Automobile fault diagnosis early warning system and method based on artificial intelligence

Abstract

The invention provides an automobile fault diagnosis early warning system and method based on artificial intelligence, and belongs to the technical field of automobile fault diagnos.The automobile fault diagnosis early warning system combines vehicle local computing resources with cloud computing resources, dynamically allocates computing tasks according to the network state of the cloud, ensures efficient utilization of the resources, and improves the working efficiency. And secondly, local computing resources of the vehicle are divided into a plurality of processing nodes, and the processing nodes are dynamically allocated according to the weight level of the parameters, so that adjustment can be performed according to actual computing requirements, the response speed and computing power of the system are improved, and the reliability of the system is improved. Important monitoring parameters can be processed in time, and a good guarantee is provided for real-time judgment of automobile faults.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile fault diagnosis, and in particular to an automobile fault diagnosis and early warning system and method based on artificial intelligence. Background Art

[0002] With the rapid development of the automobile industry and the advancement of intelligent technology, automobiles have become an indispensable means of transportation in modern society. However, with the increasing complexity of automobiles, their fault diagnosis and early warning are also facing more and more challenges. In the fault diagnosis and early warning system based on artificial intelligence, the allocation of computing resources is a key issue. For example, some automobile fault detection, its data processing is generally based on the cloud or the computing resources of the car itself. When the cloud network is unstable, the data processing is limited or difficult to process, which affects the real-time judgment of automobile faults. Directly processing data through the car's own computing resources will cause excessive load on the car system, affecting the system processing effect of the car.

[0003] Therefore, it is necessary to provide an automobile fault diagnosis and early warning system and method based on artificial intelligence to solve the above technical problems. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides an automobile fault diagnosis and warning system and method based on artificial intelligence to solve the problems of allocation of computing resources consumed by data processing and real-time data processing during existing automobile fault judgment.

[0005] The present invention provides an automobile fault diagnosis and early warning method based on artificial intelligence, and the diagnosis and early warning method comprises the following steps:

[0006] According to the preset cycle, obtain the parameter data of the current monitoring vehicle electronic control state and driving state;

[0007] Divide the vehicle's local computing resources into multiple processing nodes, and define multiple network levels in the cloud and the processing levels of various parameters corresponding to the multiple network levels;

[0008] Obtain the current network level of the cloud in real time, and divide the processing level of various parameters currently monitoring the vehicle's electronic control status and driving status;

[0009] Dynamically distribute various parameters to the cloud and local vehicle computing resources based on the processing level of various parameters currently monitoring the vehicle's electronic control status and driving status;

[0010] The parameters allocated to the vehicle's local computing resources are divided into weight levels, and the weight levels are divided to allocate a corresponding number of processing nodes to each parameter;

[0011] Based on allocating a corresponding number of processing nodes to each parameter, the vehicle's local computing resources work with the cloud to simultaneously process each parameter and extract parameter features;

[0012] The extracted parameter features are input into the pre-trained fault judgment model, and the final diagnosis results are output and an early warning is issued.

[0013] Preferably, the specific steps of obtaining the parameter data for monitoring the electronic control state and driving state of the vehicle according to a preset cycle are as follows:

[0014] Set a monitoring cycle, and obtain the parameter data of the vehicle's electronic control state and driving state in real time through sensors and data acquisition equipment according to the monitoring cycle;

[0015] The parameter data of the electric control state and driving state of the vehicle are stored in the vehicle local storage library.

[0016] Preferably, the vehicle local computing resources are divided into a plurality of processing nodes, and a plurality of network levels in the cloud and the processing levels of various parameters corresponding to the plurality of network levels are defined, and the specific steps are as follows:

[0017] Obtain information about the vehicle's local computing resources, including processor performance, memory size, storage space, and computing power;

[0018] Based on the local computing resource information of the vehicle, the local computing resources of the vehicle are divided into a plurality of processing nodes;

[0019] According to the network characteristics and status of the cloud, the cloud network is divided into multiple network levels, including high-speed network level, medium-speed network level, low-speed network level and no network level;

[0020] A corresponding parameter processing level is defined for each network level, and a mapping relationship between the network level and the processing level is established.

