Rail transit vehicle fault closed-loop tracking management system

Abstract

The application discloses a rail transit vehicle fault closed-loop tracking management system, comprising a multi-dimensional fault diagnosis subsystem, an abnormality detection data knowledge base subsystem, a vehicle key component and associated component fault prediction subsystem, and an extreme working condition adaptive protection subsystem. The rail transit vehicle fault closed-loop tracking management system of the application aims at the problems of poor real-time performance, data island, unpredictable fault risk, weak unknown fault processing capacity, poor adaptability to special working conditions and the like of the existing rail transit vehicle fault detection system, constructs a closed-loop management method and system for guaranteeing that vehicle faults are timely and effectively processed, and through the system, multi-dimensional improvement is realized in the safety guarantee, operation and maintenance efficiency, service quality and management optimization of the rail transit maglev vehicle, closed-loop tracking from fault discovery to solution is realized, and the systematicness and efficiency of the process are ensured.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, and in particular, to a closed-loop tracking and management system for rail transit vehicle faults. Background Technology

[0002] Currently, during daily operations, rail transit vehicles require effective management and tracking of vehicle malfunctions from discovery to resolution. In response to the needs of rail transit vehicle operation safety assurance and the automation and intelligent development of manual maintenance operations, there is an urgent need for a closed-loop management method and system that can effectively ensure the monitoring, diagnosis, and maintenance of rail transit vehicle malfunctions.

[0003] The publication number is CN112650201A, and the title is "Vehicle Fault Diagnosis Method, Device, Vehicle and Storage Medium". This method determines key fault data in the original fault data based on fault indicators, sends the key fault data, and diagnoses the vehicle fault problem quickly and accurately through the key fault data. However, the method cannot iteratively update the original fault data and fault indicators, which raises questions about its reliability.

[0004] The patent application, CN112363488A, entitled "Vehicle Fault Handling System and Method," describes a system that uses a pre-set neural network model to predict faults and obtain prediction results. Based on the fault classification results, it identifies the corresponding fault handling module for potential faults and sends the corresponding fault handling instructions to the module for processing. If an unmanageable potential fault is detected, the system sends the fault handling instructions to a remote diagnostic unit for processing. However, it lacks a complete fault handling method and cannot guarantee timely fault detection and effective handling.

[0005] The system, with publication number CN105159283A and titled "Remote Fault Analysis and Feedback System for Urban Rail Transit Vehicles," collects vehicle data through onboard equipment, which is then received and analyzed by ground equipment. The diagnostic results are fed back to the vehicle's intelligent terminal in real time. However, it cannot predict potential fault risks in critical vehicle components and lacks relevant solutions or suggestions for resolving these faults, resulting in limited practicality.

[0006] The publication number is CN116466677A, and the title is "Vehicle Remote Fault Diagnosis System and Method, Storage Medium, and Computer Equipment." This method analyzes the target faults of each target vehicle in real time based on target data uploaded by the on-board terminal and provides prompts to the end user, enabling rapid location of target vehicle faults without on-site inspection by maintenance personnel. However, it does not form a systematic anomaly detection system from the fault data pre-stored on the data platform, which cannot guarantee the proper handling of faults.

[0007] In conclusion, it is essential to research a closed-loop management system for monitoring, diagnosing, and repairing faults in rail transit vehicles, which can enable comprehensive prediction and rapid closed-loop response to vehicle faults. Summary of the Invention

[0008] This application provides a closed-loop tracking and management system for rail transit vehicle faults to solve the technical problems of poor real-time performance, data silos, unpredictable fault risks, weak ability to handle unknown faults, and poor adaptability to special working conditions in existing technologies.

[0009] This application is achieved through the following solution:

[0010] A closed-loop tracking and management system for rail transit vehicle faults, characterized in that it includes:

[0011] A multi-dimensional fault diagnosis subsystem is used to introduce multi-physics field coupling simulation technology, construct a digital twin of the maglev vehicle, realize real-time mapping between physical entities and virtual models, and develop a fault diagnosis decision engine based on deep reinforcement learning. On the basis of the dynamic acquisition of existing real-time vehicle operation status data, trackside inspection image data, and maintenance inspection data, environmental perception data, cross-system operation data, and cross-model / cross-line data are added. A graph neural network (GNN) is used to construct a knowledge graph of "vehicle-environment-operation" association, quantify the impact weight of environmental factors (such as high temperature) on component life (such as cooling system), and improve the scenario adaptability of fault diagnosis.

