Vehicle maintenance method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of intelligent traffic and vehicle engineering, in particular to a vehicle maintenance method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out the fault prediction of a vehicle based on a preset fault prediction model according to a preliminary anomaly detection result, and recognizing the fault maintenance priority of the vehicle according to a fault prediction result, the method comprises the steps of adjusting a fault maintenance priority by using a reinforcement learning algorithm according to a vehicle driving road condition, user driving data and a vehicle part inventory state, generating a dynamic maintenance strategy of a vehicle according to the adjusted fault maintenance priority, obtaining a recommended maintenance network point of the vehicle based on the dynamic maintenance strategy, and maintaining the vehicle according to the recommended maintenance network point. Therefore, the problems of regular maintenance limitation, fault diagnosis hysteresis quality, unreasonable resource allocation and the like in the related technology are solved, and through multi-source data fusion, a deep learning model and a dynamic maintenance strategy, the maintenance efficiency and the fault prediction accuracy are improved, and resource allocation is optimized.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation and vehicle engineering technology, and in particular to a vehicle maintenance method, device, electronic device and storage medium. Background Technology

[0002] With the development and progress of the automotive industry, especially the increasing maturity of artificial intelligence, the Internet of Things and big data analysis technologies, it has become possible to manage and maintain vehicles in a more intelligent and refined manner.

[0003] In related technologies, vehicle maintenance technology relies on fixed-cycle maintenance, human experience, or simple sensor threshold alarms, which makes it impossible to accurately identify the actual wear and tear of parts and to predict latent faults in advance. In addition, because data from vehicle sensors, OBD (On-Board Diagnostics) systems, and maintenance records are stored in a scattered manner and lack a unified analysis framework, it is also difficult to uncover potential correlations between different data sources, thus limiting a comprehensive understanding of the vehicle's health status, which urgently needs to be addressed. Summary of the Invention

[0004] This application provides a vehicle maintenance method, apparatus, electronic device, and storage medium to address the limitations of periodic maintenance, delayed fault diagnosis, and unreasonable resource allocation in related technologies.

[0005] The first aspect of this application provides a vehicle maintenance method, including the following steps: Obtain preliminary anomaly detection results for the vehicle; Based on the preset fault prediction model, the vehicle is fault predicted according to the preliminary anomaly detection results, and the fault prediction result of the vehicle is obtained. Based on the fault prediction results, the fault maintenance priority of the vehicle is identified, and the fault maintenance priority is adjusted using a reinforcement learning algorithm according to the vehicle's driving conditions, user driving data, and vehicle parts inventory status. A dynamic maintenance strategy for the vehicle is generated based on the adjusted fault maintenance priority. Recommended maintenance outlets for the vehicle are obtained based on the dynamic maintenance strategy, and the vehicle is maintained according to the recommended maintenance outlets.

[0006] According to one embodiment of this application, the step of generating a dynamic maintenance strategy for the vehicle based on the adjusted fault maintenance priority, and obtaining recommended maintenance outlets for the vehicle based on the dynamic maintenance strategy, includes: When the adjusted fault maintenance priority is high priority, the real-time running location of the vehicle is obtained, and a first maintenance strategy corresponding to the high priority is generated based on the real-time running location of the vehicle and the faulty parts of the vehicle, so as to obtain the recommended maintenance outlets for the vehicle based on the first maintenance strategy. When the adjusted fault maintenance priority is medium priority, a second maintenance strategy corresponding to the medium priority is generated based on the faulty parts of the vehicle, so as to obtain the recommended maintenance outlets for the vehicle according to the second maintenance strategy; When the adjusted fault maintenance priority is low, a third maintenance strategy corresponding to the low priority is generated based on the vehicle faulty parts and the preset maintenance cycle of the vehicle faulty parts, so as to obtain the recommended maintenance outlets for the vehicle according to the third maintenance strategy.

[0007] According to one embodiment of this application, the step of predicting a vehicle's fault based on a preset fault prediction model and the preliminary anomaly detection results includes: Obtain the historical maintenance results of the vehicle; Based on the historical maintenance results and the preliminary anomaly detection results, multi-source data fusion is performed to construct the vehicle's health knowledge graph; Based on the preset fault prediction model, the vehicle's faults are predicted according to the health knowledge graph, and the fault prediction results of the vehicle are obtained.

[0008] According to one embodiment of this application, obtaining the preliminary anomaly detection results of the vehicle includes: The system acquires vehicle operation data, vehicle driving environment data, and user behavior data, wherein the vehicle operation data, vehicle driving environment data, and user behavior data are collected by multiple on-board sensors. The vehicle operation data, the vehicle driving environment data, and the user behavior data are preprocessed, and key features are extracted from the preprocessed vehicle operation data, vehicle driving environment data, and user behavior data to obtain target data features. The vehicle is subjected to preliminary anomaly detection based on a preset rule engine and the target data features, and the preliminary anomaly detection results are obtained.

[0009] According to one embodiment of this application, the step of preprocessing the vehicle operation data, the vehicle driving environment data, and the user behavior data, and extracting key features from the preprocessed vehicle operation data, vehicle driving environment data, and user behavior data to obtain target data features, includes: Abnormal data is removed from the vehicle operation data, the vehicle driving environment data, and the user behavior data to obtain the target data of the vehicle, and the target data is then normalized. The time-series features of the vehicle operation data, the image features of the vehicle driving environment, and the user behavior features of the user behavior data are extracted from the normalized target data to obtain the target data features based on the time-series features, the image features, and the user behavior features.

