve the problem that the adaptability between the general operation and maintenance system and each automotive cloud platform is not the same, resulting in poor auxiliary diagnosis effect, the embodiments of this application propose a fault handling method, device, electronic device, and computer-readable storage medium for an intelligent connected vehicle cloud platform, which mainly involve the fault handling technology of the intelligent connected vehicle cloud platform included in cloud computing technology. The following will elaborate on these embodiments in detail.

[0039] First, please refer to Figure 1 , Figure 1 which is a flowchart of the fault handling method for an intelligent connected vehicle cloud platform shown in an exemplary embodiment of this application. This method can be specifically executed by a server. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, Content Delivery Network (CDN), as well as big data and artificial intelligence platforms. There is no limitation here.

[0040] As Figure 1 shown, in an exemplary embodiment, the fault handling method for the intelligent connected vehicle cloud platform may include steps S101 to S105, which are introduced in detail as follows:

[0041] Step S101: Obtain the construction and operation and maintenance knowledge and platform architecture diagram of the intelligent connected vehicle cloud platform, as well as the terminal data of the service terminals of the intelligent connected vehicle cloud platform.

[0042] Step S102: Fine-tune the vision language model based on the construction and operation and maintenance knowledge, platform architecture diagram, and terminal data to obtain the platform customization model of the intelligent connected vehicle cloud platform.

[0043] Step S103: Obtain the monitoring data of the service terminals, and use the platform customization model to output the fault result information of the service terminals based on the monitoring data.

[0044] Step S104: Based on the pre-configured platform fault information, determine the target fault matching data, where the target fault matching data includes the repetition rate between the platform fault information and the fault result information.

[0045] Step S105: Based on the target fault matching data and the corresponding relationship between the preset fault handling rules and the fault matching data, determine the target fault handling rules, so as to perform fault handling on the intelligent connected vehicle cloud platform based on the target fault handling rules.

[0046] As can be seen from the above, in the method provided in this embodiment, by obtaining the construction and operation and maintenance knowledge of the intelligent connected vehicle cloud platform, the platform architecture diagram, and the terminal data of the service terminal of the intelligent connected vehicle cloud platform; and fine-tuning the vision language model based on the construction and operation and maintenance knowledge, the platform architecture diagram, and the terminal data, a platform customization model of the intelligent connected vehicle cloud platform is obtained. In this way, on the one hand, through the language vision model combined with the construction and operation and maintenance knowledge, the platform architecture diagram, and the relevant terminal data of the intelligent connected vehicle cloud platform, a platform customization model customized for platform maintenance personnel is obtained, providing comprehensive, real-time, and personalized knowledge support for maintenance personnel and being more adaptable to platform requirements; on the other hand, the customized platform customization model is convenient for secondary development, reducing the maintenance difficulty and cost of the platform customization model.

[0047] After applying the platform customization model, by obtaining the monitoring data of the service terminal, using the platform customization model to output the fault result information of the service terminal based on the monitoring data; obtaining the repetition rate between the pre-configured platform fault information and the fault result information, determining the target fault matching data, and determining the target fault handling rules based on the corresponding relationship between the repetition rate included in the preset fault matching data and the fault handling rules and the target fault matching data, and then improving the fault handling efficiency of the intelligent connected vehicle cloud platform by executing the target fault handling rules. Therefore, in this application, by using the large model as the platform customization model to infer fault problems, it can help maintenance personnel better troubleshoot faults, contribute to improving the maintenance and operation efficiency of special vehicles, and ensure their safety and reliability.

[0048] In an exemplary embodiment provided by this application, the specific steps of obtaining the platform customization model of the intelligent connected vehicle cloud platform by combining the construction and operation and maintenance knowledge, the platform architecture diagram, the terminal data, and the vision language model may include:

[0049] Preprocess the construction and operation and maintenance knowledge, the platform architecture diagram, and the terminal data to obtain key-value pair data;

[0050] Convert the key-value data pairs into fine-tuning data in json format;

[0051] Fine-tune the vision-language model based on the fine-tuning data to obtain an initial customized model, where the vision-language model is the deepseek-vl2 large model;

[0052] Privately deploy the initial customized model using an open-source local large model running framework to obtain a platform customized model.

