Vehicle fault processing method and device, electronic equipment and storage medium

Abstract

The invention provides a vehicle fault processing method and device, electronic equipment and a storage medium, and belongs to the technical field of vehicle fault processing, and the method comprises the steps: obtaining multiple pieces of vehicle feature information; the multiple pieces of vehicle feature information are weighted, weighted feature information is determined, different pieces of vehicle feature information are generated after different vehicle features are collected, and different feature weights are preset for each vehicle feature; inputting the weighted feature information into a gradient lifting prediction model to obtain predicted fault information output by the gradient lifting prediction model; and sending the predicted fault information to the target server for fault diagnosis, and obtaining a fault diagnosis strategy returned by the target server. According to the vehicle fault processing method provided by the invention, the technical problem that the vehicle fault processing method in the related technology cannot accurately predict the potential fault of the vehicle and cannot process the vehicle fault in time is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle fault processing, and in particular to a vehicle fault processing method, device, electronic equipment and storage medium. Background Art

[0002] Vehicle fault detection technology is an important means to improve vehicle safety and reduce maintenance costs, and can predict potential vehicle failures in advance; the vehicle fault handling methods in related technologies are still at the basic data analysis level, and most of them push fault information after the relevant threshold of the vehicle reaches the fault threshold, which cannot meet the needs of predicting potential vehicle failures in advance, and after determining the vehicle failure, they are often limited to hardware repairs at the maintenance station, and cannot handle the vehicle failure in a timely manner.

[0003] It can be seen that the vehicle fault handling method in the related art has the technical problem of being unable to accurately predict potential vehicle faults and handle vehicle faults in a timely manner. Summary of the invention

[0004] The present invention provides a vehicle fault handling method, device, electronic device and storage medium, which are used to solve the technical problem that the vehicle fault handling method in the related art cannot accurately predict the potential vehicle fault and handle the vehicle fault in time.

[0005] The present invention provides a vehicle fault processing method, comprising the following steps: obtaining a plurality of vehicle feature information; performing weighted processing on the plurality of vehicle feature information to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; inputting the weighted feature information into a gradient boosting prediction model to obtain predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on feature information samples and their corresponding fault information labels; sending the predicted fault information to a target server for fault diagnosis, and obtaining a fault diagnosis strategy returned by the target server.

[0006] According to a vehicle fault handling method provided by the present invention, before performing weighted processing on the multiple vehicle feature information to determine the weighted feature information, the method also includes: inputting multiple vehicle features into a target evaluation model to obtain multiple feature importance scores output by the target evaluation model, wherein the target evaluation model is any one of a decision tree model, a random forest model, and a gradient boosting machine model, and the multiple feature importance scores correspond to the multiple vehicle features respectively; and determining multiple feature weights based on the multiple feature importance scores.

[0007] According to a vehicle fault handling method provided by the present invention, the gradient boosting prediction model is an XGBoost prediction model.

[0008] According to a vehicle fault handling method provided by the present invention, the XGBoost prediction model is obtained based on the following steps: initializing the model parameters of the XGBoost prediction model; obtaining a training data set; training the XGBoost prediction model using the training data set; and using the model parameters obtained when the number of iterations reaches a preset number as the model parameters of the trained XGBoost prediction model.

[0009] According to a vehicle fault handling method provided by the present invention, the predicted fault information is sent to a target server for fault diagnosis, and a fault diagnosis strategy returned by the target server is obtained, comprising: sending the predicted fault information to the target server through a target diagnosis protocol; obtaining the fault diagnosis strategy returned by the target server, wherein the fault diagnosis strategy is determined by the target server based on querying a real vehicle fault management library based on the predicted fault information.

[0010] According to a vehicle fault handling method provided by the present invention, after sending the predicted fault information to a target server for fault diagnosis and obtaining a fault diagnosis strategy returned by the target server, the method further includes: receiving an over-the-air upgrade data packet sent by the target server, wherein the over-the-air upgrade data packet is determined by the target server based on the fault diagnosis strategy; and installing the over-the-air upgrade data packet.

