Intelligent troubleshooting reasoning system and method for vehicle faults

Abstract

The invention discloses an intelligent troubleshooting reasoning system and method for vehicle faults, and the system comprises a data preprocessing unit which converts a structured document and an unstructured document into structured data in a standard format; the diagnosis and repair knowledge graph unit is used for generating a diagnosis and repair knowledge graph from the structured data; the intelligent reasoning engine is used for providing recommendation operation according to the data of the vehicle terminal, the fault sensing information of the maintenance personnel and the diagnosis and repair knowledge graph of the diagnosis and repair knowledge graph unit, and sending the recommendation operation to the intelligent diagnosis center; and the intelligent diagnosis center is used for interaction between the user and the intelligent inference engine, provides storage, processing, transceiving of all data and services, provides a maintenance case management platform for the user, and processes and analyzes data and result data in the maintenance process. The invention relates to the technical field of vehicle detection and maintenance, and can solve the problems that in the prior art, manual vehicle fault troubleshooting and maintenance efficiency is low, and errors are prone to occurring.

Description

Technical Field

[0001] This invention relates to the field of vehicle inspection and maintenance technology, and in particular to an intelligent troubleshooting reasoning system and method for vehicle faults. Background Technology

[0002] As vehicles become increasingly electrified and intelligent, the added complexity associated with autonomous driving functions, electrification, and more software increases the difficulty of vehicle fault diagnosis. For repair personnel who rely on traditional experience, the difficulty of quickly and efficiently handling complex vehicle faults increases dramatically, often leading to misdiagnosis, prolonged repair time, or even incorrect replacement of parts, resulting in additional losses. It is estimated that misdiagnosis and improper repair cost automakers (OEMs) approximately $9 billion annually in improper warranty costs.

[0003] Currently, vehicle troubleshooting and repair in related technical fields mainly rely on the experience of repair personnel and relevant troubleshooting manuals, resulting in low repair efficiency and a high risk of errors. With the development of AI technology, advanced vehicle diagnostics based on data and AI technology will become a major industry trend for automakers (OEMs). Therefore, there is a need for an intelligent troubleshooting reasoning system and method for vehicle faults, capable of solving the problems of low efficiency and high error rate in existing manual vehicle troubleshooting and repair techniques. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent troubleshooting and reasoning system and method for vehicle faults, which can solve the problems of low efficiency and error-proneness in manual vehicle fault troubleshooting and repair in the prior art.

[0005] This invention is implemented as follows:

[0006] An intelligent troubleshooting reasoning system for vehicle malfunctions includes:

[0007] The data preprocessing unit receives structured and unstructured documents related to vehicle repair and troubleshooting, and the data preprocessing unit converts the structured and unstructured documents into structured data.

[0008] The diagnosis and repair knowledge graph unit is used to generate a diagnosis and repair knowledge graph from the structured data preprocessed by the data preprocessing unit.

[0009] The intelligent reasoning engine is used to provide recommended operations based on data from the vehicle terminal, fault perception information from maintenance personnel, and the diagnostic and repair knowledge graph of the diagnostic and repair knowledge graph unit, and sends the recommended operations to the intelligent diagnostic center.

[0010] The intelligent diagnostic hub is used for user interaction with the intelligent inference engine, providing storage, processing, and transmission of all data and services. It also provides users with a maintenance case management platform to process and analyze data and results during the maintenance process.

[0011] The structured and unstructured documents for vehicle repair and troubleshooting include: after-sales parts repair units, fault code lists and related information, repair and troubleshooting manuals, after-sales repair labor hours, after-sales parts prices, disassembly and assembly manuals, repair circuit diagrams, historical repair data, and system and component description documents; among which:

[0012] The after-sales parts repair unit record includes replaceable or repairable parts during the after-sales repair process;

[0013] The fault code list and related information include the reporting module, fault code name, fault code description, severity level, triggering conditions, whether the indicator light is on, ECU after-processing, and fault code repair conditions;

[0014] The troubleshooting manual includes non-DTC faults and DTC faults. Non-DTC faults include fault descriptions, fault test conditions and test details, test results and measures. DTC faults should at least include fault codes and their descriptions, possible causes of the faults and fault diagnosis test steps.

[0015] After-sales repair man-hours include man-hour information for monitoring and repair operations within the OEM after-sales claims system;

[0016] Aftermarket parts prices include component cost information from the OEM aftermarket claims system;

[0017] The disassembly and assembly manual includes detailed instructions on disassembling and installing the components;

[0018] The maintenance circuit diagram includes a diagram showing the connection relationships between the vehicle controller, sensors, actuators, and on-board electrical components;

[0019] Historical maintenance data includes the fault number, time of occurrence, vehicle model involved, title, fault code, failure mode, solution, and frequency of occurrence;

[0020] System and component documentation includes descriptions, composition, and layout diagrams of the system and its components.