[0010] According to one embodiment of this application, identifying the fault maintenance priority of the vehicle based on the fault prediction result includes: The fault prediction results are analyzed based on a preset deep learning model to obtain the fault level, service life, maintenance cost and user needs of the vehicle faulty parts. Based on the fault level, service life, maintenance cost, and user needs, a comprehensive scoring strategy is used to determine the vehicle's fault maintenance priority.

[0011] The vehicle maintenance method according to embodiments of this application, based on a preset fault prediction model, predicts vehicle faults based on preliminary anomaly detection results, identifies vehicle fault maintenance priorities based on the fault prediction results, adjusts the fault maintenance priorities using a reinforcement learning algorithm based on vehicle driving conditions, user driving data, and vehicle parts inventory status, generates a dynamic maintenance strategy for the vehicle based on the adjusted fault maintenance priorities, obtains recommended maintenance outlets for the vehicle based on the dynamic maintenance strategy, and performs maintenance on the vehicle according to the recommended maintenance outlets. This solves the problems of limitations in periodic maintenance, delayed fault diagnosis, and unreasonable resource allocation in related technologies. Through multi-source data fusion, deep learning models, and dynamic maintenance strategies, it improves maintenance efficiency and fault prediction accuracy, and optimizes resource allocation.

[0012] A second aspect of this application provides a vehicle maintenance device, comprising: The acquisition module is used to acquire preliminary anomaly detection results for the vehicle. The fault prediction module is used to predict the fault of the vehicle based on the preliminary anomaly detection results according to the preset fault prediction model, and to obtain the fault prediction result of the vehicle. The adjustment module is used to identify the fault maintenance priority of the vehicle based on the fault prediction result, and adjust the fault maintenance priority according to the vehicle driving conditions, user driving data, and vehicle parts inventory status using a reinforcement learning algorithm. The maintenance module is used to generate a dynamic maintenance strategy for the vehicle based on the adjusted fault maintenance priority, obtain recommended maintenance outlets for the vehicle based on the dynamic maintenance strategy, and perform maintenance on the vehicle according to the recommended maintenance outlets.

[0013] According to one embodiment of this application, the maintenance module includes: The first generation unit is used to obtain the real-time running location of the vehicle when the adjusted fault maintenance priority is high priority, and generate a first maintenance strategy corresponding to the high priority based on the real-time running location of the vehicle and the faulty parts of the vehicle, so as to obtain the recommended maintenance outlets for the vehicle based on the first maintenance strategy. The second generation unit is used to generate a second maintenance strategy corresponding to the medium priority based on the faulty parts of the vehicle when the adjusted fault maintenance priority is medium priority, so as to obtain the recommended maintenance outlets for the vehicle according to the second maintenance strategy. The third generation unit is used to generate a third maintenance strategy corresponding to the low priority based on the vehicle faulty parts and the preset maintenance cycle of the vehicle faulty parts when the adjusted fault maintenance priority is low priority, so as to obtain the recommended maintenance outlets for the vehicle according to the third maintenance strategy.

[0014] According to one embodiment of this application, the fault prediction module includes: The first acquisition unit is used to acquire the historical maintenance results of the vehicle; The data fusion unit is used to perform multi-source data fusion based on the historical maintenance results and the preliminary anomaly detection results to construct the vehicle's health knowledge graph. The fault prediction unit is used to predict the faults of the vehicle based on the preset fault prediction model and the health knowledge graph, and to obtain the fault prediction result of the vehicle.

[0015] According to one embodiment of this application, the acquisition module includes: The second acquisition unit is used to acquire vehicle operation data, vehicle driving environment data and user behavior data, wherein the vehicle operation data, vehicle driving environment data and user behavior data are collected by multiple vehicle-mounted sensors. The feature extraction unit is used to preprocess the vehicle operation data, the vehicle driving environment data, and the user behavior data, and to extract key features from the preprocessed vehicle operation data, vehicle driving environment data, and user behavior data to obtain target data features. An anomaly detection unit is used to perform preliminary anomaly detection on the vehicle based on a preset rule engine and the target data features, and to obtain the preliminary anomaly detection result.

[0016] According to one embodiment of this application, the feature extraction unit includes: The normalization processing subunit is used to remove abnormal data from the vehicle operation data, the vehicle driving environment data, and the user behavior data to obtain the target data of the vehicle, and to perform normalization processing on the target data. The feature extraction subunit is used to extract the temporal features of the vehicle operation data, the image features of the vehicle driving environment, and the user behavior features of the user behavior data from the normalized target data, so as to obtain the target data features based on the temporal features, the image features, and the user behavior features.

[0017] According to one embodiment of this application, the adjustment module includes: The analysis unit is used to analyze the fault prediction results based on a preset deep learning model to obtain the fault level, service life, maintenance cost and user needs of the vehicle faulty parts. The third acquisition unit is used to obtain the vehicle's fault maintenance priority based on the fault level, service life, maintenance cost, and user needs using a comprehensive scoring strategy.