[0011] The present invention also provides a vehicle fault handling device, comprising the following modules: an acquisition module, used to acquire multiple vehicle feature information; a first execution module, used to perform weighted processing on the multiple vehicle feature information to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; a second execution module, used to input the weighted feature information into a gradient boosting prediction model to obtain predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on feature information samples and their corresponding fault information labels; a third execution module, used to send the predicted fault information to a target server for fault diagnosis, and obtain a fault diagnosis strategy returned by the target server.

[0012] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the vehicle fault handling methods described above is implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the vehicle fault handling method as described in any one of the above is implemented.

[0014] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the vehicle fault handling method as described above is implemented.

[0015] The vehicle fault handling method, device, electronic device and storage medium provided by the present invention perform weighted processing on multiple vehicle feature information to determine weighted feature information, which can increase the amount of information contained in the weighted feature information, and use the gradient boosting prediction model to predict the fault information to be determined based on the weighted feature information, so as to accurately predict the current predicted fault information of the vehicle, send the predicted fault information to the target server for fault diagnosis, obtain the fault diagnosis strategy returned by the target server, and handle the vehicle fault in a timely manner, thereby solving the technical problem that the vehicle fault handling method in the related technology cannot accurately predict the potential fault of the vehicle and handle the vehicle fault in a timely manner, thereby improving the reliability of the vehicle fault handling method. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced one by one below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a schematic diagram of a vehicle fault handling process in the related technology.

[0018] Figure 2 It is one of the flow charts of the vehicle fault handling method provided by the present invention.

[0019] Figure 3 This is the second flow chart of the vehicle fault handling method provided by the present invention.

[0020] Figure 4 It is a structural schematic diagram of the vehicle fault processing device provided by the present invention.

[0021] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] It should be noted that in the description of the present invention, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device. In the absence of further restrictions, the elements defined by the sentence "include one..." do not exclude the existence of other identical elements in the process, method, article or device including the elements. The orientation or position relationship indicated by the terms "upper", "lower" and the like is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0024] The terms "first", "second", etc. in the present invention are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present invention can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally of the same type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" means at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0025] Combine the following Figure 1-Figure 5 The present invention describes a vehicle fault handling method, device, electronic device and storage medium.

[0026] Vehicle fault detection technology is an important means to improve vehicle safety and reduce maintenance costs. It can predict potential vehicle failures in advance. Modern cars are generally equipped with on-board diagnostic systems, which can monitor the operating status of various systems in real time and automatically record fault codes when a fault occurs. By reading these fault codes, maintenance personnel can quickly locate the fault site and improve diagnostic efficiency.

[0027] Figure 1 It is a schematic diagram of the vehicle fault handling process in the related art, such as Figure 1 As shown: Most systems on the market are still at the basic data analysis level, lacking the ability of deep mining and dynamic strategy update. Most of them are triggered when the fault threshold is reached to determine whether it is a national standard fault (the national regulations stipulate that the new energy-pure electric vehicle-alarm type GB-T32960 on-board energy storage system has multiple safety alarm mechanisms, a total of 19 items). If it is judged to be yes, the fault information is pushed and handed over to after-sales for processing. Other monitored fault information that does not belong to the national standard fault is not processed. The vehicle fault handling method in the relevant technology cannot meet the panoramic monitoring needs of automobile companies.

[0028] In the relevant technologies, the nationally stipulated new energy - pure electric vehicle - alarm type GB-T32960 on-board energy storage system has multiple safety alarm mechanisms, most of which push fault information to the platform after a fault occurs, and cannot meet the needs of early warning of potential faults; with the increasing amount of vehicle networking data, traditional detection methods have been unable to cope with diversified needs. Traditional methods are difficult to conduct comprehensive and accurate detection in terms of unified fault detection thresholds across multiple models and unique fault detection of a certain model; it can be seen that the vehicle fault handling method in the relevant technology has the technical problem of being unable to accurately predict potential vehicle faults and handle vehicle faults in a timely manner.

[0029] In order to at least partially solve the above problems, the present invention provides a vehicle fault handling method; the vehicle fault handling method of the present invention can be applied to the scenario of predicting potential vehicle faults and handling vehicle faults.