[0021] The data preprocessing unit uses artificial intelligence technology, natural language processing technology, or large language models to process and extract content from structured and unstructured documents, extracts the entities required in the diagnosis and repair knowledge graph, as well as the relationship attributes between entities, and generates the diagnosis and repair knowledge graph through the diagnosis and repair knowledge graph unit.

[0022] The entities in the diagnostic and repair knowledge graph include: components, vehicle subsystems, fault codes, vehicle signals, signal processing results, fault symptoms, failure modes, testing methods, test results, repair solutions, typical cases, and the relationships between the above entities; wherein:

[0023] Components, the smallest unit used to describe vehicle repair;

[0024] Vehicle subsystem, used to describe the hierarchical relationships of the vehicle's systems;

[0025] Fault codes are used to describe fault codes triggered and reported by the vehicle controller;

[0026] Vehicle signals are used to describe the information flow transmitted by the vehicle controller to the outside world through the in-vehicle communication system;

[0027] Signal processing results are used to describe the local or global health characteristics of the vehicle;

[0028] Fault symptoms are used to describe the fault phenomena, including key characteristic information of the fault;

[0029] Failure mode is used to describe the root cause of a failure.

[0030] Testing methods are used to describe the testing methods and tools used to troubleshoot faults.

[0031] Test results are used to describe the results obtained after using test methods. The results include normal results and certain component failures or malfunctions.

[0032] A maintenance plan describes the measures to be taken for a faulty component, including repair or replacement.

[0033] Typical cases are used to describe the experience of successfully repairing similar faults in the past.

[0034] The intelligent troubleshooting reasoning system for vehicle malfunctions also includes a machine learning module, which is connected to the intelligent diagnostic center and used to perform closed-loop self-learning on the diagnostic and repair knowledge graph units.

[0035] The aforementioned maintenance case management platform includes case viewing, keyword search, and report generation; wherein:

[0036] Case viewing allows users to access the vehicle fault page by double-clicking the fault case information generated from the report.

[0037] Keyword search allows users to accurately, fuzzily, or in combination search for repair case information based on VIN code, vehicle model, and update time.

[0038] Report generation allows users to view and print fault reports on resolved cases.

[0039] The aforementioned perceived fault information refers to non-DTC faults in the maintenance and troubleshooting manual, which are used to report fault codes that cannot be obtained from the vehicle controller to the intelligent troubleshooting reasoning system in a standard format.

[0040] The user interaction with the intelligent inference engine includes the intelligent inference engine guiding the user to quickly troubleshoot and locate vehicle faults in an interactive manner. Specifically, this includes reporting perceived symptoms, reporting non-perceived symptoms, grouping and sorting fault codes, checking troubleshooting progress, inference calculations, viewing operation plans, and exporting diagnostic cases.

[0041] Symptom reporting is used by maintenance personnel to manually upload non-DTC faults to the intelligent troubleshooting and reasoning system.

[0042] Non-perceptible symptom reporting is used by the vehicle to automatically upload DTC faults to the intelligent troubleshooting and reasoning system.

[0043] Fault code grouping is used in scenarios where multiple fault codes occur. The system can identify the inherent correlation between fault codes and realize intelligent grouping and merging of fault codes for troubleshooting.

[0044] Sorting is used to arrange the recommended operations of the guided troubleshooting solution in order of priority;

[0045] Troubleshooting progress is displayed to maintenance personnel in the form of a progress bar.

[0046] Reasoning calculation refers to the use of artificial intelligence algorithms, based on diagnostic and repair knowledge graphs, to effectively isolate weakly related or irrelevant failure causes, pinpoint the root cause of the fault, and provide repair solutions based on fault symptom information. Reasoning calculation includes automatic troubleshooting and manual troubleshooting. Automatic troubleshooting refers to the intelligent troubleshooting reasoning system automatically processing vehicle signals and automatically triggering the intelligent troubleshooting reasoning system to perform reasoning calculations on the fault to achieve fault isolation and fault convergence. Manual troubleshooting is used when maintenance personnel submit operation plans during the troubleshooting process, triggering the intelligent troubleshooting reasoning system to perform reasoning calculations on the fault until the fault is repaired and resolved.

[0047] The operation plan is used by maintenance personnel to view the operation details and related diagnostic and repair data of the system's recommended operations during troubleshooting, guiding them to carry out inspections and repairs.

[0048] The diagnostic and repair case export function allows users to export historical repair case data in batches.