[0021] Preferably, the real-time acquisition of the current network level of the cloud and the classification of the processing levels of various parameters for monitoring the electronic control state and driving state of the vehicle are carried out in the following specific steps:

[0022] Monitor the network characteristics and status of the cloud, including but not limited to network bandwidth, latency, and connection stability;

[0023] Determine the current network level of the cloud based on monitoring the network characteristics and status of the cloud;

[0024] Based on the current network level of the cloud, the processing level of various parameters currently monitoring the vehicle's electronic control state and driving state is divided by establishing a mapping relationship between the network level and the processing level.

[0025] Preferably, the process of dynamically allocating various parameters to cloud and vehicle local computing resources according to the processing level of various parameters currently monitoring the electronic control state and driving state of the vehicle comprises the following specific steps:

[0026] The allocation strategy is executed according to the processing level of various parameters currently monitoring the vehicle's electronic control state and driving state and the current network level of the cloud, specifically:

[0027] For high-speed network level, the highest level of processing parameters are allocated to the cloud;

[0028] For medium-speed network level, parameters of medium level and below are allocated to the cloud, and the remaining parameters are allocated to the vehicle's local computing resources.

[0029] For low-speed network levels, the lowest-level parameters are allocated to the cloud, and the remaining parameters are allocated to the vehicle's local computing resources;

[0030] If there is no network level, all parameters are allocated to the vehicle's local computing resources;

[0031] Monitor changes in the cloud network status in real time, obtain the latest network characteristics and status of the cloud, and update the current network level of the cloud;

[0032] According to the updated current network level of the cloud, the execution allocation strategy of various parameters is adjusted in real time.

[0033] Preferably, the parameters allocated to the vehicle local computing resources are divided into weight levels, and the weight levels are divided to allocate a corresponding number of processing nodes to each parameter, and the specific steps are:

[0034] The parameters allocated to the vehicle's local computing resources are weighted according to their safety importance, urgency, and processing complexity.

[0035] According to the weight level of each parameter allocated to the vehicle's local computing resources, a corresponding number of processing nodes are allocated to each parameter. Specifically, the higher the weight level of the parameter, the more processing nodes are allocated.

[0036] Preferably, the specific steps of pre-training the fault judgment model are:

[0037] Obtain the historical electronic control state and driving state parameter data of the vehicle and the corresponding fault problems;

[0038] Preprocess the parameter data of the vehicle's historical electronic control state and driving state, extract features, and obtain the mapping relationship between features and fault problems through data analysis methods;

[0039] The mapping relationship between features and fault problems is used as the training set, and a machine learning model framework is selected for training to obtain a fault judgment model.

[0040] An automobile fault diagnosis and early warning system based on artificial intelligence, the diagnosis and early warning system comprising:

[0041] The monitoring and acquisition module is used to obtain parameter data of the current monitoring vehicle electronic control state and driving state according to a preset cycle;

[0042] A partition definition module is used to divide the vehicle's local computing resources into multiple processing nodes and define multiple network levels in the cloud and the processing levels of various parameters corresponding to the multiple network levels;

[0043] The priority classification module is used to obtain the current network level of the cloud in real time and classify the processing level of various parameters currently monitoring the vehicle's electronic control status and driving status;

[0044] A dynamic allocation module is used to dynamically allocate various parameters to the cloud and local vehicle computing resources according to the processing level of various parameters currently monitoring the vehicle's electronic control state and driving state;

[0045] A weight division module, used to divide the parameters divided into vehicle local computing resources into weight levels, and allocate a corresponding number of processing nodes to each parameter according to the divided weight level;

[0046] A data processing module is used to allocate a corresponding number of processing nodes to each parameter, and the vehicle's local computing resources and the cloud simultaneously process the parameters to extract parameter features;

[0047] The diagnosis and warning module is used to input the extracted parameter features into the pre-trained fault judgment model, output the final diagnosis results and issue a warning.

[0048] Compared with the related art, the automobile fault diagnosis and early warning system and method based on artificial intelligence provided by the present invention has the following beneficial effects:

[0049] The present invention combines the vehicle's local computing resources with the cloud computing resources, dynamically allocates computing tasks according to the network status of the cloud, thereby ensuring efficient use of resources and the continuity and real-time nature of fault diagnosis. Secondly, by dividing the vehicle's local computing resources into multiple processing nodes and dynamically allocating processing nodes according to parameter weight levels, it can be adjusted according to actual computing needs, thereby improving the system's response speed and computing power, facilitating timely processing of important monitoring parameters, and providing good protection for real-time judgment of vehicle faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1A flowchart of an automobile fault diagnosis and early warning method based on artificial intelligence of the present invention;