[0012] The anomaly detection data knowledge base subsystem is used to effectively couple onboard monitoring data, trackside inspection data, maintenance business data, environmental perception data, cross-system operation data, and cross-model / cross-line data for data information in the operation and maintenance process of rail transit vehicles, and extract effective data to form an anomaly detection data knowledge base; at the same time, through the component health transfer learning framework, it realizes cross-model knowledge transfer, constructs a component full life cycle degradation trajectory map, introduces federated learning + self-supervised learning technology and uncertainty quantification plan, constructs cross-line and cross-model unknown fault identification, and provides prediction credibility assessment;

[0013] The vehicle key component and related component fault prediction subsystem is used to realize fault diagnosis and prediction of key components of rail transit vehicles by using intelligent diagnostic and big data analysis technology. At the same time, it introduces time sequence graph network (TGN) to construct component fault propagation network, predict secondary faults that may be caused by the main fault, generate combined maintenance solutions in advance, and provide scientific decision-making for vehicle maintenance.

[0014] The extreme working condition adaptive protection subsystem is used to build a three-dimensional mapping relationship library of "working condition-failure mode-solution" through a distributed anomaly detection framework based on federated learning, so as to ensure the real-time response of the adaptive protection system under special working conditions.

[0015] Furthermore, the multi-dimensional fault diagnosis subsystem is specifically used for:

[0016] A high-fidelity digital twin model of a maglev vehicle is established, integrating a multiphysics simulation engine; real-time vehicle operation status data, trackside inspection image data, maintenance and inspection data, environmental perception data, cross-system operation data, and cross-model / cross-line data are synchronized to the digital twin model in real time to generate virtual sensor data; when the deviation between actual sensor data and virtual sensor data exceeds a threshold, deep diagnosis is triggered, and the diagnosis results are fed back to the anomaly detection data knowledge base to optimize the parameters of the digital twin model;

[0017] When the maglev vehicle is running, it transmits real-time data to the vehicle on-route status monitoring module. The vehicle on-route status monitoring module views abnormal real-time data in the digital twin model and issues an alarm, while also providing a matching solution. When the vehicle enters the depot, it transmits real-time detection data to the vehicle trackside status detection module. The detection module views abnormal detection data in the digital twin model and issues an alarm, while also matching a corresponding solution.

[0018] The vehicle en route status monitoring module / vehicle trackside status detection module distributes the solution to the maglev vehicle manager and dispatch manager, and at the same time distributes fault repair information to the vehicle maintenance dispatch module in the dispatch center to ensure rapid inspection and repair of vehicles after they enter the station.

[0019] Furthermore, the multi-dimensional fault diagnosis subsystem is specifically used for:

[0020] The vehicle maintenance scheduling module is triggered to initiate maintenance and repair processes periodically, reducing the possibility of failures and ensuring the healthy maintenance of all system components of the maglev vehicle.

[0021] Furthermore, the anomaly detection data knowledge base subsystem is specifically used for:

[0022] Establish a fault knowledge graph, screen historical inspection data of rail transit maglev vehicles, classify and collect data information according to relevant projects, and form a preliminary abnormal detection data knowledge base;

[0023] Based on the federated learning framework, while protecting the data privacy of each line, a basic feature space of components is established, fault features of multiple lines / multiple models are aggregated, and a fault feature transfer model is trained using few-shot transfer learning technology.

[0024] When abnormal data that does not match the knowledge base is detected, the self-supervised learning module automatically extracts the spatiotemporal features of the abnormal data and compares them with the fault feature transfer model to identify and mark suspected unknown faults. The spatiotemporal features include the vibration frequency and current fluctuation curve when the fault occurs.

[0025] By combining the diagnostic results of human experts, feature tags and solutions for new faults are automatically generated and updated to the anomaly detection data knowledge base in real time, realizing a closed-loop iteration of unknown fault-diagnosis-knowledge accumulation;

[0026] By analyzing and judging the on-board monitoring data and trackside detection data of maglev vehicles through the anomaly detection data knowledge base, the detected fault information is matched with the corresponding fault code. According to the different fault types, the corresponding solutions are issued and the corresponding maintenance application process is arranged.