[0053] In this embodiment, a platform customized model for the intelligent connected vehicle cloud platform is constructed by fine-tuning the vision-language model. The data used for fine-tuning is key-value pair data obtained by preprocessing construction and operation and maintenance knowledge, platform architecture diagrams, and terminal data. Preferably, the vision-language model is the deepseek-vl2 large model. Of course, the vision-language model can also be other vision-language models such as Qwen2.5-VL.

[0054] Preferably, convert the key-value data pairs into json format fine-tuning data. At the same time, use the fine-tuning data and the deepseek-vl2 large model as the input for fine-tuning, and then use the huggingface transformers (an open-source pre-trained language library) toolkit for fine-tuning. After fine-tuning, output a new large model file, which is the initial customized model defined in this embodiment. Finally, privately deploy the initial customized model using an open-source local large model running framework to obtain a platform customized model for the intelligent connected vehicle cloud platform.

[0055] In another exemplary embodiment, while privately deploying the initial customized model, a retrieval system is also deployed for knowledge retrieval. The specific steps may include:

[0056] Deploy the retrieval system using an open-source local large model running framework;

[0057] Obtain the recorded data generated during the process of troubleshooting the intelligent connected vehicle cloud platform, and store the recorded data using the retrieval system.

[0058] In this embodiment, the retrieval system can be a RAG (Retrieval-augmented Generation) system. After the private deployment is completed, it is used to store the recorded data formed after executing the target troubleshooting rules, form knowledge in the retrieval system, update the content of the pre-configured platform fault information, and can be directly retrieved and called, facilitating the quick handling of subsequent similar fault matters. The troubleshooting efficiency is improved.

[0059] In this way, through the above embodiments, the present application constructs a platform customization model specifically customized for platform maintenance personnel, providing comprehensive, real-time, and personalized knowledge support for maintenance personnel, better adapting to the platform requirements, and avoiding the problems of unsatisfactory auxiliary diagnosis effect and high fees, resulting in increased operation and maintenance costs.

[0060] In another exemplary embodiment provided by the present application, the specific steps for preprocessing the construction operation and maintenance knowledge, platform architecture diagram, and terminal data to obtain key-value pair data may include:

[0061] Extract key information from the construction operation and maintenance knowledge through a preset language large model to obtain the first key-value pair. The first key-value pair is in the key-value pair form of question and answer, and the language large model is the deepseek-r1 model;

[0062] Obtain the second key-value pair based on the platform architecture diagram. The second key-value pair is in the key-value pair form of scene name and image;

[0063] Obtain the number and metadata of the service terminal, and obtain the third key-value pair based on the number and metadata. The third key-value pair is in the key-value pair form of number and metadata. The service terminal includes vehicle terminals, road terminals, and the cloud.

[0064] In this embodiment, the preprocessing of the construction operation and maintenance knowledge is preferably to organize the key information in the construction operation and maintenance knowledge into the key-value pair form of question + answer through the language large model deepseek-r1 to obtain the first key-value pair. The construction operation and maintenance knowledge includes platform construction documents and operation and maintenance documents.

[0065] The preprocessing of the platform architecture diagram is preferably to form key-value pairs of the scene names and graphics of each scene topology diagram to obtain the second key-value pair, and the component pictures and component pictures in each scene topology can also form key-value pairs. The service terminal includes vehicle terminals, road terminals, and the cloud, and the corresponding platform architecture diagram may include the network topology diagrams of the cloud and road terminals; the data interaction topology diagrams of vehicle terminals with the cloud and road terminals.

[0066] The preprocessing of the terminal data is preferably to obtain the number and metadata of the service terminal, and obtain the third key-value pair based on the number and metadata. The third key-value pair is in the key-value pair form of the number and metadata. Among them, the number is defined in the construction stage of the intelligent connected vehicle cloud platform and represents the unique identifier of the service terminal in the intelligent connected vehicle cloud platform. In addition, the metadata of the vehicle terminal can be OBU (On-Board Unit), vehicle VIN code (Vehicle Identification Number), license plate number, IP address of the OBU, etc.; the metadata of the road terminal RSU (Roadside Unit) can be intersection or section name, longitude and latitude, IP address, etc.; the metadata of the cloud can be the server name of the cloud server, IDC name (data center name), IP address, etc. Under different service terminals, all the metadata of each device are concatenated together, and form a key-value pair form of number + metadata with the number to obtain the third key-value pair.