[0030] The vehicle fault handling method of the present invention can be executed by the vehicle, specifically, it can be executed by an ECU (Electronic Control Unit) arranged on the vehicle. The ECU is also called a driving computer, an on-board computer, etc. It is one of the core electronic components of modern automobiles. It is equivalent to the brain nerve center system of the automobile, responsible for controlling various systems of the vehicle to ensure the normal operation and performance optimization of the vehicle.

[0031] Figure 2 It is one of the flow charts of the vehicle fault handling method provided by the present invention, such as Figure 2 As shown, including but not limited to the following steps: Step 201, obtaining multiple vehicle feature information.

[0032] In this embodiment, multiple vehicle feature information of multiple vehicle features can be obtained through the ECU; specifically, the vehicle's ECU collects a series of multiple vehicle feature information corresponding to the multiple vehicle features at the current moment through other sensors.

[0033] Optionally, multiple vehicle characteristic information includes but is not limited to message time, vehicle status, charging status, vehicle speed, total voltage, total current, battery cell maximum voltage value, battery cell minimum voltage value, maximum temperature value, minimum temperature value, alarm data, drive motor temperature, insulation resistance, brake pedal travel value, backup battery power, etc. Each characteristic represents the current status or performance of a certain aspect of the vehicle.

[0034] Optionally, the vehicle fault handling method of the present invention can also be executed by a remote server, and multiple vehicle feature information of the vehicle can be uploaded to the remote server through T-BOX (Telematics Box). Here, T-BOX is the core component of the vehicle telematics system, which realizes data exchange between the vehicle and the cloud platform through a wireless communication network (such as 4G / 5G, GPS, etc.).

[0035] Step 202 , weighting processing is performed on multiple vehicle feature information to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight.

[0036] A multi-feature weighted feature extraction method is used to characterize fault features. Here, the ECU will perform weighted processing on the collected feature information. Each feature will be assigned a weight value (i.e., feature weight), which reflects the importance or influence of the feature in predicting faults. Through weighted processing, the system can focus more on those features that are more critical to fault prediction, thereby improving the accuracy of the prediction.

[0037] Specifically, the weighted feature information can be calculated by the following formula:

[0038] Here, n is the number of features.

[0039] Step 203, inputting the weighted feature information into a gradient boosting prediction model to obtain predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on feature information samples and their corresponding fault information labels.

[0040] Gradient Boosting Model, especially in the field of machine learning and data science, usually refers to an ensemble learning algorithm that builds a powerful prediction model by combining multiple weak learners (such as decision trees). This model performs well in solving prediction problems such as classification and regression, and is one of the commonly used algorithms in data science competitions and practical applications.

[0041] In this embodiment, the system inputs weighted feature information into a pre-trained gradient boosting prediction model, which outputs predicted fault information based on the input feature information. The predicted fault information may include fault type, fault code, probability of fault occurrence, or severity of fault occurrence, etc.

[0042] Step 204: Send the predicted fault information to the target server for fault diagnosis, and obtain the fault diagnosis strategy returned by the target server.

[0043] The system sends the predicted fault information to the target server (which can be a cloud server or a remote data center) for more detailed fault diagnosis and returns a fault diagnosis strategy; on the target server, more complex algorithms or models are run to further analyze the predicted fault information and generate a corresponding fault diagnosis strategy.

[0044] The fault diagnosis strategy may include maintenance suggestions, fault repair steps or emergency treatment measures, etc. Specifically, after the fault diagnosis strategy is sent back to the vehicle, the relevant maintenance personnel can perform maintenance operations, and the vehicle can also perform corresponding fault repairs for predicted faults through OTA (Over-The-Air) upgrades.

[0045] Optionally, after the predicted fault information is sent to the target server for fault diagnosis, the target server automatically determines the fault priority and pushes it to after-sales service, which contacts the user or dealer to deal with the faulty vehicle in a timely manner. At this time, the fault diagnosis strategy is to send the corresponding fault resolution strategy back to the vehicle.