[0051] Figure 2 The system block diagram of an automobile fault diagnosis and early warning system based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0052] The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

[0053] Embodiment 1

[0054] like Figure 1 As shown, an automobile fault diagnosis and early warning method based on artificial intelligence includes the following steps:

[0055] S100, obtaining parameter data for monitoring the electronic control state and driving state of the vehicle according to a preset cycle;

[0056] In the specific implementation process, the specific steps of step S100 are:

[0057] S101, setting a monitoring period, and acquiring parameter data of the electric control state and driving state of the vehicle in real time through sensors and data acquisition equipment according to the monitoring period;

[0058] Specifically, in this embodiment, the monitoring cycle is set to once a day, and the parameter data of the vehicle's electronic control state and driving state are obtained in real time through sensors including but not limited to temperature sensors, pressure sensors, and speed sensors. The parameter data specifically include the vehicle's electronic control state (such as engine control unit data), driving state (such as wheel speed, steering angle, etc.) and other related parameters.

[0059] S102, storing the parameter data of the electric control state and the driving state of the vehicle in a local storage library of the vehicle;

[0060] Specifically, the parameter data of the vehicle's electronic control state and driving state are obtained and a suitable storage format is selected according to the type and characteristics of the data. In this embodiment, for time series data, a database or file system can be selected for storage, wherein the storage location is specifically a built-in storage device of the vehicle, such as a hard disk or a solid-state drive.

[0061] S200, dividing the vehicle local computing resources into a plurality of processing nodes, and defining a plurality of network levels in the cloud and processing levels of various parameters corresponding to the plurality of network levels;

[0062] In the specific implementation process, the specific steps of step S200 are:

[0063] S201, obtaining local computing resource information of the vehicle, including processor performance, memory size, storage space, and computing power;

[0064] Specifically, first, the vehicle's built-in computer hardware is tested, including the processor (CPU), memory (RAM), storage devices (such as hard disks or SSDs), etc., and the processor's computing speed, memory's read and write speed, and storage device's read and write speed and capacity are evaluated through existing test algorithms. Based on the hardware performance evaluation results, the computing power of the vehicle's local computing resources is calculated to determine the amount of data and complexity that can be processed.

[0065] S202, based on the local computing resource information of the vehicle, dividing the local computing resources of the vehicle into a plurality of processing nodes;

[0066] Specifically, the vehicle's local computing resources are divided into multiple processing nodes based on the amount of data and the computing power of the complexity that can be processed. It should be noted that each processing node can be an independent computing unit, such as a CPU core, or a computing cluster composed of multiple cores.

[0067] S203, dividing the cloud network into multiple network levels according to the network characteristics and status of the cloud, specifically including a high-speed network level, a medium-speed network level, a low-speed network level, and no network level;

[0068] Specifically, through specific network monitoring tools, the network characteristics of the cloud, such as bandwidth, latency and connection stability, are monitored in real time, and the cloud network is divided into multiple network levels, including high-speed network level (high bandwidth, low latency), medium-speed network level (medium bandwidth, moderate latency), low-speed network level (low bandwidth, high latency) and no network level (unable to connect to the cloud).

[0069] S204: Define a corresponding parameter processing level for each network level, and establish a mapping relationship between the network level and the processing level.

[0070] Specifically, for high-speed network levels, the highest processing level is set, for medium-speed network levels, a medium processing level is set, and for low-speed network levels and no network levels, the lowest processing level is set or no data processing is performed at all. In this way, a mapping relationship between these network levels and processing levels is established to ensure that the parameter processing level can be automatically adjusted when the network level changes.

[0071] S300, obtaining the current network level of the cloud in real time, and classifying the processing level of various parameters currently monitoring the electronic control state and driving state of the vehicle;

[0072] In the specific implementation process, the specific steps of step S300 are:

[0073] S301. Monitor the network characteristics and status of the cloud, including but not limited to network bandwidth, latency, and connection stability;

[0074] Specifically, the network characteristics and status of the cloud are monitored by a network monitoring tool installed on the cloud server. In this embodiment, the network monitoring tool used is Network Monitor. For example, the bandwidth status of the network is evaluated by calculating the average value and fluctuation range of the bandwidth over a period of time; the network delay is calculated by measuring the round-trip time of the data packet; and the stability of the connection is evaluated by monitoring the number and duration of connection interruptions.