[0049] An intelligent troubleshooting reasoning method using a vehicle fault intelligent troubleshooting reasoning system includes the following steps:

[0050] S1: The DTC fault of the vehicle controller is sent to the intelligent troubleshooting and reasoning system through the near-field diagnostic instrument or remote diagnostic system. At the same time, the maintenance personnel manually input the perceived symptoms into the intelligent troubleshooting and reasoning system through the terminal equipment. The above fault codes and perceived symptoms are reported to the intelligent diagnostic center of the intelligent troubleshooting and reasoning system in the cloud server in a unified form.

[0051] S2: The intelligent diagnostic center sends the fault data to the intelligent inference engine. The intelligent inference engine retrieves the diagnostic and repair knowledge graph of the diagnostic and repair knowledge graph unit, creates fault cases for the faults and groups the cases, performs inference calculations for each group of fault cases, and sends the case creation status and inference calculation results to the terminal through the intelligent diagnostic center.

[0052] S3: Maintenance personnel can click on any case to enter the case details page and view the case details;

[0053] S4: The maintenance personnel view the corresponding recommended operations and their operation plans in the case details, inspect the vehicle according to the recommended operations, and submit the inspection results to the intelligent diagnostic center;

[0054] S5: The intelligent diagnostic center sends the examination results to the intelligent inference engine. The intelligent inference engine performs inference calculations based on the examination results and displays recommended operations in the recommended order.

[0055] S6: Repeat S4 and S5 until the fault is repaired and resolved during vehicle inspection, then close the fault case.

[0056] In step 1, the method for maintenance personnel to input perceived symptoms is as follows: maintenance personnel request a list of perceived symptoms existing in the system from the intelligent troubleshooting reasoning system through a terminal device, select possible perceived symptoms based on the maintenance personnel's judgment, and report them to the intelligent troubleshooting reasoning system.

[0057] Compared with the prior art, the present invention has the following advantages:

[0058] 1. The intelligent troubleshooting reasoning system of the present invention can not only reason and troubleshoot based on module fault codes, but also perform calculations based on real-time vehicle data streams. At the same time, it can further reason and calculate based on perceived fault information and inspection results, create relevant case groups, and provide corresponding recommended operations for maintenance personnel. This facilitates quick and accurate identification of fault causes, improves vehicle maintenance and troubleshooting efficiency and accuracy, reduces maintenance costs, and increases user satisfaction.

[0059] 2. This invention can support the acquisition of vehicle fault information and inject perceived fault information into the system. The system automatically matches the diagnostic knowledge graph corresponding to the vehicle model, and quickly analyzes the vehicle fault based on the system's own reasoning mechanism. It also provides guided troubleshooting guidance to help maintenance personnel quickly and accurately pinpoint the cause of the fault. This reduces the reliance on the professional knowledge of maintenance personnel, improves the accuracy and efficiency of fault troubleshooting, thereby reducing maintenance costs and increasing user satisfaction. Attached Figure Description

[0060] Figure 1 This is a structural block diagram of the intelligent troubleshooting and reasoning system for vehicle malfunctions of the present invention;

[0061] Figure 2 This is a flowchart of the intelligent troubleshooting reasoning method for vehicle faults according to the present invention.

[0062] In the diagram, 1 is the data preprocessing unit, 2 is the diagnostic and repair knowledge graph unit, 3 is the machine learning module, 4 is the intelligent diagnostic center, 5 is the intelligent inference engine, and 6 is the vehicle terminal. Detailed Implementation

[0063] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0064] Please see the appendix Figure 1 An intelligent troubleshooting reasoning system for vehicle malfunctions, comprising:

[0065] In the data preprocessing unit 1, structured and unstructured documents related to vehicle repair and troubleshooting are input into the data preprocessing unit 1, and the data preprocessing unit 1 converts the structured and unstructured documents into structured data.

[0066] Diagnosis and Repair Knowledge Graph Unit 2 is used to generate a diagnosis and repair knowledge graph from the structured data preprocessed by Data Preprocessing Unit 1.

[0067] The intelligent reasoning engine 5 is used to provide recommended operations based on the data from the vehicle terminal 6 and the diagnostic and repair knowledge graph of the diagnostic and repair knowledge graph unit 2, and to send the recommended operations to the intelligent diagnostic center 4.

[0068] The intelligent diagnostic hub 4 is used for interaction between users such as vehicle terminals 6 and maintenance personnel and the intelligent inference engine 5. It provides storage, processing, and transmission of all data and services, and provides users with a maintenance case management platform to process and analyze data and results during the maintenance process.

[0069] The data preprocessing unit 1, the diagnostic knowledge graph unit 2, the intelligent diagnostic center 4, and the intelligent reasoning engine 5 can be configured based on computer technology and AI technology. By replacing manual inspection and troubleshooting through data calculation and processing, they can improve the accuracy and efficiency of vehicle maintenance and troubleshooting, avoid unnecessary maintenance costs, and thus improve user satisfaction.

