Intelligent fault diagnosis system and method

Abstract

The invention discloses an intelligent fault diagnosis system and method, and belongs to the field of vehicle fault diagnosis. The system comprises a big data preprocessing layer, a data analysis processing layer and a user application layer, and the big data preprocessing layer is used for obtaining vehicle data, preprocessing the vehicle data and then sending the vehicle data to the data analysis processing layer; the data analysis processing layer is used for performing fault prediction and early warning according to the input data of the big data preprocessing layer, performing diagnosis analysis on the found fault, generating a solution and outputting the solution to the user application layer; and the user application layer is used for receiving and analyzing the solution, performing task allocation and data forwarding according to a preset processing mechanism, and executing corresponding task operation. According to the invention, fault prediction and early warning are realized, a solution is intelligently generated and executed, and the diagnosis efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention belongs to the field of vehicle fault diagnosis, and in particular, relates to an intelligent fault diagnosis system and method. Background Art

[0002] New energy vehicles are experiencing a surge in development, with electrification, intelligence, and connectivity becoming inevitable trends in the industry. People are placing higher demands on environmentally friendly, intelligent, and convenient mobility solutions, hoping to enjoy more innovative and high-quality travel experiences. New energy vehicles are undoubtedly a key force in driving future mobility.

[0003] However, in the critical area of ​​new energy vehicle fault diagnosis, the traditional fault diagnosis methods currently used have numerous shortcomings. For one thing, traditional diagnostic methods primarily rely on vehicle testing using OBD diagnostic instruments and monitoring via manual remote platforms. This approach is not only slow to respond but also prone to misjudgments, resulting in low diagnostic efficiency and failing to meet users' needs for fast and accurate vehicle fault diagnosis. Furthermore, neither diagnostic instruments nor remote platforms offer fault prevention, intelligent early warning and prediction, or real-time intelligent analysis and effective solutions when faults occur. Furthermore, existing diagnostic service models are relatively limited, requiring users to drive their vehicles to a repair shop and use diagnostic instruments to determine the cause of the fault, which undoubtedly brings significant inconvenience.

[0004] To this end, the present invention proposes an intelligent fault diagnosis system and method. Summary of the Invention

[0005] The present invention aims to overcome the deficiencies of the prior art and proposes an intelligent fault diagnosis system and method to achieve the following objectives: to achieve fault prediction and early warning, to intelligently generate and execute solutions, and to improve diagnostic efficiency and accuracy.

[0006] In order to achieve the above objectives, the technical solution adopted by the present invention is: an intelligent fault diagnosis system, which includes a big data preprocessing layer, a data analysis and processing layer, and a user application layer, wherein:

[0007] The big data preprocessing layer is used to obtain vehicle data and preprocess it before sending it to the data analysis and processing layer;

[0008] The data analysis and processing layer is used to perform fault prediction and early warning based on the input data of the big data pre-processing layer, and to diagnose and analyze the found faults and generate solutions to output to the user application layer;

[0009] The user application layer is used to receive and analyze the solution, perform task allocation and data forwarding according to a preset processing mechanism, and execute corresponding task operations.

[0010] Preferably, data is transferred and stored between the big data pre-processing layer, the data analysis and processing layer, and the user application layer through a cloud server.

[0011] Preferably, the vehicle data acquired by the big data preprocessing layer includes:

[0012] Direct data, i.e., vehicle operating data, including vehicle T-box data, diagnostic data, and cloud data;

[0013] Indirect data, namely system operation support data, includes vehicle operation control logic strategy information, vehicle configuration and diagnostic configuration information, software information, maintenance manual information, design prevention stage failure mode information and test data, case information, technical requests and customer requests, etc.

[0014] Preferably, the preprocessing operations of the big data preprocessing layer include: removing noise, invalid values ​​and abnormal values ​​in the acquired vehicle data, and converting unstructured data into a structured data format.

[0015] Preferably, the data analysis and processing layer includes an early warning prediction and fault model, and a diagnosis and analysis model, wherein:

[0016] The early warning prediction and fault model is used to predict vehicle faults based on the input data of the big data preprocessing layer and generate corresponding early warning signals, and the predicted vehicle faults are sent to the diagnostic analysis model;

[0017] The diagnostic analysis model is used to perform attribution diagnosis based on the predicted vehicle fault, and generate a targeted solution and send it to the user application layer.