[0075] S302, determining the current network level of the cloud based on the network characteristics and status of the cloud;

[0076] Specifically, if the network characteristics and status of the detected cloud are that the bandwidth is greater than 80Mbps and the connection stability is good, it is determined to be a high-speed network level; if the network characteristics and status of the detected cloud are that the bandwidth is between 80Mbps-50Mbps, it is determined to be a medium-speed network level; if the network characteristics and status of the detected cloud are that the bandwidth is below 50Mbps, it is determined to be a low-speed network level; if the network characteristics and status of the cloud are none or below 10Mbps, it is determined to be no network level.

[0077] S303. Based on the current network level of the cloud, the processing level of various parameters currently monitoring the electronic control state and driving state of the vehicle is divided by establishing a mapping relationship between the network level and the processing level.

[0078] S400, dynamically allocating various parameters to cloud and vehicle local computing resources according to the processing levels of various parameters currently monitoring the electronic control state and driving state of the vehicle;

[0079] In the specific implementation process, the specific steps of step S400 are:

[0080] S401: Execute the allocation strategy according to the processing level of various parameters currently monitoring the electronic control state and driving state of the vehicle and the current network level of the cloud, specifically:

[0081] For high-speed network level, the highest level of processing parameters are allocated to the cloud;

[0082] For medium-speed network level, parameters of medium level and below are allocated to the cloud, and the remaining parameters are allocated to the vehicle's local computing resources.

[0083] For low-speed network levels, the lowest-level parameters are allocated to the cloud, and the remaining parameters are allocated to the vehicle's local computing resources;

[0084] If there is no network level, all parameters are allocated to the vehicle's local computing resources;

[0085] S402, monitor the changes of the cloud network status in real time, obtain the latest network characteristics and status of the cloud, and update the current network level of the cloud;

[0086] Specifically, the network monitoring tool Network Monitor is used to monitor the changes in the network status of the cloud in real time, including fluctuations in characteristics such as network bandwidth, latency, and connection stability, and updates the current network level of the cloud based on the latest network characteristics and status information obtained.

[0087] S403: According to the updated current network level of the cloud, the execution allocation strategy of various parameters is adjusted in real time.

[0088] Specifically, according to the updated network level information, the execution allocation strategy of each parameter is adjusted in real time, including reallocating parameters with high processing levels to the cloud or local vehicle computing resources.

[0089] For example, when it is determined that the current network level of the cloud has changed from a high-speed network level to a medium-speed network level, the allocation strategy is adjusted immediately. For example, certain parameters with medium or high processing levels originally allocated to the cloud may be reallocated to the vehicle's local computing resources for processing to ensure that these parameters can still be processed in a timely manner even when the network conditions are poor.

[0090] S500, dividing the parameters allocated to the vehicle local computing resources into weight levels, and allocating a corresponding number of processing nodes to each parameter according to the divided weight levels;

[0091] In the specific implementation process, the specific steps of step S500 are:

[0092] S501, weighting various parameters allocated to the vehicle local computing resources according to safety importance, urgency, and processing complexity factors of the parameters;

[0093] Specifically, the impact of evaluation parameters on the safe operation of the vehicle, specifically parameters related to the vehicle's braking system and steering system, which are of high safety importance; the real-time requirements of evaluation parameters during vehicle operation, specifically parameters such as the vehicle's speed and acceleration, which are of high urgency; the computing resources and time required to process these parameters are evaluated, specifically certain advanced driver assistance system (ADAS) functions, such as lane keeping and automatic emergency braking. According to the evaluation results, the parameters are divided into three levels: high, medium and low. The high-level parameters include key parameters directly related to safety, the medium-level parameters include important parameters related to vehicle performance, and the low-level parameters include some non-critical auxiliary parameters.

[0094] S502. According to the weight level of each parameter allocated to the vehicle local computing resources, a corresponding number of processing nodes are allocated to each parameter. Specifically, the higher the weight level of the parameter, the more processing nodes are allocated.

[0095] Specifically, high-level parameters have higher safety importance and urgency, so more processing nodes are allocated to them to ensure that these parameters can be processed in a timely and accurate manner. Specifically, for the parameters of the vehicle braking system, multiple high-performance CPU or GPU computing units can be allocated to them to ensure rapid response in emergency situations; the processing node allocation of medium-level parameters can be adjusted according to their urgency and processing complexity. Specifically, for parameters such as vehicle speed and acceleration, an appropriate amount of processing nodes can be allocated to them to ensure the real-time and accuracy of these parameters; a small number of processing nodes are allocated to low-level parameters. Specifically, for some non-critical auxiliary parameters such as in-vehicle temperature and audio volume, a small number of processing nodes can be allocated to them.