[0018] The vehicle maintenance device according to embodiments of this application, based on a preset fault prediction model, predicts vehicle faults based on preliminary anomaly detection results, identifies the vehicle's fault maintenance priority based on the fault prediction results, adjusts the fault maintenance priority using a reinforcement learning algorithm based on vehicle driving conditions, user driving data, and vehicle parts inventory status, generates a dynamic maintenance strategy for the vehicle based on the adjusted fault maintenance priority, obtains recommended maintenance outlets for the vehicle based on the dynamic maintenance strategy, and performs maintenance on the vehicle according to the recommended maintenance outlets. This solves the problems of limited periodic maintenance, delayed fault diagnosis, and unreasonable resource allocation in related technologies. Through multi-source data fusion, deep learning models, and dynamic maintenance strategies, it improves maintenance efficiency and fault prediction accuracy, and optimizes resource allocation.

[0019] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the vehicle maintenance method as described in the above embodiments.

[0020] A fourth aspect of this application provides a computer-readable storage medium storing computer instructions for causing the computer to perform a vehicle maintenance method as described in the above embodiments.

[0021] A fifth aspect of this application provides a computer program product, including a computer program that is executed to implement the vehicle maintenance method described in the above embodiments.

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

[0023] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating a vehicle maintenance method according to an embodiment of this application; Figure 2 This is a schematic diagram of a system architecture according to an embodiment of this application; Figure 3 This is a flowchart illustrating the multi-source data fusion process according to one embodiment of this application; Figure 4 This is a diagram illustrating the architecture of a fault prediction model according to an embodiment of this application. Figure 5 A logic diagram is generated for a dynamic maintenance strategy according to one embodiment of this application; Figure 6 This is a block diagram illustrating a vehicle maintenance device according to an embodiment of this application; Figure 7 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application.

[0024] Among them, 10-vehicle maintenance device, 100-acquisition module, 200-fault prediction module, 300-adjustment module, and 400-maintenance module. Detailed Implementation

[0025] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0026] The following description, with reference to the accompanying drawings, describes a vehicle maintenance method, apparatus, electronic device, and storage medium according to embodiments of this application. Addressing the limitations of periodic maintenance, delayed fault diagnosis, and unreasonable resource allocation in related technologies mentioned in the background, this application provides a vehicle maintenance method. In this method, based on a preset fault prediction model, vehicle faults are predicted according to preliminary anomaly detection results. The vehicle's fault maintenance priority is identified based on the fault prediction results. The fault maintenance priority is adjusted using a reinforcement learning algorithm based on vehicle driving conditions, user driving data, and vehicle parts inventory status. A dynamic maintenance strategy for the vehicle is generated based on the adjusted fault maintenance priority. Recommended maintenance service points for the vehicle are obtained based on the dynamic maintenance strategy, and maintenance is performed on the vehicle according to the recommended service points. This solves the problems of periodic maintenance limitations, delayed fault diagnosis, and unreasonable resource allocation in related technologies. Through multi-source data fusion, deep learning models, and dynamic maintenance strategies, maintenance efficiency and fault prediction accuracy are improved, and resource allocation is optimized.

[0027] Specifically, Figure 1 This is a schematic flowchart illustrating a vehicle maintenance method provided in an embodiment of this application.

[0028] like Figure 1 As shown, the maintenance method for this vehicle includes the following steps: In step S101, preliminary anomaly detection results of the vehicle are obtained.

[0029] In step S102, based on the preset fault prediction model, the vehicle fault is predicted according to the preliminary anomaly detection results to obtain the vehicle fault prediction result.

[0030] In step S103, the vehicle's fault maintenance priority is identified based on the fault prediction results, and the fault maintenance priority is adjusted using a reinforcement learning algorithm based on the vehicle's driving conditions, user driving data, and vehicle parts inventory status.

[0031] In step S104, a dynamic maintenance strategy for the vehicle is generated based on the adjusted fault maintenance priority, a recommended maintenance site for the vehicle is obtained based on the dynamic maintenance strategy, and the vehicle is maintained according to the recommended maintenance site.

[0032] The preset fault prediction model can be set by those skilled in the art according to the actual intelligent maintenance and management needs, and no specific limitations are made here.

[0033] Specifically, such as Figure 2As shown, the system architecture involved in this application embodiment mainly includes a perception layer, an edge computing layer, and a cloud platform layer. In the perception layer, various types of vehicle-mounted sensors (such as vibration sensors, temperature sensors, cameras, OBD interfaces, etc.) are deployed to collect vehicle operation data, vehicle driving environment data, and user behavior data in real time. In the edge computing layer, vehicle-mounted edge computing nodes perform data preprocessing on the collected vehicle operation data, vehicle driving environment data, and user behavior data, such as cleaning, feature extraction, and preliminary anomaly detection, to reduce cloud transmission pressure. The cloud platform layer mainly includes a data fusion module and an AI (Artificial Intelligence) model layer. This allows the data fusion module to integrate multi-source heterogeneous data (such as sensor data, historical maintenance results, road condition information, etc.) to construct a vehicle health knowledge graph. The AI ​​model layer then uses fault prediction models (such as LSTM (Long Short-Term Memory) and CNN (Convolutional Neural Network) to perform these models. Deep learning architectures such as convolutional neural networks (CNNs) are used to learn features from time-series data, predict the remaining lifespan of components, and then the model is optimized through maintenance strategies. Combined with user driving data, vehicle parts inventory status, and service network distribution, a dynamic maintenance strategy for the vehicle is generated. Finally, through the user interaction module, the mobile APP or in-vehicle display provides real-time vehicle health reports, maintenance reminders, and repair service appointments. These are pushed to the user through the APP and synchronized to the service network dispatch system.