[0018] Preferably, the data analysis and processing layer also includes an AI deep learning model, which is used to obtain the historical working data of the early warning prediction and fault model, as well as the diagnostic analysis model, and input it into the deep learning algorithm for training optimization, and obtain the optimized model parameters to adjust the early warning prediction and fault model, as well as the diagnostic analysis model.

[0019] Preferably, the attribution diagnosis method of the diagnostic analysis model includes Bayesian network, decision tree, etc.

[0020] Preferably, the user application layer includes an intelligent integrated processing module and a functional service module, wherein:

[0021] The intelligent comprehensive processing module is used to receive and analyze the solution, and then forward the task and corresponding data to the corresponding functional business module according to a preset processing mechanism;

[0022] The functional business module is used to receive the task and corresponding data and execute the corresponding task operation according to its own business logic.

[0023] Preferably, the functional business modules include an early warning platform, an intelligent maintenance platform, a preventive tracking platform, a remote monitoring platform, a remote diagnosis platform, a remote maintenance scheduling platform, etc.

[0024] This application also proposes an intelligent fault diagnosis method, using the above-mentioned intelligent fault diagnosis system, the method comprising:

[0025] The data preprocessing layer collects vehicle data, preprocesses it, and then sends it to the data analysis and processing layer;

[0026] The data analysis and processing layer performs fault prediction and early warning based on the input data of the big data preprocessing layer, and generates solutions after diagnosing and analyzing the discovered faults and outputs them to the user application layer;

[0027] After receiving and analyzing the solution, the user application layer performs task allocation and data forwarding according to a preset processing mechanism and executes corresponding task operations.

[0028] The technical effects of the present invention are:

[0029] (1) The present invention performs real-time fault prediction and early warning for vehicles, intelligently generates solutions, and achieves "zero" perception of faults by users through rapid response and early repair.

[0030] (2) The present invention can monitor vehicle data in real time and predict possible failures, thereby reducing the impact of failures on users and improving the real-time performance of the system.

[0031] (3) The present invention reduces manual intervention and improves the efficiency and accuracy of fault handling; at the same time, based on efficient data processing, it improves diagnostic efficiency and timeliness, ensuring the operation of users' vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A schematic diagram of the structure of an intelligent fault diagnosis system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The following is a further detailed description of the specific embodiments of the present invention through the description of the invention with reference to the accompanying drawings, with the aim of helping those skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present invention and to facilitate its implementation. It should be noted that the terms "first", "second" and the like described in this application are only used to facilitate the description of the technical solution to distinguish components, and the corresponding component configurations may be the same or different, and are not intended to limit this application. In order to make the technical solution of the present invention clearer, the present invention is explained through the following invention.

[0034] The present invention provides an intelligent fault diagnosis system, such as Figure 1 As shown, the system includes a big data pre-processing layer, a data analysis and processing layer, and a user application layer, wherein:

[0035] The big data preprocessing layer is used to obtain vehicle data and preprocess it before sending it to the data analysis and processing layer;

[0036] The data analysis and processing layer is used to perform fault prediction and early warning based on the input data of the big data pre-processing layer, and to diagnose and analyze the found faults and generate solutions to output to the user application layer;

[0037] The user application layer is used to receive and analyze the solution, perform task allocation and data forwarding according to a preset processing mechanism, and execute corresponding task operations.

[0038] Specifically, the present invention utilizes cloud servers for data transmission and storage between the big data preprocessing layer, data analysis and processing layer, and user application layer. Cloud servers are typically deployed in high-speed network environments, enabling rapid data transmission and ensuring efficient operation of the present application's system. Furthermore, cloud servers possess vast storage resources that can meet the large amounts of data storage required by the present application's system during operation. Compared to physical storage devices, cloud servers are more convenient to operate, easier to expand, and less expensive.