[0096] S600, based on allocating a corresponding number of processing nodes to each parameter, the local computing resources of the vehicle and the cloud simultaneously process the parameters to extract parameter features;

[0097] S700. Input the extracted parameter features into a pre-trained fault judgment model, output the final diagnosis result and issue a warning. If the final diagnosis result is a fault, a corresponding warning is issued. In this embodiment, the warning is issued by voice or information notification on the vehicle interface or the mobile terminal vehicle-related software.

[0098] In the specific implementation process, the specific steps of pre-training the fault judgment model in step S700 are:

[0099] S701, obtaining parameter data of the historical electronic control state and driving state of the vehicle and corresponding fault problems;

[0100] S702, preprocessing the parameter data of the historical electronic control state and driving state of the vehicle, extracting features, and obtaining a mapping relationship between the features and the fault problem through a data analysis method;

[0101] S703. Using the mapping relationship between features and fault problems as a training set, a machine learning model framework is selected for training to obtain a fault judgment model.

[0102] Embodiment 2

[0103] like Figure 2 As shown, an artificial intelligence-based automobile fault diagnosis and warning system is applied to an artificial intelligence-based automobile fault diagnosis and warning method, and the diagnosis and warning system specifically includes:

[0104] The monitoring and acquisition module is used to obtain parameter data of the current monitoring vehicle electronic control state and driving state according to a preset cycle;

[0105] A partition definition module is used to divide the vehicle's local computing resources into multiple processing nodes and define multiple network levels in the cloud and the processing levels of various parameters corresponding to the multiple network levels;

[0106] The priority classification module is used to obtain the current network level of the cloud in real time and classify the processing level of various parameters currently monitoring the vehicle's electronic control status and driving status;

[0107] A dynamic allocation module is used to dynamically allocate various parameters to the cloud and local vehicle computing resources according to the processing level of various parameters currently monitoring the vehicle's electronic control state and driving state;

[0108] A weight division module, used to divide the parameters divided into vehicle local computing resources into weight levels, and allocate a corresponding number of processing nodes to each parameter according to the divided weight level;

[0109] A data processing module is used to allocate a corresponding number of processing nodes to each parameter, and the vehicle's local computing resources and the cloud simultaneously process the parameters to extract parameter features;

[0110] The diagnosis and warning module is used to input the extracted parameter features into the pre-trained fault judgment model, output the final diagnosis results and issue a warning.

[0111] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0112] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0113] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. An automobile fault diagnosis and early warning method based on artificial intelligence, characterized in that: The diagnostic and early warning method comprises the following steps: According to the preset cycle, obtain the parameter data of the current monitoring vehicle electronic control state and driving state; Divide the vehicle's local computing resources into multiple processing nodes, and define multiple network levels in the cloud and the processing levels of various parameters corresponding to the multiple network levels; Obtain the current network level of the cloud in real time, and divide the processing level of various parameters currently monitoring the vehicle's electronic control status and driving status; Dynamically distribute various parameters to the cloud and local vehicle computing resources based on the processing level of various parameters currently monitoring the vehicle's electronic control status and driving status; The parameters allocated to the vehicle's local computing resources are divided into weight levels, and the weight levels are divided to allocate a corresponding number of processing nodes to each parameter; Based on allocating a corresponding number of processing nodes to each parameter, the vehicle's local computing resources work with the cloud to simultaneously process each parameter and extract parameter features; The extracted parameter features are input into the pre-trained fault judgment model, and the final diagnosis results are output and an early warning is issued.

2. The automobile fault diagnosis and early warning method based on artificial intelligence according to claim 1 is characterized in that: The specific steps of obtaining the parameter data of the current monitoring vehicle electronic control state and driving state according to the preset cycle are as follows: Set a monitoring cycle, and obtain the parameter data of the vehicle's electronic control state and driving state in real time through sensors and data acquisition equipment according to the monitoring cycle; The parameter data of the electric control state and driving state of the vehicle are stored in the vehicle local storage library.