[0067] In this way, through the above embodiments, the present application uses the key-value pair data obtained by preprocessing the above-mentioned construction and operation and maintenance knowledge, platform architecture diagram, and terminal data as the basis for fine-tuning the vision-language model, so that the platform customization model can provide comprehensive, real-time, and personalized knowledge support for maintenance personnel to help maintenance personnel better troubleshoot faults.

[0068] In an exemplary embodiment provided by the present application, after the platform customization model is constructed, the specific steps of obtaining the monitoring data of the service terminal and using the platform customization model to output the fault result information of the service terminal based on the monitoring data may include:

[0069] Obtain the monitoring data of the service terminal and configure the threshold conditions corresponding to the monitoring data;

[0070] When the monitoring data of at least two terminals in the service terminal meet the corresponding threshold conditions, select the monitoring data that meets the corresponding threshold conditions as the fault monitoring data;

[0071] Use the platform customization model to output the corresponding fault result information based on the fault monitoring data.

[0072] In the embodiment provided by the present application, if the monitoring data meets the corresponding threshold conditions, it is considered that the service terminal corresponding to the monitoring data has a fault condition. In another exemplary embodiment, the service terminal includes a vehicle terminal, a road terminal, and a cloud. The specific steps of obtaining the monitoring data of the service terminal and configuring the threshold conditions corresponding to the monitoring data may include:

[0073] Obtain the vehicle device status data of the vehicle terminal and the roadside device status data of the road terminal;

[0074] Obtain the server status data, platform application status data, database status data, and middleware status data in the cloud, and add tag information to obtain the cloud monitoring items in the cloud. The tag information includes the number and metadata of the cloud.

[0075] Use an open-source monitoring system to form time-series data based on the cloud monitoring data.

[0076] Use a neural network to obtain the predicted monitoring values corresponding to the cloud monitoring data based on the time-series data, and send the predicted monitoring values back to the open-source monitoring system to obtain new cloud monitoring items.

[0077] Configure the threshold conditions corresponding to the vehicle device status data, roadside device status data, and new cloud monitoring items.

[0078] In this embodiment, the vehicle device status data (such as the online status of the OBU) is obtained through the vehicle management platform on the cloud, the roadside device status data (such as the temperature inside the floor box / pole box) is obtained through the roadside management platform, and the server status data (such as the CPU load), platform application status data (such as the api interface call status code), database status data (the read and write frequencies of the recorded mysql), and middleware status data (such as the backlog situation of the queue) are obtained through the cloud monitoring terminal.

[0079] After that, further process the data obtained from the cloud. Specifically, it can include adding tag information during collection. The tag information includes the number and metadata of the cloud to form the cloud monitoring items in the cloud. Then, integrate the cloud monitoring items in the cloud into the cloud monitoring Prometheus (open-source monitoring system) service to form time-series data. Send the time-series data into the LSTM neural network (Long Short-Term Memory), predict the future situation to obtain the predicted monitoring values, and send them back to Prometheus to generate new cloud monitoring items and new time-series data.

[0080] Finally, configure the threshold conditions corresponding to the vehicle device status data, roadside device status data, and new cloud monitoring items, such as the threshold temperature for determining that the temperature of the pole box is too high, the threshold number of times for determining that the mysql read and write frequency is too high, etc.

[0081] In addition, while configuring the threshold conditions corresponding to the monitoring data, platform fault information will also be configured. The configured platform fault information is common problems of the intelligent connected vehicle cloud platform, and the content can include: the monitoring data and alarm data involved (alarm notifier, alarm information), problem description, problem handling script. In this way, when it is identified that there is a duplication between the fault result information and the configured platform fault information, the problem handling script and other information of the overlapping platform fault information can be directly analyzed to handle the faults in this part.

[0082] In this way, the present application, through the above-mentioned embodiments, configures targeted threshold conditions for all monitoring data to accurately identify whether there is a platform failure and handle it in a timely manner.