[0027] After the maglev vehicle fault repair is completed, the anomaly detection data knowledge base updates the corresponding fault knowledge information. At the same time, the vehicle maintenance scheduling module arranges regular maintenance services for the maglev vehicle. During the maintenance process, the fault information and solutions of the vehicle will be updated to the anomaly detection data knowledge base in a synchronized manner.

[0028] Furthermore, the relevant items include the vehicle brand where the fault occurred, vehicle model, specific faulty system, specific faulty component, specific fault location, fault code, fault keywords, description of fault phenomenon, fault cause, and fault solution.

[0029] Furthermore, based on the different types of faults, solutions are distributed accordingly and corresponding workflows are arranged.

[0030] The specific requirements for a maintenance application include:

[0031] If the fault type is emergency, a temporary solution will be sent to the maglev car driving supervisor and the dispatch room supervisor via SMS. At the same time, the corresponding fault repair application and other related application processes will be initiated in the vehicle maintenance and dispatch module.

[0032] If the fault type is general, the suggested solution will be sent to the person in charge of the dispatching room via SMS. The dispatcher will then arrange the corresponding maintenance application based on the maintenance status of the maglev vehicles in the depot.

[0033] If the fault type is minor, the dispatcher will be notified via message in the vehicle maintenance dispatch module to arrange the corresponding repair application.

[0034] Furthermore, the vehicle key component and related component fault prediction subsystem is specifically used for:

[0035] Based on the fault knowledge graph, the physical connections and fault propagation paths between key components are analyzed, and a component fault propagation network is constructed. On the basis of the original single component life prediction, a time series graph network (TGN) is introduced. Real-time status data of multiple components (such as current, temperature, vibration) are input to predict the secondary faults that may be caused by the main fault and generate combined maintenance plans in advance.

[0036] After the key components of the maglev vehicle are detected and analyzed through the anomaly detection data knowledge base, a life trend map of the key components and related components is generated, including the component name, type, wear value, and moving average life trend information of the key components and related components.

[0037] Historical testing data of similar key components and related components are combined to form a life trend moving average chart of key components and related components. The current key components and related components being tested display their current wear analysis and life prediction information. If the wear value and life prediction value of key components and related components exceed the threshold, a message is sent to the person in charge of the dispatch room via SMS, and the relevant maintenance process is initiated in the vehicle maintenance dispatch module.

[0038] Furthermore, the vehicle key component and related component fault prediction subsystem is specifically used for:

[0039] After the key components of the maglev vehicle are inspected and maintained, the abnormal detection data knowledge base will update the corresponding fault knowledge information. At the same time, the various maintenance and inspection processes initiated by the vehicle maintenance scheduling module will update the health status of the key components of the maglev vehicle. During the maintenance process, various information values ​​of key components and related components will be updated to the key component life trend chart.

[0040] Furthermore, the extreme condition adaptive protection subsystem is specifically used for:

[0041] Hardware redundancy and anti-interference upgrades enhance electromagnetic interference resistance and ensure data acquisition reliability by using anti-interference sensors and redundantly deployed sensors.

[0042] The adaptive diagnostic model for operating conditions has a preset "operating condition-model parameter" mapping table, which adjusts the detection threshold and weight of the diagnostic model in real time based on environmental data; reinforcement learning is introduced to allow the system to autonomously optimize the diagnostic strategy based on historical processing experience under special operating conditions.

[0043] The emergency response plan database is integrated with cross-system collaboration to establish a special operating condition failure plan database, which includes emergency handling procedures for scenarios such as extreme weather and sudden vibrations; it connects with meteorological and geological monitoring departments to obtain early warning information on extreme weather and construction around the line in advance, automatically triggering preventive maintenance, and continuously updating the extreme operating condition response strategies for each vehicle through federated learning.

[0044] Furthermore, data acquisition through anti-interference sensors and redundantly deployed sensors specifically includes:

[0045] Fiber optic sensing technology is used to replace traditional electromagnetic sensors at key sensing nodes to improve the resistance to electromagnetic interference. Dual backup sensors are deployed near components that are susceptible to extreme temperatures to ensure the reliability of data acquisition.