[0034] Therefore, this application combines three major technological inventions—multi-source data fusion algorithm, deep learning fault prediction model, and dynamic maintenance strategy generation—to achieve comprehensive monitoring and accurate prediction of vehicle health status, significantly improving maintenance efficiency, reducing maintenance costs, and optimizing resource allocation. The main features include using a multi-source data fusion algorithm to break down data silos, enabling a comprehensive assessment of vehicle health status and providing a solid foundation for subsequent fault prediction; utilizing a deep learning fault prediction model to improve the accuracy of fault prediction; and generating personalized dynamic maintenance strategies based on fault prediction results, vehicle driving conditions, user driving data, and vehicle parts inventory status, ensuring optimal maintenance results with limited resources. Specific implementation methods are described below.

[0035] Step S101 includes the following aspects: According to one embodiment of this application, obtaining preliminary anomaly detection results for a vehicle includes: acquiring vehicle operation data, vehicle driving environment data, and user behavior data, wherein the vehicle operation data, vehicle driving environment data, and user behavior data are collected by multiple on-board sensors; performing data preprocessing on the vehicle operation data, vehicle driving environment data, and user behavior data, and extracting key features from the preprocessed vehicle operation data, vehicle driving environment data, and user behavior data to obtain target data features; and performing preliminary anomaly detection on the vehicle based on a preset rule engine and target data features to obtain preliminary anomaly detection results.

[0036] According to one embodiment of this application, vehicle operation data, vehicle driving environment data, and user behavior data are preprocessed, and key features are extracted from the preprocessed vehicle operation data, vehicle driving environment data, and user behavior data to obtain target data features. This includes: removing abnormal data from the vehicle operation data, vehicle driving environment data, and user behavior data to obtain target data for the vehicle, and normalizing the target data; extracting time-series features of the vehicle operation data, image features of the vehicle driving environment, and user behavior features of the user behavior data from the normalized target data to obtain target data features based on time-series features, image features, and user behavior features.

[0037] The preset rule engine can be set by those skilled in the art according to actual intelligent maintenance and management needs, and no specific limitations are made here.

[0038] Specifically, such as Figure 3 As shown, firstly, in the perception layer of the system architecture of this application embodiment, multiple types of vehicle-mounted sensors are deployed, such as vibration sensors, temperature sensors, tire pressure sensors, cameras, OBD interfaces, etc., to collect vehicle operation data (such as engine speed, coolant temperature), vehicle driving environment data (such as tire pressure, vehicle body damage images, etc.) and user behavior data (such as rapid acceleration frequency) in real time, and transmit the collected data to the edge computing layer for preliminary processing.

[0039] Secondly, after receiving vehicle operation data, vehicle driving environment data, and user behavior data transmitted by the perception layer, the edge computing layer performs data preprocessing on these data. For example, it removes abnormal data (such as sensor transient noise) from the vehicle operation data, vehicle driving environment data, and user behavior data, and performs noise reduction processing on these data through edge computing nodes to obtain the target data of the vehicle. Then, it performs normalization processing on the target data (for example, scaling temperature, voltage, and other data to a uniform range of [0, 1]) to ensure data quality.

[0040] Again, such as Figure 4As shown, the Fast Fourier Transform is used to extract key features from vehicle operation data, vehicle driving environment data, and user behavior data. Then, based on deep learning architectures such as LSTM and CNN, the normalized target data is used to learn features of time series features. For example, LSTM is used to extract time series features of vehicle operation data (such as engine vibration and battery voltage), and CNN layers are used to extract image features of the vehicle driving environment (such as tire wear images) and user behavior features of user behavior data. The target data features are obtained based on time series features, image features, and user behavior features.

[0041] Finally, based on the preset rule engine (e.g., triggering an alarm when tire pressure is below a threshold) and target data features, the vehicle is initially detected for anomalies, and the initial anomaly detection results are obtained. For example, when the tire pressure is below a threshold (e.g., 1.8 Bar), a local alarm is triggered; when the engine temperature exceeds the safe range, a power reduction operation suggestion is given.

[0042] Furthermore, in order to understand important information about the vehicle's past status and to help accurately assess the vehicle's current health status, this application embodiment also needs to extract the vehicle's historical maintenance records from the historical maintenance database through the API (Application Programming Interface) interface. These records may include the time of each service, the type of service performed (such as changing engine oil or brake pads), the problems found and their solutions, etc., thereby obtaining the vehicle's historical maintenance results and preparing for subsequent multi-source data fusion.

[0043] Step S102 includes the following aspects: According to one embodiment of this application, based on a preset fault prediction model, a vehicle fault is predicted based on preliminary anomaly detection results, including: obtaining the vehicle's historical maintenance results; performing multi-source data fusion based on the historical maintenance results and preliminary anomaly detection results to construct a vehicle health knowledge graph; and predicting vehicle faults based on the preset fault prediction model and the health knowledge graph to obtain the vehicle fault prediction result.