[0046] Through the embodiments provided by the present application, multiple vehicle feature information is obtained; multiple vehicle feature information is weighted to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; the weighted feature information is input into a gradient boosting prediction model to obtain predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on feature information samples and their corresponding fault information labels; the predicted fault information is sent to a target server for fault diagnosis, and a fault diagnosis strategy returned by the target server is obtained, which can solve the technical problem that the vehicle fault handling method in the related technology cannot accurately predict potential vehicle faults and handle vehicle faults in a timely manner, thereby improving the reliability of the vehicle fault handling method.

[0047] As an optional solution, before performing weighted processing on the plurality of vehicle feature information and determining the weighted feature information, the method further includes: S11, inputting a plurality of vehicle features into a target evaluation model to obtain a plurality of feature importance scores output by the target evaluation model, wherein the target evaluation model is any one of a decision tree model, a random forest model, and a gradient boosting model, and the plurality of feature importance scores correspond to the plurality of vehicle features respectively; S21, determining multiple feature weights based on multiple feature importance scores.

[0048] In this embodiment, the target evaluation model may include any one of a decision tree model, a random forest model, and a gradient boosting machine model to evaluate the feature importance score corresponding to the vehicle feature.

[0049] Decision tree model: The decision tree splits the data through a series of judgment rules and calculates the information gain or Gini impurity reduction of each feature in the segmentation process to evaluate the importance of the feature.

[0050] Random Forest Model: Random Forest is an ensemble model consisting of multiple decision trees that assesses the overall importance of features by averaging the feature importance scores of each tree, which generally provides more stable and reliable scores.

[0051] Gradient boosting machine model: During the training process, the gradient boosting machine calculates the contribution of each feature in reducing the loss function to evaluate the importance of the feature.

[0052] Specifically, inputting a plurality of vehicle features into a target evaluation model and obtaining a plurality of feature importance scores output by the target evaluation model may include the following steps: 1. Data preparation: First, you need to collect and organize a data set containing multiple vehicle features. These features may include the vehicle’s brand, model, year, mileage, motor power, fuel consumption, safety rating, price, etc.

[0053] 2. Select the target evaluation model: Select a decision tree model, random forest model or gradient boosting machine model as the target evaluation model. These models are all tree-based, can handle classification and regression problems well, and provide feature importance evaluation methods.

[0054] 3. Model training: Use the prepared dataset to train the selected model. During the training process, the model learns how to map the input features (vehicle characteristics) to the output (such as the vehicle's value assessment, performance rating, etc.).

[0055] 4. Obtain feature importance scores: After training, the model will output the importance scores of each feature. These scores reflect the contribution of the feature in the prediction process of the gradient boosting prediction model, that is, the influence of the feature on the model prediction results.

[0056] Multiple weight values ​​are determined based on multiple feature importance scores. Since the feature importance scores obtained by different models or the same model on different data sets may not be on the same scale, these scores usually need to be normalized so that they are in the same numerical range (such as between 0 and 1).

[0057] Specifically, the feature importance scores can be normalized to the [0, 1] interval using the following formula to ensure that they are comparable to each other:

[0058] Among them, the target normalized score is the normalized score corresponding to the target feature importance score, the minimum score is the minimum value among multiple feature importance scores, and the maximum score is the maximum value among multiple feature importance scores.

[0059] Furthermore, the normalized feature importance scores can be directly used as weight values, or these scores can be further transformed according to specific rules to obtain weight values.

[0060] Through this embodiment, multiple feature importance scores are determined through a target evaluation model, and multiple weight values ​​are determined based on the multiple feature importance scores, which can effectively improve the accuracy of the multiple weight values.

[0061] As an optional solution, the gradient boosting prediction model is an XGBoost prediction model.

[0062] The XGBoost (eXtreme Gradient Boosting) prediction model is an efficient, flexible and scalable machine learning algorithm. It is an efficient, flexible and scalable implementation of the Gradient Boosting Decision Tree (GBDT).

[0063] Through this embodiment, the accuracy of the obtained predicted fault information can be effectively improved.

[0064] As an optional solution, the XGBoost prediction model is obtained based on the following steps: S21, initialize the model parameters of the XGBoost prediction model; S22, obtaining a training data set; S23, train the XGBoost prediction model using the training data set; S24, using the model parameters obtained when the number of iterations reaches a preset number as the model parameters of the trained XGBoost prediction model.