[0070] The documents in data preprocessing unit 1 include, but are not limited to: after-sales parts repair unit, fault code list and related information, repair and troubleshooting manual, after-sales repair hours, after-sales parts prices, disassembly and assembly manual, repair circuit diagram, historical repair data and system and component description documents.

[0071] Wherein: the after-sales parts repair unit records the parts that can be replaced or repaired during the after-sales repair process;

[0072] The fault code list and related information include the reporting module, fault code name, fault code description, severity level, triggering conditions, whether the indicator light is on, ECU after-processing, and fault code repair conditions.

[0073] The maintenance and troubleshooting manual includes non-DTC faults and DTC faults. The non-DTC faults include fault descriptions, fault test conditions and test details, test results and measures. The DTC faults should at least include fault codes and their fault descriptions, possible causes of the faults and fault diagnosis test steps.

[0074] The after-sales repair time record is the time information for monitoring and repair operations in the OEM after-sales claim system;

[0075] The after-sales component price record contains component cost information in the OEM after-sales claim system;

[0076] The disassembly and assembly manual records detailed instructions for disassembling and installing the components;

[0077] The maintenance circuit diagram records the connection relationships between the vehicle controller, sensors, actuators, and on-board electrical components.

[0078] The historical maintenance data includes the fault number, occurrence time, vehicle model involved, title, fault code, failure mode, solution, and frequency of occurrence.

[0079] The system and component documentation includes descriptions, composition, and layout diagrams of the system and its components.

[0080] Data preprocessing unit 1 processes and extracts content from structured and unstructured documents. It uses artificial intelligence technology, natural language processing (NLP) technology, or large language model (LLM) to preprocess the original document content and extract the entities required in the diagnostic and repair knowledge graph, as well as the relationship attributes between entities.

[0081] The diagnostic and repair knowledge graph unit 2 is used to convert unified structured data into corresponding diagnostic and repair knowledge graphs, which facilitates the retrieval of intelligent reasoning engine 5 and the investigation and diagnosis of faults. The entities in the diagnostic and repair knowledge graph include, but are not limited to, parts, vehicle subsystems, fault codes, vehicle signals, signal processing results, fault symptoms, failure modes, testing methods, test results, repair plans, and typical cases, as well as the relationships between the above entities.

[0082] Wherein: the component is used to describe the smallest component unit for vehicle diagnosis and repair;

[0083] The vehicle subsystem is used to describe the hierarchical relationship of the vehicle system;

[0084] The fault code is used to describe the fault code triggered and reported by the vehicle controller;

[0085] The vehicle signal is used to describe the information flow transmitted by the vehicle controller to the outside through the in-vehicle communication system;

[0086] The signal processing results are used to describe the local or global health characteristics of the vehicle;

[0087] The fault symptoms are used to describe the fault phenomena, including key characteristic information of the fault;

[0088] The failure modes are used to describe the root causes of the failures.

[0089] The aforementioned testing methods are used to describe the testing methods and tools used for troubleshooting.

[0090] The test results are used to describe the results obtained after using the test methods, including normal results and certain component failures or malfunctions.

[0091] The repair plan describes the measures to be taken for the faulty component, including repair or replacement;

[0092] The typical cases described are used to illustrate the experience of successfully repairing similar faults in the past.

[0093] The intelligent reasoning engine 5 uses data and AI technology, combined with data and diagnostic knowledge graphs, to troubleshoot problems. After identifying fault cases, it outputs relevant recommended actions as a reference for maintenance personnel to troubleshoot vehicles.

[0094] The intelligent diagnostic center 4 sends fault data to the intelligent inference engine 5, and sends the inference results from the intelligent inference engine 5 to the vehicle terminal 6. The intelligent diagnostic center 4 and the vehicle terminal 6 interact via wireless or wired network. Through interactive guidance, maintenance personnel can quickly troubleshoot and locate faults, thus realizing intelligent troubleshooting function.

[0095] The maintenance case management platform includes case viewing, keyword search, and report generation, among which:

[0096] The case viewing feature allows users to access the vehicle fault page by double-clicking the reported fault case information;

[0097] The keyword search is used by users to accurately / fuzzily / combinedly query repair case information based on VIN code, vehicle model, and update time.

[0098] The report generation is used by users to view and print fault reports of resolved cases.

[0099] Please see the appendix Figure 1 The intelligent troubleshooting reasoning system for vehicle faults also includes a machine learning module 3, which is connected to the intelligent diagnostic center 4 and is used to perform closed-loop self-learning on the diagnostic knowledge graph unit 2 to improve the troubleshooting accuracy of the intelligent troubleshooting reasoning system.

[0100] Machine learning module 3 can use neural network-based machine learning methods to create a training set from historical maintenance data and integrate the information from the historical maintenance data into the diagnostic knowledge graph. The diagnostic knowledge graph is regularly upgraded to keep it up-to-date, thereby improving the accuracy and efficiency of the intelligent inference engine 5 in troubleshooting and locating faults.