[0039] The big data preprocessing layer communicates with the vehicle through the API interface to obtain vehicle data. The acquired vehicle data includes direct data and indirect data, including:

[0040] Direct data refers to the vehicle's operating data, including vehicle T-box data, diagnostic data, and cloud data. Specifically, it involves real-time vehicle data, DTC (Disturbance Code) and health report data (i.e., on-board diagnostic data), vehicle maintenance monitoring data, and the aggregate data defined by the CAN communication matrix.

[0041] Indirect data, or system operation support data, refers to various types of information that does not directly derive from the vehicle's operating status but supports vehicle function implementation, system analysis, and fault diagnosis. This includes comprehensive information such as vehicle operation control logic strategy information, vehicle configuration and diagnostic configuration information, software information, maintenance manual information (including diagnostic fault scripts and circuit diagrams), failure mode information and test data from the design prevention phase, case information (such as technical support and claims information), technical requests, and customer requests.

[0042] For the acquired vehicle data, the big data preprocessing layer performs corresponding preprocessing operations, including removing noise, invalid values ​​and abnormal values ​​in the acquired vehicle data, and clearly marking the data to facilitate data classification management. At the same time, there is a large amount of unstructured data in the acquired vehicle data. For this type of data, the machine cannot directly extract the logical relationship between the data, and the processing efficiency is low. Therefore, the present invention also converts the unstructured data into a structured data format stored in the form of tables and key-value pairs to ensure that high-quality and high-availability structured data input is provided for subsequent analysis and processing links, and the data is properly stored in the cloud server so that it can be called at any time. After converting the unstructured data into structured data, the machine can directly perform logical analysis and extraction, which improves work efficiency.

[0043] The data analysis and processing layer is the core of the intelligent fault diagnosis system. It receives structured data input from the big data preprocessing layer and uses a variety of pre-established algorithm models to process the data, improving the system's intelligence and reducing manual intervention. The data analysis and processing layer of the present invention includes early warning prediction and fault models, as well as diagnostic analysis models.

[0044] The early warning prediction and fault model is used to predict vehicle faults based on the input data from the big data preprocessing layer and generate corresponding early warning signals. The predicted vehicle faults are sent to the diagnostic analysis model to provide clear guidance for subsequent diagnostic analysis. During the vehicle design and manufacturing phase, various failure conditions are simulated, and the results are used to construct an early warning prediction and fault model. For example, for a failure condition involving unstable engine power output, characteristics such as the engine speed fluctuation range, torque changes, and sensor signal anomalies are used as model input parameters. Appropriate thresholds and judgment rules are set, and the resulting early warning prediction and fault model can be used to predict engine power system faults.

[0045] The diagnostic analysis model is used to perform attribution diagnosis based on the predicted vehicle fault, and generate targeted solutions and send them to the user application layer. Specifically, the diagnostic analysis model combines the predicted vehicle fault, traces the root cause of the fault through the attribution diagnosis method, and analyzes why the fault occurred. The attribution diagnosis method includes Bayesian networks, decision trees, etc., all of which use existing technologies, and this application does not elaborate on their working principles. For example, according to vehicle historical data, unstable engine output failure may be caused by sensor failure, temperature abnormality, etc. These data are processed through attribution diagnosis methods such as Bayesian networks and decision trees to model the relationship between the fault and various fault causes, and then analyze the root cause of the fault.

[0046] After determining the root cause of the fault, the diagnostic analysis model can intelligently generate targeted solutions. For example, the diagnostic analysis model can match corresponding solutions from a historical solution library based on the root cause of the fault.

[0047] In addition, the data analysis and processing layer of the present invention also includes an AI deep learning model, which is used to obtain the historical working data of the early warning prediction and fault model, as well as the diagnostic analysis model, and input it into the deep learning algorithm for training optimization, and obtain the optimized model parameters to adjust the early warning prediction and fault model, as well as the diagnostic analysis model. Exemplarily, the AI ​​deep learning model trains a neural network model (such as a convolutional neural network, etc.) based on the massive historical working data of the early warning prediction and fault model, as well as the diagnostic analysis model, so as to obtain the optimized model parameters and feed them back to the early warning prediction and fault model, as well as the diagnostic analysis model. For example, for the early warning prediction and fault model, a more accurate fault judgment threshold is provided to reduce misjudgment, and for the diagnostic analysis model, a more efficient and reliable attribution logic is provided to improve attribution accuracy.