[0065] In this embodiment, it is necessary to perform model training on the XGBoost prediction model.

[0066] Initialize the model parameters of the XGBoost prediction model. Here, you need to define the parameters of the XGBoost prediction model. These parameters will affect the training process and performance of the model. Optional parameters include but are not limited to: max_depth: The maximum depth of the tree. Increasing this value will make the model more complex and may also lead to overfitting. learning_rate (or eta): learning rate, which controls the contribution of each tree to the final result; n_estimators (or num_boost_round): the total number of trees to be built, which is actually a limit on the number of iterations; objective: specifies the learning task and the corresponding learning objective, such as 'reg:squarederror' for regression tasks and 'binary:logistic' for binary classification tasks; subsample: the sample ratio used to train each tree; colsample_bytree: the proportion of features sampled when building a tree; gamma (or min_split_loss): the minimum loss reduction required to split a node; reg_alpha and reg_lambda: weights of the L1 and L2 regularization terms, used to control the complexity of the model.

[0067] Get the training dataset. The data in the training dataset has been preprocessed and divided into feature matrices and label vectors.

[0068] It should be noted that the training data set is training data collected from multiple vehicles under the target model. Here, the target model refers to the model of the vehicle that executes the vehicle fault handling method.

[0069] Use the training data set to train the XGBoost prediction model; here, you need to specify the number of iterations, which is set in advance.

[0070] The model parameters obtained when the number of iterations reaches the preset number are used as the model parameters of the trained XGBoost prediction model; during the training process, the model parameters will be optimized to minimize the loss function. When the number of iterations reaches the preset value, the model parameters will be fixed and used for subsequent predictions.

[0071] Through this embodiment, by training the XGBoost prediction model, the complexity of model features can be simplified, model parameters can be optimized, the processing efficiency of the XGBoost prediction model for real-time data can be improved, and the accuracy of fault prediction can be improved.

[0072] As an optional solution, the predicted fault information is sent to the target server for fault diagnosis, and a fault diagnosis strategy returned by the target server is obtained, including: S31, sending the predicted fault information to the target server through the target diagnosis protocol; S32, obtaining a fault diagnosis strategy returned by the target server, wherein the fault diagnosis strategy is determined by the target server by querying a real vehicle fault management library based on predicted fault information.

[0073] Optionally, the target diagnostic protocol is the UDS (Unified Diagnostic Services) diagnostic protocol. The UDS diagnostic protocol is a widely used industry standard that defines diagnostic services and protocols for ECU and network communications. The UDS protocol allows external test equipment (such as diagnostic tools) to communicate with the vehicle ECU to read diagnostic information, clear fault codes, program and configure the ECU, etc.

[0074] The target server is equipped with a real vehicle fault management library, which is a system that centrally stores and manages vehicle fault information. It usually contains a large number of fault cases, diagnostic processes, maintenance guides and related technical documents. The real vehicle fault management library is used to support vehicle fault diagnosis.

[0075] The target server can be a cloud server or a remote data center. On the target server, more complex algorithms or models will be run to query the actual vehicle fault management library based on the predicted fault information to further analyze the predicted fault information and generate corresponding fault diagnosis strategies.

[0076] The fault diagnosis strategy may include instructing the vehicle to go to a maintenance station for hardware repair, or may include repairing the vehicle fault through intelligent remote diagnosis and OTA upgrade technology.

[0077] Through this embodiment, the predicted fault information is sent to the target server through the target diagnosis protocol, and the fault diagnosis strategy returned by the target server is obtained, which can improve the stability and accuracy of the obtained fault diagnosis strategy.

[0078] As an optional solution, after sending the predicted fault information to the target server for fault diagnosis and obtaining the fault diagnosis strategy returned by the target server, the method further includes: S41, receiving an over-the-air upgrade data packet sent by a target server, wherein the over-the-air upgrade data packet is determined by the target server based on a fault diagnosis strategy; S42, install the over-the-air upgrade data package.