[0083] In another exemplary embodiment provided by the present application, the specific steps of using the platform customization model to output corresponding fault result information based on fault monitoring data may include:

[0084] Call the conversation interface of the open source local big model running framework, input the conversation data and problem data corresponding to the fault monitoring data into the platform customized model, and the platform customized model is deployed in the open source local big model running framework;

[0085] Use the platform-customized model to obtain the fault scenario topology map based on the conversation data;

[0086] Using the platform customized model, the root cause terminal is obtained based on the problem data and the fault scenario architecture diagram through image reasoning method;

[0087] Use the platform customized model to call construction and operation knowledge to determine the fault cause and repair suggestions corresponding to the fault monitoring data;

[0088] Obtain fault result information based on the fault scenario topology diagram, root cause terminal, fault cause and repair suggestions.

[0089] In this embodiment, the dialogue interface of the open source local large model running framework called is the API interface exposed by the open source local large model running framework, and the interface request body is JSON data, which is used to simulate the dialogue and chat with the platform customized model.

[0090] After calling the dialogue interface, the problem data and dialogue data are automatically sent to the deepseek-vl2 large model. The dialogue data sent can be the monitoring name, label, monitoring value, alarm information, platform topology map, etc. The problem data can be the location scenario of the current fault, the root cause service or device (which one should be found to be the root cause of multiple alarms), the cause of the problem of the device, and the repair suggestions.

[0091] The device metadata learned by the platform customized model during the fine-tuning process can correspond one-to-one with the received monitoring data. Therefore, the platform customized model can be used to obtain the device metadata of the device under the corresponding service terminal based on the monitoring data in the conversation data, thereby locating the scene topology map of the device metadata learned during the fine-tuning process.

[0092] Afterwards, through the image inference method of the platform customized model, analyze the location scenario in the current fault and the location in the overall platform architecture diagram based on the obtained scenario topology diagram, so as to deduce the root cause terminal in the scenario topology diagram, such as the service or device under the service terminal. At the same time, give the fault cause and repair suggestions for this problem based on the general or fine-tuned platform construction and operation and maintenance knowledge learned by the platform customized model, and then obtain the fault result information including the fault scenario topology diagram, root cause terminal, fault cause and repair suggestions.

[0093] In this way, through the above embodiments, the present application can quickly obtain the fault result information corresponding to the platform fault, which helps the maintenance personnel to better troubleshoot the fault and improves the maintenance and operation efficiency of special vehicles.

[0094] In another exemplary embodiment, first, the alarm data will be configured synchronously when configuring the threshold conditions. After obtaining the fault result information including the fault cause and repair suggestions, feedback and announcements will also be made based on the alarm data. Specifically, after obtaining the monitoring data of the service terminal and configuring the threshold conditions corresponding to the monitoring data, the method further includes:

[0095] Configure the alarm data corresponding to the monitoring data, and the alarm data includes the alarm notifier;

[0096] When obtaining the fault result information output by the platform customized model, send the fault cause and repair suggestions included in the fault result information to the corresponding alarm notifier.

[0097] In this embodiment, the alarm data includes alarm notifications, alarm notifiers, and alarm information (such as high server CPU load, a large number of 504 api interface status codes, too high temperature in the pole box, too high mysql read and write frequency).

[0098] In this way, through the above embodiments, the present application can promptly inform the fault cause and repair suggestions, thereby improving the platform maintenance efficiency.

[0099] In an exemplary embodiment provided by the present application, based on the target fault matching data and the corresponding relationship between the preset fault handling rules and the fault matching data, the specific steps for determining the target fault handling rules may include:

[0100] If the repetition rate included in the target fault matching data indicates that the fault result information completely repeats the platform fault information, then based on the corresponding relationship between the preset fault handling rules and the fault matching data, determine that the target fault handling rule corresponding to the fault result information is the rule for automatically executing the script;

[0101] If the repetition rate indicates that the fault result information partially overlaps with the platform fault information, then based on the corresponding relationship, the target fault handling rule corresponding to the fault result information is determined as the rule of manual verification and script execution;

[0102] If the repetition rate indicates that the fault result information does not overlap with the platform fault information at all, then based on the corresponding relationship, the target fault handling rule corresponding to the fault result information is determined as the rule of manual handling.