[0044] Specifically, in order to achieve semantic alignment and unified representation of multi-source heterogeneous data and solve the data silo problem, this application embodiment requires multi-source data fusion based on historical maintenance results and preliminary anomaly detection results to construct a vehicle health knowledge graph, which is then transmitted to the cloud platform layer for analysis. This allows for a more accurate reflection of the vehicle's overall health status, providing a solid foundation for subsequent fault prediction and maintenance strategy generation. At the same time, by constructing the vehicle health knowledge graph, various advanced analysis techniques (such as deep learning and reinforcement learning) can be easily applied to perform fault prediction, anomaly detection, and maintenance optimization for the vehicle. Furthermore, it helps the fault prediction model better understand the correlation between different data, thereby improving the accuracy and efficiency of the analysis.

[0045] Step S103 includes the following aspects: According to one embodiment of this application, based on the fault prediction results, the fault maintenance priority of a vehicle is identified, including: analyzing the fault prediction results based on a preset deep learning model to obtain the fault level, service life, repair cost, and user needs of the faulty parts of the vehicle; and obtaining the fault maintenance priority of the vehicle using a comprehensive scoring strategy based on the fault level, service life, repair cost, and user needs.

[0046] The preset deep learning model can be set by those skilled in the art according to actual intelligent maintenance and management needs, and no specific limitations are made here.

[0047] Specifically, embodiments of this application can generate dynamic maintenance strategies for vehicles based on the vehicle health knowledge graph constructed above, combined with vehicle driving conditions, user driving data, and vehicle parts inventory status, to achieve optimal resource allocation. For example, if the prediction results indicate that a critical component is about to malfunction, but the current vehicle parts inventory is insufficient, procurement can be arranged in advance to avoid service delays caused by waiting for parts.

[0048] Specifically, such as Figure 5 As shown, firstly, this embodiment of the application utilizes the AI ​​model layer in the cloud platform layer to perform fault prediction and obtain fault prediction results. That is, it performs fault prediction on the vehicle based on the preset fault prediction model in the AI ​​model layer. For example, the LSTM model can be used to predict the wear degree of the engine piston rings and give the corresponding suggested replacement time. The LSTM captures the long-term dependency relationship of time series data such as engine vibration and battery voltage. The YOLOv5 model analyzes the images of the vehicle camera to detect visual faults such as tire wear and body damage, so as to predict the remaining service life of key components.

[0049] Secondly, based on the fault prediction results, the vehicle's fault maintenance priority is identified. In this embodiment, the fault prediction results can be analyzed based on a preset deep learning model. For example, the remaining lifespan of the components (e.g., 200 hours remaining for the engine), the fault probability score (e.g., 85% probability of transmission failure), and the visual inspection results (e.g., tire crack length ≥ 5mm) can be predicted by the preset fault prediction model. Thus, the fault level, lifespan, and maintenance cost of the vehicle's faulty parts can be obtained. At the same time, user needs can be obtained based on the user's driving habits (frequency of rapid acceleration / braking). Therefore, the vehicle's fault maintenance priority can be obtained by using a comprehensive scoring strategy based on the fault level, lifespan, maintenance cost, and user needs.

[0050] In this application embodiment, a comprehensive score can be achieved using standardized scoring and threshold triggering. For example, standardized scoring may include: fault level (weight 40%) + service life (weight 30%) + maintenance cost (weight 20%) + user needs (weight 10%). Threshold triggering may include: total score > 7 points, immediate maintenance recommended (e.g., brake system failure); 5-7 points, maintenance recommended within 7 days (e.g., tire wear); < 5 points, delayed maintenance recommended (e.g., headlight failure).

[0051] For example, if engine piston ring wear is detected, according to the standardized scoring: fault level 8 points + service life (30 hours) 9 points + maintenance cost 7 points + user needs (no plan) 3 points, the total score = 8×0.4 + 9×0.3 + 7×0.2 + 3×0.1 = 7.6 points > 7 points. Therefore, its fault maintenance priority is high priority, and immediate maintenance is recommended.

[0052] Finally, based on vehicle driving conditions, user driving data, and vehicle parts inventory status, reinforcement learning algorithms can be used to dynamically adjust fault maintenance priorities. For example, if the initial maintenance priority is engine > tires > transmission, then based on user driving data (e.g., frequent emergency braking may lead to a higher transmission maintenance priority) and vehicle driving conditions (e.g., uneven roads may lead to a higher tire maintenance priority), reinforcement learning algorithms can be used to dynamically adjust fault maintenance priorities. Then, based on vehicle driving conditions, user driving data, and vehicle parts inventory status (e.g., brake pad inventory and delivery cycle), a dynamic maintenance strategy for the vehicle can be generated. Thus, by dynamically adjusting fault maintenance priorities, faults that may lead to serious accidents (e.g., brake system failure, steering failure, etc.) can be dealt with first, thereby reducing the risk of sudden failures.

[0053] Step S104 includes the following aspects: A dynamic maintenance strategy for the vehicle is generated based on the adjusted fault maintenance priority. Recommended maintenance locations for the vehicle are then determined based on this strategy. This includes: when the adjusted fault maintenance priority is high, obtaining the vehicle's real-time operating location and generating a first maintenance strategy corresponding to the high priority based on the vehicle's real-time operating location and the faulty parts, thereby determining recommended maintenance locations; when the adjusted fault maintenance priority is medium, generating a second maintenance strategy corresponding to the medium priority based on the faulty parts, thereby determining recommended maintenance locations; and when the adjusted fault maintenance priority is low, generating a third maintenance strategy corresponding to the low priority based on the faulty parts and their preset maintenance cycles, thereby determining recommended maintenance locations.