[0079] In this embodiment, the over-the-air upgrade data packet refers to an OTA upgrade data packet. After the target server determines the fault diagnosis strategy, if it determines that the fault can be repaired based on OTA upgrade, the target server can send an over-the-air upgrade data packet to the vehicle; the vehicle can repair the fault by accepting the over-the-air upgrade data packet sent by the target server and installing the over-the-air upgrade data packet.

[0080] Through this embodiment, the repair of vehicle failures and the restoration of vehicle functions are not limited to hardware repairs at maintenance stations. Through intelligent remote diagnosis and OTA upgrade technology, users can have a more convenient and economical maintenance experience.

[0081] Figure 3 This is the second flow chart of the vehicle fault handling method provided by the present invention, see Figure 3 The vehicle fault handling method of the embodiment of the present application is described below in conjunction with an optional example. In this optional example, the following steps may be included: 1. From multiple dimensions such as different vehicle models and different users, a multi-feature weighted feature extraction method is used to characterize fault features and determine weighted feature information; here, multiple vehicle features can include Figure 3 The vehicle speed, single battery voltage and message time are shown.

[0082] 2. Build an XGBoost model, iterate the feature description of fault features multiple times, simplify the model feature complexity, optimize model parameters, improve the processing efficiency of real-time data, and improve the precision of the fault prediction model; after building the XGBoost model, use the XGBoost model to make predictions based on weighted feature information to obtain predicted fault information.

[0083] 3. With the UDS diagnostic protocol and real vehicle fault management library as the core, deeply integrate and develop vehicle remote diagnosis and OTA upgrade technology to achieve efficient and accurate remote diagnosis and fault repair, greatly reduce maintenance costs, bring more convenient and economical maintenance experience to car owners, and fully enable vehicle intelligence to make travel smarter and safer; that is, send the predicted fault information to the target server for fault diagnosis, and obtain the fault diagnosis strategy returned by the target server.

[0084] Optionally, after the predicted fault information is sent to the target server for fault diagnosis, the target server automatically determines the fault priority and pushes it to after-sales service, which contacts the user or dealer to deal with the faulty vehicle in a timely manner. At this time, the fault diagnosis strategy is to send the corresponding fault resolution strategy back to the vehicle.

[0085] It can be understood that the present invention is not limited to shallow data analysis. It combines big data technology to extract features from vehicle data and deeply mine the value of data. It uses existing data to build a fault prediction model and continuously optimize it to generate custom faults, providing more comprehensive assistance for vehicle safety. At the same time, the repair of vehicle faults and the restoration of vehicle functions are not limited to hardware repairs at maintenance stations. Through intelligent remote diagnosis and OTA upgrade technology, it brings car owners a more convenient and economical maintenance experience.

[0086] Figure 4 is a structural block diagram of the vehicle fault processing device provided by the present invention, such as Figure 4 As shown, including but not limited to the following modules: An acquisition module 401 is used to acquire multiple vehicle feature information; The first execution module 402 is used to perform weighted processing on multiple vehicle feature information to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; The second execution module 403 is used to input the weighted feature information into the gradient boosting prediction model to obtain the predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on the feature information samples and their corresponding fault information labels; The third execution module 404 is used to send the predicted fault information to the target server for fault diagnosis, and obtain the fault diagnosis strategy returned by the target server.

[0087] Through the embodiments of the present application, multiple vehicle feature information is obtained; multiple vehicle feature information is weighted to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; the weighted feature information is input into a gradient boosting prediction model to obtain predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on feature information samples and their corresponding fault information labels; the predicted fault information is sent to a target server for fault diagnosis, and a fault diagnosis strategy returned by the target server is obtained, which can solve the technical problem that the vehicle fault handling method in the related technology cannot accurately predict potential vehicle faults and handle vehicle faults in a timely manner, thereby improving the reliability of the vehicle fault handling method.

[0088] It should be noted that the vehicle fault handling device provided by the present invention can execute the vehicle fault handling method described in any of the above embodiments during specific operation, which will not be elaborated in this embodiment.