[0101] The intelligent troubleshooting and reasoning system for vehicle faults also includes a display screen (not shown in the figure) for displaying data. The data displayed on the display screen includes fault code analysis, repair rate, regional repair rate, trend changes in diagnostic behavior, trend changes in vehicle model fault rate, and percentage of fault phenomena.

[0102] Preferably, the display screen can be deployed in the vehicle manufacturing, management and sales sites to facilitate better management of vehicle design, manufacturing and sales, and reduce the frequency and recurrence of vehicle malfunctions.

[0103] The fault data includes fault codes sent by the vehicle terminal 6, manually entered perceived fault information, and the inspection results of the vehicle by maintenance personnel.

[0104] Fault codes are automatically generated by vehicle controllers, T-boxes, and other devices based on vehicle fault conditions. This facilitates troubleshooting based on fault codes, quickly pinpointing the cause of the fault, and can also be calculated based on the vehicle's real-time data stream for pre-testing and maintenance.

[0105] The perceived fault information can be selectively input by maintenance personnel based on the fault situation. The perceived fault information can be non-DTC faults in the maintenance and troubleshooting manual, and is used to report fault codes that cannot be obtained from the vehicle controller to the intelligent troubleshooting reasoning system in a standard format.

[0106] The inspection results can be entered by the maintenance personnel after inspecting the vehicle according to the recommended results. The inspection results include vehicle fault repair and resolution, vehicle faults that have not been repaired and resolved, and their corresponding fault codes and perceived fault information.

[0107] The intelligent troubleshooting and reasoning system for vehicle malfunctions is deployed on a cloud server and connected to terminal devices. Repair personnel manually input perceived fault information into the system via the terminal devices, which simultaneously provide a confirmation page for troubleshooting (including repair conditions for fault codes), ensuring that repair personnel correctly execute verification operations and that fault elimination is effective. Repair personnel can report problems encountered during the repair process to the intelligent troubleshooting and reasoning system. The system records these reported problems and displays them in a standardized list format on the terminal devices, including problem descriptions, submitter information, etc., and supports adding, deleting, modifying, and querying problem records. OEMs (equipment manufacturers) can process reported problems during the repair process via the terminal devices, display them in a list, and perform addition, deletion, modification, and querying of problem records. The intelligent troubleshooting and reasoning system can generate and export work order reports according to customer-defined format requirements, and can also export all retrieved historical case data. This data contains complete troubleshooting information and can be used for data analysis.

[0108] Preferably, the terminal device may include mobile devices such as after-sales diagnostic instruments, tablet computers, and mobile robots with user interfaces. The mobile robot with a user interface allows maintenance personnel to upload faults requiring troubleshooting via its interface. Since the intelligent troubleshooting reasoning system is deployed on a cloud server, the terminal device needs to connect via a mobile network or Wi-Fi to access the system.

[0109] The intelligent troubleshooting reasoning system can manage the fault cases and recommended actions generated by the intelligent reasoning engine 5. It can display all cases created by all 4S stores nationwide in a list, including case topic, submitter, publication status, submission status, etc., which facilitates searching from different dimensions. It also makes it easier for repair personnel to perform the recommended actions of the cases according to different statuses and scenario requirements, and facilitates the review, editing, rejection, viewing details, and viewing operation logs of repair cases.

[0110] Please see the appendix Figure 1 and attached Figure 2 A method for intelligent troubleshooting of vehicle malfunctions includes the following steps:

[0111] S1: The DTC fault of the vehicle controller is sent to the intelligent troubleshooting and reasoning system through the near-field diagnostic instrument or the remote diagnostic and repair system. At the same time, the maintenance personnel can manually input the perceived symptoms into the intelligent troubleshooting and reasoning system through the terminal. The above fault codes and perceived symptoms are reported to the intelligent diagnostic center 4 of the intelligent troubleshooting and reasoning system in the cloud server in a unified form.

[0112] Fault information can be obtained by maintenance personnel through a terminal device requesting a list of perceived symptoms stored in the intelligent troubleshooting reasoning system. The maintenance personnel can then select possible perceived symptoms based on their own judgment and report them to the intelligent troubleshooting reasoning system.

[0113] S2: The intelligent diagnostic center 4 sends the fault data to the intelligent inference engine 5. The intelligent inference engine 5 retrieves the diagnostic and repair knowledge graph of the diagnostic and repair knowledge graph unit 2, creates fault cases for the faults and groups the cases, performs inference calculations for each group of fault cases, and sends the case creation status and inference calculation results to the vehicle terminal 6, the terminal equipment of maintenance personnel and other terminals through the intelligent diagnostic center 4.