3. The automobile fault diagnosis and early warning method based on artificial intelligence according to claim 1 is characterized in that: The specific steps of dividing the vehicle local computing resources into multiple processing nodes and defining multiple network levels in the cloud and the processing levels of various parameters corresponding to the multiple network levels are as follows: Obtain information about the vehicle's local computing resources, including processor performance, memory size, storage space, and computing power; Based on the local computing resource information of the vehicle, the local computing resources of the vehicle are divided into a plurality of processing nodes; According to the network characteristics and status of the cloud, the cloud network is divided into multiple network levels, including high-speed network level, medium-speed network level, low-speed network level and no network level; A corresponding parameter processing level is defined for each network level, and a mapping relationship between the network level and the processing level is established.

4. The automobile fault diagnosis and early warning method based on artificial intelligence according to claim 1 is characterized in that: The specific steps of obtaining the current network level of the cloud in real time and dividing the processing level of various parameters of the current monitoring vehicle electronic control state and driving state are as follows: Monitor the network characteristics and status of the cloud, including but not limited to network bandwidth, latency, and connection stability; Determine the current network level of the cloud based on monitoring the network characteristics and status of the cloud; Based on the current network level of the cloud, the processing level of various parameters currently monitoring the vehicle's electronic control state and driving state is divided by establishing a mapping relationship between the network level and the processing level.

5. The automobile fault diagnosis and early warning method based on artificial intelligence according to claim 1 is characterized in that: The specific steps of dynamically allocating various parameters to the cloud and the vehicle's local computing resources according to the processing levels of various parameters currently monitoring the vehicle's electronic control state and driving state are as follows: The allocation strategy is executed according to the processing level of various parameters currently monitoring the vehicle's electronic control state and driving state and the current network level of the cloud, specifically: For high-speed network level, the highest level of processing parameters are allocated to the cloud; For medium-speed network level, parameters of medium level and below are allocated to the cloud, and the remaining parameters are allocated to the vehicle's local computing resources. For low-speed network levels, the lowest-level parameters are allocated to the cloud, and the remaining parameters are allocated to the vehicle's local computing resources; If there is no network level, all parameters are allocated to the vehicle's local computing resources; Monitor changes in the cloud network status in real time, obtain the latest network characteristics and status of the cloud, and update the current network level of the cloud; According to the updated current network level of the cloud, the execution allocation strategy of various parameters is adjusted in real time.

6. The automobile fault diagnosis and early warning method based on artificial intelligence according to claim 1 is characterized in that: The parameters allocated to the local computing resources of the vehicle are divided into weight levels, and the weight levels are divided to allocate a corresponding number of processing nodes to each parameter, and the specific steps are: The parameters allocated to the vehicle's local computing resources are weighted according to their safety importance, urgency, and processing complexity. According to the weight level of each parameter allocated to the vehicle's local computing resources, a corresponding number of processing nodes are allocated to each parameter. Specifically, the higher the weight level of the parameter, the more processing nodes are allocated.

7. The automobile fault diagnosis and early warning method based on artificial intelligence according to claim 1 is characterized in that: The specific steps of pre-training the fault judgment model are: Obtain the historical electronic control state and driving state parameter data of the vehicle and the corresponding fault problems; Preprocess the parameter data of the vehicle's historical electronic control state and driving state, extract features, and obtain the mapping relationship between features and fault problems through data analysis methods; The mapping relationship between features and fault problems is used as the training set, and a machine learning model framework is selected for training to obtain a fault judgment model.

8. An automobile fault diagnosis and early warning system based on artificial intelligence, applied to an automobile fault diagnosis and early warning method based on artificial intelligence as claimed in any one of claims 1 to 7, characterized in that: The diagnostic early warning system comprises: The monitoring and acquisition module is used to obtain parameter data of the current monitoring vehicle electronic control state and driving state according to a preset cycle; A partition definition module is used to divide the vehicle's local computing resources into multiple processing nodes and define multiple network levels in the cloud and the processing levels of various parameters corresponding to the multiple network levels; The priority classification module is used to obtain the current network level of the cloud in real time and classify the processing level of various parameters currently monitoring the vehicle's electronic control status and driving status; A dynamic allocation module is used to dynamically allocate various parameters to the cloud and local vehicle computing resources according to the processing level of various parameters currently monitoring the vehicle's electronic control state and driving state; A weight division module, used to divide the parameters divided into vehicle local computing resources into weight levels, and allocate a corresponding number of processing nodes to each parameter according to the divided weight level; A data processing module is used to allocate a corresponding number of processing nodes to each parameter, and the vehicle's local computing resources and the cloud simultaneously process the parameters to extract parameter features; The diagnosis and warning module is used to input the extracted parameter features into the pre-trained fault judgment model, output the final diagnosis results and issue a warning.

Images