[0054] The preset maintenance cycle can be set by those skilled in the art based on the actual maintenance needs of the vehicle, and no specific limitations are made here.

[0055] Specifically, since the cloud platform layer of this application embodiment can monitor the vehicle status in real time, an early warning will be triggered and an emergency response process will be initiated when the vehicle is in an abnormal state. At this time, repair outlets can be selected according to distance, and the repair outlets can prepare vehicle-related parts in advance according to the maintenance plan and provide door-to-door service or appointment for work.

[0056] Specifically, when the adjusted fault maintenance priority is high priority, it indicates that the vehicle's faulty parts have malfunctioned and require immediate inspection or repair; otherwise, it will affect driving safety, such as the engine being about to fail. Therefore, when the adjusted fault maintenance priority is high priority, it is first necessary to obtain the vehicle's real-time operating location and generate the first maintenance strategy corresponding to the high priority based on the vehicle's real-time operating location and the faulty parts. In other words, it is necessary to generate the optimal recommended maintenance location for the vehicle based on the current fault status of the vehicle, the current inventory of vehicle parts, and the usage of maintenance outlets. Thus, when the vehicle travels to the recommended maintenance outlet, it can be inspected or repaired.

[0057] Optionally, when the adjusted fault maintenance priority is medium priority, it indicates that the vehicle's faulty parts have potential for failure, but will not affect the normal use of the vehicle. For example, the brake pads are worn close to the critical value. Therefore, when the adjusted fault maintenance priority is medium priority, a second maintenance strategy corresponding to the medium priority is generated. That is, based on the wear condition of the vehicle's faulty parts, it is recommended to check or maintain them within the next X kilometers or Y days, and the optimal recommended maintenance outlet is generated for the vehicle.

[0058] Optionally, when the adjusted fault maintenance priority is low, it means that the faulty parts of the vehicle have not yet shown any potential for failure, but are close to their maintenance cycle. Therefore, when the adjusted fault maintenance priority is low, a third maintenance strategy corresponding to the low priority is generated based on the faulty parts of the vehicle and their preset maintenance cycle (e.g., one month). Thus, when the adjusted fault maintenance priority is low, the optimal recommended maintenance outlets will be generated for the vehicle based on the usage of the faulty parts, and the vehicle will be maintained according to the preset maintenance cycle.

[0059] In summary, the embodiments of this application can achieve the following beneficial effects: (1) By integrating multi-source data (sensors, maintenance records, user behavior, etc.) and utilizing deep learning models (such as LSTM, YOLOv5), the system can more accurately predict potential faults. For example, the advance time of fault warning can reach 5-7 days. Compared with the traditional simple alarm method based on threshold, the accuracy is improved by 300%, thereby improving the accuracy and foresight of fault prediction. (2) The embodiments of this application realize intelligent collaboration between vehicle health status, user needs and maintenance resources (spare parts, outlets, work stations). For example, the spare parts inventory turnover rate of maintenance outlets is expected to increase by 40%, reducing inventory backlog and capital occupation. The average maintenance waiting time for users is expected to be shortened by 50%. Since maintenance outlets can prepare spare parts and work stations in advance according to the maintenance plan, service efficiency and resource allocation can be greatly optimized. (3) The embodiments of this application can generate an “on-demand maintenance” plan based on the actual health condition of the vehicle, thereby avoiding unnecessary over-maintenance and is expected to save users 20%-30% of maintenance costs.

[0060] The vehicle maintenance method according to embodiments of this application, based on a preset fault prediction model, predicts vehicle faults based on preliminary anomaly detection results, identifies vehicle fault maintenance priorities based on the fault prediction results, adjusts the fault maintenance priorities using a reinforcement learning algorithm based on vehicle driving conditions, user driving data, and vehicle parts inventory status, generates a dynamic maintenance strategy for the vehicle based on the adjusted fault maintenance priorities, obtains recommended maintenance outlets for the vehicle based on the dynamic maintenance strategy, and performs maintenance on the vehicle according to the recommended maintenance outlets. This solves the problems of limitations in periodic maintenance, delayed fault diagnosis, and unreasonable resource allocation in related technologies. Through multi-source data fusion, deep learning models, and dynamic maintenance strategies, it improves maintenance efficiency and fault prediction accuracy, and optimizes resource allocation.

[0061] Next, the vehicle maintenance device proposed according to the embodiments of this application is described with reference to the accompanying drawings.

[0062] Figure 6This is a block diagram of a vehicle maintenance device according to an embodiment of this application.

[0063] like Figure 6 As shown, the vehicle maintenance device 10 includes: an acquisition module 100, a fault prediction module 200, an adjustment module 300, and a maintenance module 400.

[0064] The acquisition module 100 is used to acquire the preliminary anomaly detection results of the vehicle. The fault prediction module 200 is used to predict vehicle faults based on a preset fault prediction model and preliminary anomaly detection results, and obtain the vehicle fault prediction results. The adjustment module 300 is used to identify the vehicle's fault maintenance priority based on the fault prediction results, and adjust the fault maintenance priority using reinforcement learning algorithms according to the vehicle's driving conditions, user driving data, and vehicle parts inventory status. The maintenance module 400 is used to generate a dynamic maintenance strategy for the vehicle based on the adjusted fault maintenance priority, obtain recommended maintenance outlets for the vehicle based on the dynamic maintenance strategy, and perform maintenance on the vehicle according to the recommended maintenance outlets.