[0114] The intelligent inference engine 5 intelligently groups fault codes based on data and AI technology, and performs inference calculations on each group of fault cases in conjunction with the diagnostic and repair knowledge graph. This is used to assist in vehicle fault analysis, and through cases, it helps repair personnel quickly and accurately pinpoint the cause of the fault, reducing reliance on the professional knowledge of repair personnel, and improving the problems of low efficiency and error-proneness of manual repair by repair personnel.

[0115] S3: Maintenance personnel can click on any set of cases to enter the case details page and view the case details.

[0116] By creating case studies, we can provide guidance for repair personnel in troubleshooting vehicle malfunctions, which can help improve vehicle repair efficiency and reduce vehicle repair costs.

[0117] S4: The maintenance personnel view the corresponding recommended operations and their operation plans in the case details, inspect the vehicle according to the recommended operations, and submit the inspection results to the intelligent diagnostic center 4.

[0118] The inspection results include troubleshooting information for vehicle malfunctions and feedback from maintenance personnel. This facilitates the recording of reported issues and provides a standardized list display on the terminal device, including issue descriptions and submitter information. It can also be used to add, delete, modify, and query issue records. Furthermore, it allows OEMs (equipment manufacturers) to process issues reported during the maintenance process, display them in a list, and add, delete, modify, and query issue records via the terminal device. After maintenance is completed, a work order report can be generated and exported according to customer-defined format requirements. It can also export all historical case data containing complete troubleshooting information that has been retrieved.

[0119] S5: The intelligent diagnostic center 4 sends the examination results to the intelligent inference engine 5. The intelligent inference engine 5 performs inference calculations again based on the examination results and displays recommended operations in the recommended order.

[0120] By utilizing the feedback from maintenance personnel on vehicle inspection results and the further reasoning and calculation of the intelligent reasoning engine 5, guided troubleshooting guidance is provided to maintenance personnel through repeated operations. This helps maintenance personnel quickly and accurately pinpoint the cause of the fault, reducing reliance on the professional knowledge of maintenance personnel and improving the accuracy and efficiency of troubleshooting.

[0121] Interactive methods guide users to quickly troubleshoot and locate faults, including reporting perceived symptoms, reporting non-perceived symptoms, grouping and sorting fault codes, troubleshooting progress, reasoning calculations (including automatic and manual troubleshooting), viewing operation plans, and exporting diagnostic and repair cases.

[0122] Wherein: the perceived symptom reporting is used by maintenance personnel to manually upload non-DTC faults to the intelligent troubleshooting reasoning system;

[0123] The non-perceptible symptom reporting is used by the vehicle to automatically upload DTC faults to the intelligent troubleshooting and reasoning system.

[0124] The fault code grouping is used in scenarios where multiple fault codes occur. The system can identify the inherent correlation between fault codes and realize intelligent grouping and merging of fault codes for troubleshooting.

[0125] The sorting is used to arrange the recommended operations of the guided troubleshooting scheme in order of priority;

[0126] The troubleshooting progress is used to display the troubleshooting progress to the maintenance personnel in the form of a progress bar.

[0127] The aforementioned reasoning calculation specifically refers to the application of artificial intelligence algorithms, based on a diagnostic knowledge graph, to effectively isolate weakly related or irrelevant failure causes based on fault symptom information, pinpoint the root cause of the fault, and provide a repair solution. The automatic troubleshooting refers to the system automatically processing vehicle signals and automatically triggering the system to perform reasoning calculations on the fault to achieve fault isolation and convergence; the manual troubleshooting refers to the process where maintenance personnel submit their operation plans during troubleshooting, triggering the system to perform reasoning calculations on the fault until the fault is repaired and resolved.

[0128] The viewing operation plan is used by maintenance personnel to view the operation details and relevant diagnostic and repair data of the system's recommended operations during the troubleshooting process, guiding maintenance personnel to carry out inspections and repairs.

[0129] The diagnostic and repair case export function is used by users to export historical repair case data in batches.

[0130] S6: Repeat S4 and S5 until the fault is repaired and resolved during vehicle inspection, then close the fault case.

[0131] The intelligent troubleshooting reasoning system and method for vehicle faults of the present invention provides a standard troubleshooting system development process and toolchain. It adopts an intelligent diagnostic hub deployed in the cloud to collect vehicle fault codes and real-time data uploaded by the vehicle. Based on the intelligent reasoning engine and diagnostic knowledge graph, it infers faults and provides guided troubleshooting support to maintenance personnel through the terminal. It has a simple and transparent operation method, which can improve the efficiency of vehicle fault troubleshooting and maintenance. There is no need to consult the maintenance guide provided by the manufacturer, reducing the reliance on professional personnel. At the same time, it can also improve diagnostic accuracy and reduce maintenance time and costs.