[0103] In this embodiment, the pre-configured platform fault information includes problem handling scripts and alarm data. The alarm data includes alarm notifiers and alarm information. Therefore, when the repetition rate indicates that the fault result information completely or partially overlaps with the platform fault information, while processing the overlapping part of the fault through the problem handling scripts included in the platform fault information, the alarm information will also be notified to the corresponding alarm notifiers.

[0104] Figure 2 It is a block diagram of a fault handling device 200 for an intelligent connected vehicle cloud platform shown in an exemplary embodiment of the present application. As Figure 2 shown, the device includes:

[0105] An acquisition unit 201, configured to obtain the construction and operation and maintenance knowledge of the intelligent connected vehicle cloud platform, the platform architecture diagram, and the terminal data of the service terminals of the intelligent connected vehicle cloud platform;

[0106] A model construction unit 202, configured to fine-tune the vision language model based on the construction and operation and maintenance knowledge, the platform architecture diagram, and the terminal data to obtain a platform customized model for the intelligent connected vehicle cloud platform;

[0107] A processing unit 203, configured to obtain the monitoring data of the service terminals, and use the platform customized model to output the fault result information of the service terminals based on the monitoring data;

[0108] A determination unit 204, configured to determine target fault matching data based on the pre-configured platform fault information, where the target fault matching data includes the repetition rate between the platform fault information and the fault result information;

[0109] An execution unit 205, configured to determine a target fault handling rule based on the target fault matching data and the corresponding relationship between the preset fault handling rules and the fault matching data, so as to perform fault handling on the intelligent connected vehicle cloud platform based on the target fault handling rule.

[0110] This device applies the fault handling method of the intelligent connected vehicle cloud platform provided by this application. The acquisition unit 201 obtains the construction and operation and maintenance knowledge, platform architecture diagram of the intelligent connected vehicle cloud platform, and the terminal data of the service terminals of the intelligent connected vehicle cloud platform. The model construction unit 202 fine-tunes the vision-language model based on the construction and operation and maintenance knowledge, platform architecture diagram, and terminal data to obtain a platform customized model for the intelligent connected vehicle cloud platform. In this way, by combining the language vision model with the construction and operation and maintenance knowledge, platform architecture diagram, and relevant terminal data of the intelligent connected vehicle cloud platform, a platform customized model customized for platform maintenance personnel is obtained, which is more suitable for platform requirements. Moreover, the customized platform customized model is convenient for secondary development, reducing the maintenance difficulty and cost of the platform customized model.

[0111] When applying this platform customized model later, the processing unit 203 obtains the monitoring data of the service terminal, and the platform customized model outputs the fault result information of the service terminal based on the monitoring data. The determination unit 204 obtains the repetition rate between the pre-configured platform fault information and the fault result information, determines the target fault matching data, and the execution unit 205 determines the target fault handling rule based on the target fault matching data and the corresponding relationship between the preset fault handling rule and the fault matching data, and executes the target fault handling rule. In this way, inferring fault problems through the large model that is the platform customized model can help maintenance personnel better troubleshoot faults, contribute to improving the maintenance and operation efficiency of special vehicles, and ensure their safety and reliability.

[0112] In another exemplary embodiment, the model construction unit 202 is further configured to preprocess the construction and operation and maintenance knowledge, platform architecture diagram, and terminal data to obtain key-value pair data; convert the format of the key-value data pair to obtain fine-tuning data in json format; fine-tune the vision-language model based on the fine-tuning data to obtain an initial customized model, and the vision-language model is the deepseek-vl2 large model; privately deploy the initial customized model using an open-source local large model operation framework to obtain a platform customized model.

[0113] In another exemplary embodiment, the model construction unit 202 is further configured to deploy a retrieval system using an open-source local large model operation framework; obtain the recorded data generated during the process of handling faults for the intelligent connected vehicle cloud platform, and store the recorded data using the retrieval system.