[0065] According to one embodiment of this application, the maintenance module 400 includes: The first generation unit is used to obtain the real-time running location of the vehicle when the adjusted fault maintenance priority is high priority, and generate a first maintenance strategy corresponding to the high priority based on the real-time running location of the vehicle and the faulty parts of the vehicle, so as to obtain the recommended maintenance outlets for the vehicle based on the first maintenance strategy. The second generation unit is used to generate a second maintenance strategy corresponding to the medium priority based on the faulty parts of the vehicle when the adjusted fault maintenance priority is medium priority, so as to obtain the recommended maintenance outlets for the vehicle according to the second maintenance strategy. The third generation unit is used to generate a third maintenance strategy corresponding to the low priority when the adjusted fault maintenance priority is low priority, based on the vehicle faulty parts and the preset maintenance cycle of the vehicle faulty parts, so as to obtain the recommended maintenance outlets for the vehicle according to the third maintenance strategy.

[0066] According to one embodiment of this application, the fault prediction module 200 includes: The first acquisition unit is used to acquire the vehicle's historical maintenance results; The data fusion unit is used to fuse multi-source data based on historical maintenance results and preliminary anomaly detection results to construct a vehicle health knowledge graph. The fault prediction unit is used to predict vehicle faults based on a preset fault prediction model and a health knowledge graph, and obtain the vehicle fault prediction results.

[0067] According to one embodiment of this application, the acquisition module 100 includes: The second acquisition unit is used to acquire vehicle operation data, vehicle driving environment data and user behavior data, wherein the vehicle operation data, vehicle driving environment data and user behavior data are collected by multiple vehicle-mounted sensors; The feature extraction unit is used to preprocess vehicle operation data, vehicle driving environment data and user behavior data, and to extract key features from the preprocessed vehicle operation data, vehicle driving environment data and user behavior data to obtain target data features. The anomaly detection unit is used to perform preliminary anomaly detection on vehicles based on a preset rule engine and target data features, and obtain preliminary anomaly detection results.

[0068] According to one embodiment of this application, the feature extraction unit includes: The normalization processing subunit is used to remove abnormal data from vehicle operation data, vehicle driving environment data and user behavior data to obtain the target data of the vehicle, and to perform normalization processing on the target data. The feature extraction subunit is used to extract the temporal features of vehicle operation data, the image features of the vehicle driving environment, and the user behavior features of user behavior data from the normalized target data, so as to obtain the target data features based on the temporal features, image features, and user behavior features.

[0069] According to one embodiment of this application, the adjustment module 300 includes: The analysis unit is used to analyze the fault prediction results based on a preset deep learning model to obtain the fault level, service life, maintenance cost and user needs of the vehicle faulty parts. The third acquisition unit is used to obtain the vehicle's fault maintenance priority based on the fault level, service life, maintenance cost, and user needs using a comprehensive scoring strategy.

[0070] The vehicle maintenance device according to embodiments of this application, based on a preset fault prediction model, predicts vehicle faults based on preliminary anomaly detection results, identifies the vehicle's fault maintenance priority based on the fault prediction results, adjusts the fault maintenance priority using a reinforcement learning algorithm based on vehicle driving conditions, user driving data, and vehicle parts inventory status, generates a dynamic maintenance strategy for the vehicle based on the adjusted fault maintenance priority, obtains recommended maintenance outlets for the vehicle based on the dynamic maintenance strategy, and performs maintenance on the vehicle according to the recommended maintenance outlets. This solves the problems of limited periodic maintenance, delayed fault diagnosis, and unreasonable resource allocation in related technologies. Through multi-source data fusion, deep learning models, and dynamic maintenance strategies, it improves maintenance efficiency and fault prediction accuracy, and optimizes resource allocation.

[0071] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 701, the processor 702, and the computer program stored on the memory 701 and executable on the processor 702.

[0072] When the processor 702 executes the program, it implements the vehicle maintenance method provided in the above embodiments.

[0073] Furthermore, electronic devices also include: Communication interface 703 is used for communication between memory 701 and processor 702.

[0074] The memory 701 is used to store computer programs that can run on the processor 702.

[0075] The memory 701 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0076] If the memory 701, processor 702, and communication interface 703 are implemented independently, then the communication interface 703, memory 701, and processor 702 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized into address buses, data buses, control buses, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0077] Optionally, in a specific implementation, if the memory 701, processor 702, and communication interface 703 are integrated on a single chip, then the memory 701, processor 702, and communication interface 703 can communicate with each other through an internal interface.

[0078] The processor 702 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0079] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the vehicle maintenance method described above.

[0080] This embodiment also provides a computer program product, including a computer program that is executed to implement the vehicle maintenance method of the above embodiment.

[0081] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0082] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0083] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0084] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0085] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0086] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium. When executed, the program includes one or a combination of the steps of the method embodiments.

[0087] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0088] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for maintaining a vehicle, characterized in that, Includes the following steps: Obtain preliminary anomaly detection results for the vehicle; Based on the preset fault prediction model, the vehicle is fault predicted according to the preliminary anomaly detection results, and the fault prediction result of the vehicle is obtained. Based on the fault prediction results, the fault maintenance priority of the vehicle is identified, and the fault maintenance priority is adjusted using a reinforcement learning algorithm according to the vehicle's driving conditions, user driving data, and vehicle parts inventory status. A dynamic maintenance strategy for the vehicle is generated based on the adjusted fault maintenance priority. Recommended maintenance outlets for the vehicle are obtained based on the dynamic maintenance strategy, and the vehicle is maintained according to the recommended maintenance outlets.