[0132] For automakers, this invention can effectively improve the efficiency of diagnostic and repair capabilities and reduce their costs, standardize after-sales maintenance practices, enhance product user experience, and empower the full lifecycle management of vehicles.

[0133] For repair service providers, this invention can effectively lower the barriers to repair, improve repair efficiency, and increase service revenue.

[0134] For car owners, this invention can effectively reduce the risks of car use and eliminate car-related anxiety.

[0135] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the invention. Therefore, any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An intelligent troubleshooting reasoning system for vehicle malfunctions, characterized by: include: The data preprocessing unit (1) inputs structured and unstructured documents for vehicle maintenance and troubleshooting into the data preprocessing unit (1), and the data preprocessing unit (1) converts the structured and unstructured documents into structured data. The diagnosis and repair knowledge graph unit (2) is used to generate a diagnosis and repair knowledge graph from the structured data preprocessed by the data preprocessing unit (1); The intelligent reasoning engine (5) is used to provide recommended operations based on the data of the vehicle terminal (6), the fault information perceived by the maintenance personnel and the diagnostic knowledge graph of the diagnostic knowledge graph unit (2), and send the recommended operations to the intelligent diagnostic center (4). The intelligent diagnostic hub (4) is used for users to interact with the intelligent inference engine (5), providing storage, processing, sending and receiving of all data and services, while providing users with a maintenance case management platform to process and analyze data and results during the maintenance process.

2. The intelligent troubleshooting and reasoning system for vehicle faults according to claim 1, characterized in that: The structured and unstructured documents for vehicle repair and troubleshooting include: after-sales parts repair units, fault code lists and related information, repair and troubleshooting manuals, after-sales repair labor hours, after-sales parts prices, disassembly and assembly manuals, repair circuit diagrams, historical repair data, and system and component description documents; among which: The after-sales parts repair unit record includes replaceable or repairable parts during the after-sales repair process; The fault code list and related information include the reporting module, fault code name, fault code description, severity level, triggering conditions, whether the indicator light is on, ECU after-processing, and fault code repair conditions; The troubleshooting manual includes non-DTC faults and DTC faults. Non-DTC faults include fault descriptions, fault test conditions and test details, test results and measures. DTC faults should at least include fault codes and their descriptions, possible causes of the faults and fault diagnosis test steps. After-sales repair man-hours include man-hour information for monitoring and repair operations within the OEM after-sales claims system; Aftermarket parts prices include component cost information from the OEM aftermarket claims system; The disassembly and assembly manual includes detailed instructions on disassembling and installing the components; The maintenance circuit diagram includes a diagram showing the connection relationships between the vehicle controller, sensors, actuators, and on-board electrical components; Historical maintenance data includes the fault number, time of occurrence, vehicle model involved, title, fault code, failure mode, solution, and frequency of occurrence; System and component documentation includes descriptions, composition, and layout diagrams of the system and its components.

3. The intelligent troubleshooting and reasoning system for vehicle faults according to claim 1, characterized in that: The data preprocessing unit (1) uses artificial intelligence technology, natural language processing technology or large language model to process and extract the content of structured and unstructured documents, extract the entities required in the diagnosis and repair knowledge graph, as well as the relationship attributes between entities, and generate the diagnosis and repair knowledge graph through the diagnosis and repair knowledge graph unit (2).

4. The intelligent troubleshooting and reasoning system for vehicle faults according to claim 1, characterized in that: The entities in the diagnostic and repair knowledge graph include: components, vehicle subsystems, fault codes, vehicle signals, signal processing results, fault symptoms, failure modes, testing methods, test results, repair solutions, typical cases, and the relationships between the above entities; wherein: Components, the smallest unit used to describe vehicle repair; Vehicle subsystem, used to describe the hierarchical relationships of the vehicle's systems; Fault codes are used to describe fault codes triggered and reported by the vehicle controller; Vehicle signals are used to describe the information flow transmitted by the vehicle controller to the outside world through the in-vehicle communication system; Signal processing results are used to describe the local or global health characteristics of the vehicle; Fault symptoms are used to describe the fault phenomena, including key characteristic information of the fault; Failure mode is used to describe the root cause of a failure. Testing methods are used to describe the testing methods and tools used to troubleshoot faults. Test results are used to describe the results obtained after using test methods. The results include normal results and certain component failures or malfunctions. A maintenance plan describes the measures to be taken for a faulty component, including repair or replacement. Typical cases are used to describe the experience of successfully repairing similar faults in the past.

5. The intelligent troubleshooting and reasoning system for vehicle faults according to claim 1, characterized in that: The intelligent troubleshooting reasoning system for vehicle faults also includes a machine learning module (3), which is connected to the intelligent diagnostic center (4) and is used to perform closed-loop self-learning on the diagnostic knowledge graph unit (2).