[0048] The data analysis and processing layer of this invention intelligently identifies faults and generates corresponding solutions. Furthermore, the AI ​​deep learning model empowers the system with self-evolutionary capabilities. Through machine self-learning and deep learning mechanisms, the system continuously refines and optimizes the model, continuously improving the precision and accuracy of fault prediction analysis. This enables the intelligent fault diagnosis system to accurately make judgments and decisions in complex and ever-changing vehicle fault scenarios, achieving intelligent and standardized fault handling.

[0049] In the application, the user application layer is the key link for the close connection between the intelligent fault diagnosis system and the user's actual business scenarios, including the intelligent comprehensive processing module and the functional business module, among which:

[0050] The intelligent comprehensive processing module is used to receive and analyze the solution and then forward the task and corresponding data to the corresponding functional business module according to a preset processing mechanism. The functional business module is used to receive the task and corresponding data and execute the corresponding task operation according to its own business logic. Common functional business modules include early warning platforms, intelligent maintenance platforms, preventive tracking platforms, remote diagnosis platforms, remote monitoring platforms, and remote maintenance scheduling platforms. During specific implementation, other functional business modules can be introduced according to actual needs.

[0051] The pre-set processing mechanisms within the intelligent integrated processing module essentially map the solution's data features to the business modules. For example, when a fault requires real-time monitoring and data analysis to predict potential risks and trigger alerts in advance, the early warning platform is invoked. When a fault can be fixed with a software update, an OTA update is performed via the remote diagnosis platform. When a fault requires real-time collection and visualization of global operational status, the remote monitoring platform is invoked. When a fault requires on-site repair, the remote repair dispatch platform automatically queries DMS inventory information, promptly notifies regional branches to prepare spare parts and technical tools, and schedules appointments with the customer through the call center, flexibly arranging mobile or in-store service to fully meet the user's repair and maintenance needs. When faults require continuous tracking and safety risk prevention, the preventive tracking platform is invoked. When fault resolution requires parts maintenance, the intelligent maintenance platform is invoked. The user application layer transforms the efficiency advantages of the intelligent fault diagnosis system into a practical user experience.

[0052] The intelligent fault diagnosis system of the present invention also reserves multiple API interfaces for communicating with external systems, for example, synchronizing fault diagnosis results to a quality feedback system to continuously optimize vehicle performance.

[0053] At the same time, the present invention also proposes an intelligent fault diagnosis method, using the above-mentioned intelligent fault diagnosis system, the method comprising:

[0054] The data preprocessing layer collects vehicle data, preprocesses it, and then sends it to the data analysis and processing layer;

[0055] The data analysis and processing layer performs fault prediction and early warning based on the input data of the big data preprocessing layer, and generates solutions after diagnosing and analyzing the discovered faults and outputs them to the user application layer;

[0056] After receiving and analyzing the solution, the user application layer performs task allocation and data forwarding according to a preset processing mechanism and executes corresponding task operations.

[0057] An embodiment of the present invention is as follows:

[0058] During the operation of the vehicle, the real-time data of the vehicle is sent to the intelligent fault diagnosis system through the T-Box and the cloud. The big data pre-processing layer will collect the vehicle data for data pre-processing and transmit it to the data analysis and processing layer. The early warning prediction and fault model of the data analysis and processing layer will find potential health hazards in the vehicle's battery system through analysis and issue early warning signals in a timely manner. The diagnostic analysis model generates a detailed battery maintenance plan. After receiving the plan, the intelligent comprehensive processing system of the user application layer forwards it to the remote maintenance scheduling platform according to the preset processing mechanism (notifying the corresponding outlets to prepare spare parts, tools, and technology) and makes an appointment for the customer through the call center. After the appointment is confirmed, a reminder is sent through the customer APP. After the customer arrives at the store, the maintenance personnel quickly complete the battery maintenance work according to the plan, ensuring the normal operation of the vehicle, reducing the vehicle downtime caused by battery failure, and improving the vehicle's operating time and user experience.

[0059] In summary, this invention provides real-time vehicle fault prediction and early warning, intelligently generates solutions, and, through rapid response and proactive repair, ensures users have zero awareness of faults. This overcomes the limitations of traditional diagnostic methods and provides a more efficient, accurate, and convenient intelligent fault diagnosis solution.