[0114] In another exemplary embodiment, the model construction unit 202 is further configured to extract key information from the construction and operation and maintenance knowledge through a preset language large model to obtain a first key-value pair. The first key-value pair is in the form of a key-value pair of a question and an answer, and the language large model is the deepseek-r1 model; obtain a second key-value pair based on the platform architecture diagram. The second key-value pair is in the form of a key-value pair of a scenario name and an image; obtain the number and metadata of the service terminal, and obtain a third key-value pair based on the number and metadata. The third key-value pair is in the form of a key-value pair of the number and metadata. The service terminal includes a vehicle terminal, a road terminal, and a cloud.

[0115] In another exemplary embodiment, the processing unit 203 is further configured to obtain the monitoring data of the service terminal and configure the threshold conditions corresponding to the monitoring data; when the monitoring data of at least two terminals in the service terminal meet the corresponding threshold conditions, select the monitoring data that meets the corresponding threshold conditions as the fault monitoring data; use the platform customized model to output the corresponding fault result information based on the fault monitoring data.

[0116] In another exemplary embodiment, the service terminal includes a vehicle terminal, a road terminal, and a cloud; the processing unit 203 is further configured to obtain the vehicle device status data of the vehicle terminal and the roadside device status data of the road terminal; obtain the server status data, platform application status data, database status data, and middleware status data of the cloud and add tag information to obtain the cloud monitoring items of the cloud. The tag information includes the number and metadata of the cloud; use the open source monitoring system to form time series data based on the cloud monitoring data; use the neural network to obtain the predicted monitoring value corresponding to the cloud monitoring data based on the time series data, and send the predicted monitoring value back to the open source monitoring system to obtain new cloud monitoring items; configure the threshold conditions corresponding to the vehicle device status data, roadside device status data, and new cloud monitoring items.

[0117] In another exemplary embodiment, the processing unit 203 is further configured to call the dialogue interface of the open source local large model operation framework, input the dialogue data and question data corresponding to the fault monitoring data into the platform customized model, and the platform customized model is deployed in the open source local large model operation framework; use the platform customized model to obtain the fault scenario topology diagram based on the dialogue data; use the platform customized model to obtain the root cause terminal based on the question data and the fault scenario structure diagram through an image reasoning method; use the platform customized model to call the construction and operation and maintenance knowledge to determine the fault cause and repair suggestion corresponding to the fault monitoring data; obtain the fault result information based on the fault scenario topology diagram, root cause terminal, fault cause, and repair suggestion.

[0118] In another exemplary embodiment, the processing unit 203 is further configured to configure the alarm data corresponding to the monitoring data, where the alarm data includes the alarm notifier; and when obtaining the fault result information output by the platform customization model, send the fault cause and repair suggestion included in the fault result information to the corresponding alarm notifier.

[0119] In another exemplary embodiment, the execution unit 205 is further configured to: if the repetition rate included in the target fault matching data indicates that the fault result information is exactly the same as the platform fault information, determine that the target fault processing rule corresponding to the fault result information is the rule for automatically executing the script based on the corresponding relationship between the preset fault processing rule and the fault matching data; if the repetition rate indicates that the fault result information is partially the same as the platform fault information, determine that the target fault processing rule corresponding to the fault result information is the rule for manual verification and script execution based on the corresponding relationship; if the repetition rate indicates that the fault result information is completely different from the platform fault information based on the corresponding relationship, determine that the target fault processing rule corresponding to the fault result information is the rule for manual processing.

[0120] It should be noted that the fault processing device of the intelligent connected vehicle cloud platform provided in the above embodiments belongs to the same concept as the fault processing method of the intelligent connected vehicle cloud platform provided in the above embodiments. The specific manners in which each module and unit perform operations have been described in detail in the method embodiments and will not be elaborated here. In practical applications, the fault processing device of the intelligent connected vehicle cloud platform provided in the above embodiments can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here either.

[0121] An embodiment of the present application further provides an electronic device, including: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, enabling the electronic device to implement the fault processing method of the intelligent connected vehicle cloud platform provided in each of the above embodiments.

[0122] Figure 3 The structural diagram of a computer system of an electronic device suitable for implementing the embodiments of the present application is shown. It should be noted that Figure 3 The computer system 300 of the electronic device shown is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present application.