6. The intelligent troubleshooting and reasoning system for vehicle faults according to claim 1, characterized in that: The aforementioned maintenance case management platform includes case viewing, keyword search, and report generation; wherein: Case viewing allows users to access the vehicle fault page by double-clicking the fault case information generated from the report. Keyword search allows users to accurately, fuzzily, or in combination search for repair case information based on VIN code, vehicle model, and update time. Report generation allows users to view and print fault reports on resolved cases.

7. The intelligent troubleshooting and reasoning system for vehicle faults according to claim 1, characterized in that: The aforementioned perceived fault information refers to non-DTC faults in the maintenance and troubleshooting manual, which are used to report fault codes that cannot be obtained from the vehicle controller to the intelligent troubleshooting reasoning system in a standard format.

8. The intelligent troubleshooting and reasoning system for vehicle faults according to claim 1, characterized in that: The user interaction with the intelligent inference engine (5) includes the intelligent inference engine (5) guiding the user to quickly troubleshoot and locate vehicle faults in an interactive manner, specifically including reporting perceived symptoms, reporting non-perceived symptoms, grouping and sorting fault codes, troubleshooting progress, inference calculation, viewing operation plans, and exporting diagnostic cases; wherein: Symptom reporting is used by maintenance personnel to manually upload non-DTC faults to the intelligent troubleshooting and reasoning system. Non-perceptible symptom reporting is used by the vehicle to automatically upload DTC faults to the intelligent troubleshooting and reasoning system. Fault code grouping is used in scenarios where multiple fault codes occur. The system can identify the inherent correlation between fault codes and realize intelligent grouping and merging of fault codes for troubleshooting. Sorting is used to arrange the recommended operations of the guided troubleshooting solution in order of priority; Troubleshooting progress is displayed to maintenance personnel in the form of a progress bar. Reasoning calculation refers to the use of artificial intelligence algorithms, based on diagnostic and repair knowledge graphs, to effectively isolate weakly related or irrelevant failure causes, pinpoint the root cause of the fault, and provide repair solutions based on fault symptom information. Reasoning calculation includes automatic troubleshooting and manual troubleshooting. Automatic troubleshooting refers to the intelligent troubleshooting reasoning system automatically processing vehicle signals and automatically triggering the intelligent troubleshooting reasoning system to perform reasoning calculations on the fault to achieve fault isolation and fault convergence. Manual troubleshooting is used when maintenance personnel submit operation plans during the troubleshooting process, triggering the intelligent troubleshooting reasoning system to perform reasoning calculations on the fault until the fault is repaired and resolved. The operation plan is used by maintenance personnel to view the operation details and related diagnostic and repair data of the system's recommended operations during troubleshooting, guiding them to carry out inspections and repairs. The diagnostic and repair case export function allows users to export historical repair case data in batches.

9. An intelligent troubleshooting reasoning method using the intelligent troubleshooting reasoning system for vehicle faults as described in any one of claims 1-8, characterized in that: Includes the following steps: S1: The DTC fault of the vehicle controller is sent to the intelligent troubleshooting and reasoning system through the near-field diagnostic instrument or remote diagnostic system. At the same time, the maintenance personnel manually input the perceived symptoms into the intelligent troubleshooting and reasoning system through the terminal equipment. The above fault codes and perceived symptoms are reported to the intelligent diagnostic center of the intelligent troubleshooting and reasoning system in the cloud server in a unified form (4). S2: The intelligent diagnostic center (4) sends the fault data to the intelligent reasoning engine (5). The intelligent reasoning engine (5) retrieves the diagnostic knowledge graph of the diagnostic knowledge graph unit (2), creates fault cases for the fault and groups the cases, performs reasoning calculations for each group of fault cases, and sends the case creation status and reasoning calculation results to the terminal through the intelligent diagnostic center (4). S3: Maintenance personnel can click on any case to enter the case details page and view the case details; S4: The maintenance personnel will check the corresponding recommended operations and their operation plans in the case details, inspect the vehicle according to the recommended operations, and submit the inspection results to the intelligent diagnostic center (4); S5: The intelligent diagnostic center (4) sends the examination results to the intelligent reasoning engine (5). The intelligent reasoning engine (5) performs reasoning calculations again based on the examination results and displays the recommended operations in the recommended order. S6: Repeat S4 and S5 until the fault is repaired and resolved during vehicle inspection, then close the fault case.

10. The intelligent troubleshooting reasoning method according to claim 9, characterized in that: in In step 1, the method for maintenance personnel to input perceived symptoms is as follows: maintenance personnel request a list of perceived symptoms existing in the intelligent troubleshooting and reasoning system through a terminal device, select possible perceived symptoms based on the maintenance personnel's judgment, and report them to the intelligent troubleshooting and reasoning system.

Images