[0060] The present invention has been described above with reference to the accompanying drawings. Obviously, the specific implementation of the present invention is not limited to the above-described method. Any non-substantial improvements made using the method concepts and technical solutions of the present invention, or any direct application of the above-described concepts and technical solutions to other situations without modification, fall within the scope of protection of the present invention.

Claims

1. An intelligent fault diagnosis system, characterized by: The system includes a big data pre-processing layer, a data analysis and processing layer, and a user application layer, wherein: The big data preprocessing layer is used to obtain vehicle data and preprocess it before sending it to the data analysis and processing layer; The data analysis and processing layer is used to perform fault prediction and early warning based on the input data of the big data pre-processing layer, and to diagnose and analyze the found faults and generate solutions to output to the user application layer; The user application layer is used to receive and analyze the solution, perform task allocation and data forwarding according to a preset processing mechanism, and execute corresponding task operations.

2. An intelligent fault diagnosis system according to claim 1, characterized in that: Data is transferred and stored between the big data pre-processing layer, the data analysis and processing layer, and the user application layer through the cloud server.

3. The intelligent fault diagnosis system according to claim 1, characterized in that: The vehicle data obtained by the big data preprocessing layer includes: Direct data, i.e., vehicle operating data, including vehicle T-box data, diagnostic data, and cloud data; Indirect data, namely system operation support data, includes vehicle operation control logic strategy information, vehicle configuration and diagnostic configuration information, software information, maintenance manual information, design prevention stage failure mode information and test data, case information, technical requests and customer requests.

4. An intelligent fault diagnosis system according to claim 1 or 3, characterized in that: The preprocessing operations of the big data preprocessing layer include: removing noise, invalid values ​​and abnormal values ​​in the acquired vehicle data, and converting unstructured data into a structured data format.

5. The intelligent fault diagnosis system according to claim 1, characterized in that: The data analysis and processing layer includes early warning prediction and fault models, and diagnostic analysis models, among which: The early warning prediction and fault model is used to predict vehicle faults based on the input data of the big data preprocessing layer and generate corresponding early warning signals, and the predicted vehicle faults are sent to the diagnostic analysis model; The diagnostic analysis model is used to perform attribution diagnosis based on the predicted vehicle fault, and generate a targeted solution and send it to the user application layer.

6. An intelligent fault diagnosis system according to claim 5, characterized in that: The data analysis and processing layer also includes an AI deep learning model, which is used to obtain the historical working data of the early warning prediction and fault model, as well as the diagnostic analysis model, and input it into the deep learning algorithm for training and optimization, and obtain optimized model parameters to adjust the early warning prediction and fault model, as well as the diagnostic analysis model.

7. The intelligent fault diagnosis system according to claim 5, characterized in that: The attribution diagnosis method of the diagnostic analysis model includes Bayesian network, decision tree, etc.

8. The intelligent fault diagnosis system according to claim 1, characterized in that: The user application layer includes an intelligent integrated processing module and a functional service module, wherein: The intelligent comprehensive processing module is used to receive and analyze the solution, and then forward the task and corresponding data to the corresponding functional business module according to a preset processing mechanism; The functional business module is used to receive the task and corresponding data and execute the corresponding task operation according to its own business logic.

9. The intelligent fault diagnosis system according to claim 8, characterized in that: The functional business modules include an early warning platform, an intelligent maintenance platform, a remote monitoring platform, a preventive tracking platform, a remote diagnosis platform, and a remote maintenance scheduling platform.

10. An intelligent fault diagnosis method, using an intelligent fault diagnosis system according to any one of claims 1 to 9, characterized in that: The method comprises: The data preprocessing layer collects vehicle data, preprocesses it, and then sends it to the data analysis and processing layer; The data analysis and processing layer performs fault prediction and early warning based on the input data of the big data preprocessing layer, and generates solutions after diagnosing and analyzing the discovered faults and outputs them to the user application layer; After receiving and analyzing the solution, the user application layer performs task allocation and data forwarding according to a preset processing mechanism and executes corresponding task operations.

Images