[0123] Such as Figure 3As shown, the computer system 300 includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the Read-Only Memory (ROM) 302 or the program loaded from the storage section 308 into the Random Access Memory (RAM) 303, such as executing the methods in the above embodiments. In the RAM 303, various programs and data required for system operations are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0124] The following components are connected to the I / O interface 305: an input section 306 including a keyboard, a mouse, etc.; an output section 307 including, for example, a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), etc., and a speaker, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read from it can be installed into the storage section 308 as needed.

[0125] Specifically, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the Central Processing Unit (CPU) 301, various functions defined in the system of the present application are executed.

[0126] It should be noted that the computer-readable medium shown in the embodiments of the present application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0127] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0128] The units involved in the embodiments described in this application can be implemented in software or in hardware, and the described units can also be provided in a processor. Among them, the names of these units do not constitute a limitation to the units themselves in some cases.

[0129] Another aspect of this application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the fault handling method of the intelligent connected vehicle cloud platform as described above. The computer-readable storage medium can be included in the electronic device described in the above embodiments, or can exist alone without being assembled into the electronic device.

[0130] Another aspect of this application also provides a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the fault handling method of the intelligent connected vehicle cloud platform provided in the above various embodiments.

[0131] The above are only the preferred embodiments of this application, and are not intended to limit this application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of this application shall be included in the protection scope of this application.

Claims

1. A fault handling method for an intelligent networked vehicle cloud platform, characterized in that, The method includes: Obtaining the construction and operation and maintenance knowledge of the intelligent connected vehicle cloud platform, the platform architecture diagram, and the terminal data of the service terminals of the intelligent connected vehicle cloud platform; Fine-tuning the vision language model based on the construction and operation and maintenance knowledge, the platform architecture diagram, and the terminal data to obtain the platform customization model of the intelligent connected vehicle cloud platform; Obtaining the monitoring data of the service terminals, and using the platform customization model to output the fault result information of the service terminals based on the monitoring data; Determining the target fault matching data based on the pre-configured platform fault information, where the target fault matching data includes the repetition rate between the platform fault information and the fault result information; Determining the target fault handling rule based on the target fault matching data and the corresponding relationship between the preset fault handling rules and the fault matching data, so as to perform fault handling on the intelligent connected vehicle cloud platform based on the target fault handling rule.

2. The method according to claim 1, wherein The fine-tuning the vision language model based on the construction and operation and maintenance knowledge, the platform architecture diagram, and the terminal data to obtain the platform customization model of the intelligent connected vehicle cloud platform includes: Preprocessing the construction and operation and maintenance knowledge, the platform architecture diagram, and the terminal data to obtain key-value pair data; Converting the format of the key-value data pairs to obtain fine-tuning data in json format; Fine-tuning the vision language model based on the fine-tuning data to obtain an initial customization model, where the vision language model is the deepseek-vl2 large model; Privately deploying the initial customization model using the open-source local large model operation framework to obtain the platform customization model.

3. The method according to claim 2, wherein The method further includes: Deploying a retrieval system using the open-source local large model operation framework; Obtaining the record data generated during the process of performing fault handling on the intelligent connected vehicle cloud platform, and storing the record data using the retrieval system.

4. The method according to claim 2, wherein The preprocessing the construction and operation and maintenance knowledge, the platform architecture diagram, and the terminal data to obtain key-value pair data includes: Extracting key information from the construction and operation and maintenance knowledge through a preset language large model to obtain the first key-value pair, where the first key-value pair is in the form of a key-value pair of questions and answers, and the language large model is the deepseek-r1 model; Obtaining the second key-value pair based on the platform architecture diagram, where the second key-value pair is in the form of a key-value pair of a scene name and an image; Obtaining the numbers and metadata of the service terminals, and obtaining the third key-value pair based on the numbers and metadata, where the third key-value pair is in the form of a key-value pair of numbers and metadata, and the service terminals include vehicle terminals, road terminals, and cloud terminals.