2. The method according to claim 1, characterized in that, The step of generating a dynamic maintenance strategy for the vehicle based on the adjusted fault maintenance priority, and obtaining recommended maintenance outlets for the vehicle based on the dynamic maintenance strategy, includes: When the adjusted fault maintenance priority is high priority, the real-time running location of the vehicle is obtained, and a first maintenance strategy corresponding to the high priority is generated based on the real-time running location of the vehicle and the faulty parts of the vehicle, so as to obtain the recommended maintenance outlets for the vehicle based on the first maintenance strategy. When the adjusted fault maintenance priority is medium priority, a second maintenance strategy corresponding to the medium priority is generated based on the faulty parts of the vehicle, so as to obtain the recommended maintenance outlets for the vehicle according to the second maintenance strategy; When the adjusted fault maintenance priority is low, a third maintenance strategy corresponding to the low priority is generated based on the vehicle faulty parts and the preset maintenance cycle of the vehicle faulty parts, so as to obtain the recommended maintenance outlets for the vehicle according to the third maintenance strategy.

3. The method according to claim 1, characterized in that, The fault prediction model based on a preset fault prediction model, which predicts faults in the vehicle based on the preliminary anomaly detection results, includes: Obtain the historical maintenance results of the vehicle; Based on the historical maintenance results and the preliminary anomaly detection results, multi-source data fusion is performed to construct the vehicle's health knowledge graph; Based on the preset fault prediction model, the vehicle's faults are predicted according to the health knowledge graph, and the fault prediction results of the vehicle are obtained.

4. The method according to claim 1, characterized in that, The acquisition of preliminary anomaly detection results for the vehicle includes: The system acquires vehicle operation data, vehicle driving environment data, and user behavior data, wherein the vehicle operation data, vehicle driving environment data, and user behavior data are collected by multiple on-board sensors. The vehicle operation data, the vehicle driving environment data, and the user behavior data are preprocessed, and key features are extracted from the preprocessed vehicle operation data, vehicle driving environment data, and user behavior data to obtain target data features. The vehicle is subjected to preliminary anomaly detection based on a preset rule engine and the target data features, and the preliminary anomaly detection results are obtained.

5. The method according to claim 4, characterized in that, The process involves preprocessing the vehicle operation data, vehicle driving environment data, and user behavior data, and then extracting key features from the preprocessed vehicle operation data, vehicle driving environment data, and user behavior data to obtain target data features, including: Abnormal data is removed from the vehicle operation data, the vehicle driving environment data, and the user behavior data to obtain the target data of the vehicle, and the target data is then normalized. The time-series features of the vehicle operation data, the image features of the vehicle driving environment, and the user behavior features of the user behavior data are extracted from the normalized target data to obtain the target data features based on the time-series features, the image features, and the user behavior features.

6. The method according to claim 1, characterized in that, The step of identifying the vehicle's fault maintenance priority based on the fault prediction results includes: The fault prediction results are analyzed based on a preset deep learning model to obtain the fault level, service life, maintenance cost and user needs of the vehicle faulty parts. Based on the fault level, service life, maintenance cost, and user needs, a comprehensive scoring strategy is used to determine the vehicle's fault maintenance priority.

7. A vehicle maintenance device, characterized in that, include: The acquisition module is used to acquire preliminary anomaly detection results for the vehicle. The fault prediction module is used to predict the fault of the vehicle based on the preliminary anomaly detection results according to the preset fault prediction model, and to obtain the fault prediction result of the vehicle. The adjustment module is used to identify the fault maintenance priority of the vehicle based on the fault prediction result, and adjust the fault maintenance priority according to the vehicle driving conditions, user driving data, and vehicle parts inventory status using a reinforcement learning algorithm. The maintenance module is used to generate a dynamic maintenance strategy for the vehicle based on the adjusted fault maintenance priority, obtain recommended maintenance outlets for the vehicle based on the dynamic maintenance strategy, and perform maintenance on the vehicle according to the recommended maintenance outlets.

8. The apparatus according to claim 7, characterized in that, The maintenance module includes: The first generation unit is used to obtain the real-time running location of the vehicle when the adjusted fault maintenance priority is high priority, and generate a first maintenance strategy corresponding to the high priority based on the real-time running location of the vehicle and the faulty parts of the vehicle, so as to obtain the recommended maintenance outlets for the vehicle based on the first maintenance strategy. The second generation unit is used to generate a second maintenance strategy corresponding to the medium priority based on the faulty parts of the vehicle when the adjusted fault maintenance priority is medium priority, so as to obtain the recommended maintenance outlets for the vehicle according to the second maintenance strategy. The third generation unit is used to generate a third maintenance strategy corresponding to the low priority based on the vehicle faulty parts and the preset maintenance cycle of the vehicle faulty parts when the adjusted fault maintenance priority is low priority, so as to obtain the recommended maintenance outlets for the vehicle according to the third maintenance strategy.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the vehicle maintenance method as described in any one of claims 1-6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the vehicle maintenance method as described in any one of claims 1-6.

Images