[0089] Figure 5 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the vehicle fault processing method, which includes: obtaining multiple vehicle feature information; performing weighted processing on the multiple vehicle feature information to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; inputting the weighted feature information into the gradient boosting prediction model to obtain the predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on the feature information sample and its corresponding fault information label; sending the predicted fault information to the target server for fault diagnosis, and obtaining the fault diagnosis strategy returned by the target server.

[0090] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0091] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the vehicle fault handling method provided by the above-mentioned embodiments, the method including: obtaining multiple vehicle feature information; performing weighted processing on multiple vehicle feature information to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; inputting the weighted feature information into a gradient boosting prediction model to obtain predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on feature information samples and their corresponding fault information labels; sending the predicted fault information to a target server for fault diagnosis, and obtaining a fault diagnosis strategy returned by the target server.

[0092] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the vehicle fault handling method provided by the above-mentioned embodiments, the method comprising: obtaining multiple vehicle feature information; performing weighted processing on the multiple vehicle feature information to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; inputting the weighted feature information into a gradient boosting prediction model to obtain predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on feature information samples and their corresponding fault information labels; sending the predicted fault information to a target server for fault diagnosis, and obtaining a fault diagnosis strategy returned by the target server.

[0093] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0094] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vehicle fault handling method, characterized in that: include: Obtain multiple vehicle feature information; Performing weighted processing on the plurality of vehicle feature information to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; Inputting the weighted feature information into a gradient boosting prediction model to obtain predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on feature information samples and their corresponding fault information labels; The predicted fault information is sent to a target server for fault diagnosis, and a fault diagnosis strategy returned by the target server is obtained.

2. The vehicle fault handling method according to claim 1, characterized in that: Before performing weighted processing on the plurality of vehicle feature information to determine weighted feature information, the method further includes: Inputting a plurality of vehicle features into a target evaluation model to obtain a plurality of feature importance scores output by the target evaluation model, wherein the target evaluation model is any one of a decision tree model, a random forest model, and a gradient boosting model, and the plurality of feature importance scores respectively correspond to the plurality of vehicle features; A plurality of feature weights are determined based on the plurality of feature importance scores.

3. The vehicle fault handling method according to claim 1, characterized in that: The gradient boosting prediction model is an XGBoost prediction model.

4. The vehicle fault handling method according to claim 3, characterized in that: The XGBoost prediction model is obtained based on the following steps: Initialize model parameters of the XGBoost prediction model; Get the training dataset; Using the training data set to train the XGBoost prediction model; The model parameters obtained when the number of iterations reaches a preset number are used as the model parameters of the trained XGBoost prediction model.

5. The vehicle fault handling method according to claim 1, characterized in that: The sending the predicted fault information to a target server for fault diagnosis and obtaining a fault diagnosis strategy returned by the target server includes: Sending the predicted fault information to the target server via a target diagnosis protocol; Obtain a fault diagnosis strategy returned by the target server, wherein the fault diagnosis strategy is determined by the target server by querying a real vehicle fault management library based on the predicted fault information.

6. The vehicle fault handling method according to any one of claims 1 to 5, characterized in that: After sending the predicted fault information to the target server for fault diagnosis and obtaining the fault diagnosis strategy returned by the target server, the method further includes: receiving an over-the-air upgrade data packet sent by the target server, wherein the over-the-air upgrade data packet is determined by the target server based on the fault diagnosis strategy; Install the over-the-air update package.

7. A vehicle fault handling device, characterized in that: include: An acquisition module, used for acquiring multiple vehicle feature information; A first execution module is used to perform weighted processing on the plurality of vehicle feature information to determine weighted feature information, wherein different vehicle feature information is generated after collecting different vehicle features, and each vehicle feature is preset with a different feature weight; A second execution module is used to input the weighted feature information into a gradient boosting prediction model to obtain predicted fault information output by the gradient boosting prediction model, wherein the gradient boosting prediction model is obtained after training based on feature information samples and their corresponding fault information labels; The third execution module is used to send the predicted fault information to the target server for fault diagnosis, and obtain the fault diagnosis strategy returned by the target server.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the vehicle fault handling method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the vehicle fault handling method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the vehicle fault handling method according to any one of claims 1 to 6 is implemented.

Images