5. The method according to claim 1, wherein The obtaining the monitoring data of the service terminals, and using the platform customization model to output the fault result information of the service terminals based on the monitoring data includes: Obtaining the monitoring data of the service terminals, and configuring the threshold conditions corresponding to the monitoring data; When the monitoring data of at least two terminals in the service terminals meet the corresponding threshold conditions, selecting the monitoring data that meets the corresponding threshold conditions as the fault monitoring data; The platform-customized model is used to output corresponding fault result information based on the fault monitoring data.

6. The method according to claim 5, characterized in that, The service terminal includes a vehicle terminal, a road terminal, and a cloud; obtaining the monitoring data of the service terminal and configuring the threshold conditions corresponding to the monitoring data includes: Obtaining the vehicle device status data of the vehicle terminal and the roadside device status data of the road terminal; Obtaining the server status data, platform application status data, database status data, and middleware status data of the cloud and adding tag information to obtain the cloud monitoring items of the cloud, where the tag information includes the number and metadata of the cloud; Using an open-source monitoring system to form time-series data based on the cloud monitoring data; Using a neural network to obtain the predicted monitoring value corresponding to the cloud monitoring data based on the time-series data and transmitting the predicted monitoring value back to the open-source monitoring system to obtain new cloud monitoring items; Configuring the threshold conditions corresponding to the vehicle device status data, the roadside device status data, and the new cloud monitoring items.

7. The method according to claim 5, wherein The using the platform-customized model to output corresponding fault result information based on the fault monitoring data includes: Invoking the dialogue interface of the open-source local large model operation framework, and inputting the dialogue data and problem data corresponding to the fault monitoring data into the platform-customized model, where the platform-customized model is deployed in the open-source local large model operation framework; Using the platform-customized model to obtain a fault scenario topology diagram based on the dialogue data; Using the platform-customized model, through an image inference method, to obtain a root cause terminal based on the problem data and the fault scenario architecture diagram; Using the platform-customized model to call the construction and operation and maintenance knowledge to determine the fault cause and repair suggestions corresponding to the fault monitoring data; Obtaining fault result information based on the fault scenario topology diagram, the root cause terminal, the fault cause, and the repair suggestions.

8. The method according to claim 7, wherein After obtaining the monitoring data of the service terminal and configuring the threshold conditions corresponding to the monitoring data, the method further includes: Configuring the alarm data corresponding to the monitoring data, where the alarm data includes an alarm notifier; In the case of obtaining the fault result information output by the platform-customized model, sending the fault cause and the repair suggestions included in the fault result information to the corresponding alarm notifier.

9. The method according to claim 1, wherein The determining the target fault handling rule based on the target fault matching data and the corresponding relationship between the preset fault handling rule and the fault matching data includes: If the repetition rate included in the target fault matching data indicates that the fault result information is exactly the same as the platform fault information, then based on the corresponding relationship between the preset fault handling rule and the fault matching data, determining the target fault handling rule corresponding to the fault result information as the rule for automatically executing a script; If the repetition rate indicates that the fault result information is partially the same as the platform fault information, then based on the corresponding relationship, determining the target fault handling rule corresponding to the fault result information as the rule for manual verification and script execution; If the repetition rate indicates that the fault result information and the platform fault information are completely non - repetitive, then based on the corresponding relationship, the target fault handling rule corresponding to the fault result information is determined to be a manual handling rule.

10. A fault handling device for an intelligent networked vehicle cloud platform, characterized in that, The device includes: A collection unit, configured to obtain the construction and operation and maintenance knowledge of the intelligent connected vehicle cloud platform, the platform architecture diagram, and the terminal data of the service terminal of the intelligent connected vehicle cloud platform; A model construction unit, configured to fine - tune the vision - language model based on the construction and operation and maintenance knowledge, the platform architecture diagram, and the terminal data to obtain the platform customized model of the intelligent connected vehicle cloud platform; A processing unit, configured to obtain the monitoring data of the service terminal, and use the platform customized model to output the fault result information of the service terminal based on the monitoring data; A determination unit, configured to determine target fault matching data based on the pre - configured platform fault information, where the target fault matching data includes the repetition rate between the platform fault information and the fault result information; An execution unit, configured to determine a target fault handling rule based on the target fault matching data and the corresponding relationship between the preset fault handling rules and the fault matching data, so as to perform fault handling on the intelligent connected vehicle cloud platform based on the target fault handling rule.

Images