[0046] Compared with the prior art, this application has the following beneficial effects:

[0047] This application addresses the problems of poor real-time performance, data silos, unpredictable fault risks, weak handling capabilities for unknown faults, and poor adaptability to special operating conditions in existing rail transit vehicle fault detection systems. It constructs a closed-loop management system to ensure timely and effective handling of vehicle faults. This achieves multi-dimensional improvements in safety assurance, operation and maintenance efficiency, service quality, and management optimization for rail transit maglev vehicles. Compared with traditional vehicle fault detection technologies, the main advantages of this invention include:

[0048] (1) Ensure safe and reliable vehicle operation. By using digital twins to map physical entities and virtual models in real time, a multi-dimensional fault diagnosis system, and an adaptive protection mechanism for extreme working conditions, the system can achieve real-time perception of vehicle status, early warning of anomalies, and rapid response, thereby reducing the risk of missed or false fault detections. Multi-layered protection is built from hardware to software to ensure the safety and stability of the vehicle under complex working conditions.

[0049] (2) Improve the efficiency and accuracy of fault handling. Based on the closed-loop fault process (fault discovery-diagnosis-repair-prevention), combined with the real-time diagnosis of digital twins, the correlation analysis of graph neural networks (GNN) and the secondary fault prediction of time sequence graph networks (TGN), the fault can be quickly located and the solution can be accurately matched, reducing vehicle downtime; at the same time, the efficiency of maintenance execution can be improved by optimizing the maintenance plan and simplifying the approval process.

[0050] (3) Optimize maintenance processes and service quality. Continuously optimize maintenance strategies by leveraging the iterative upgrades of the anomaly detection data knowledge base, cross-system knowledge transfer through federated learning, and unknown fault identification through self-supervised learning; reduce the failure rate through preventative maintenance to improve passenger travel experience and service reliability.

[0051] (4) Achieve precise vehicle management and improve efficiency and energy saving. Based on the knowledge graph of the relationship between “vehicle-environment-operation”, the impact of the environment on the lifespan of components is quantified. Combined with the degradation trajectory map of the entire life cycle of key components, the lifespan of components can be accurately predicted and maintained on demand, avoiding over-maintenance or under-maintenance. Through the virtual simulation and optimization of digital twins, the actual test cost is reduced, the energy utilization efficiency is improved, and the precise management of the entire life cycle of vehicles is achieved.

[0052] In addition to the purposes, features, and advantages described above, this application has other purposes, features, and advantages. A further detailed description of this application will be provided below with reference to the figures. Attached Figure Description

[0053] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0054] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein:

[0055] Figure 1 This is a schematic diagram of the subsystems of the rail transit vehicle fault closed-loop tracking management system according to a preferred embodiment of this application;

[0056] Figure 2 This is a schematic diagram of the operation flow of the multi-dimensional fault diagnosis subsystem according to a preferred embodiment of this application;

[0057] Figure 3 This is a schematic diagram of the operation flow of the anomaly detection data knowledge base subsystem according to a preferred embodiment of this application;

[0058] Figure 4 This is a schematic diagram of the operation process of the vehicle key component and related component fault prediction subsystem according to a preferred embodiment of this application;

[0059] Figure 5 This is a schematic diagram of the operation process of the extreme condition adaptive protection subsystem according to a preferred embodiment of this application;

[0060] Figure 6 This is a schematic block diagram of an electronic device according to a preferred embodiment of this application;

[0061] Figure 7 This is an internal structural diagram of a computer device according to a preferred embodiment of this application. Detailed Implementation

[0062] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0063] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0064] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or a rail transit vehicle fault closed-loop tracking and management device capable of performing the above functions. The following description uses a rail transit vehicle fault closed-loop tracking and management device as the executing entity to illustrate this embodiment and the subsequent embodiments.

[0065] like Figure 1As shown, a preferred embodiment of this application provides a closed-loop tracking and management system for rail transit vehicle faults, including:

[0066] A multi-dimensional fault diagnosis subsystem is used to introduce multi-physics field coupling simulation technology, construct a digital twin of the maglev vehicle, realize real-time mapping between physical entities and virtual models, and develop a fault diagnosis decision engine based on deep reinforcement learning. On the basis of the dynamic acquisition of existing real-time vehicle operation status data, trackside inspection image data, and maintenance inspection data, environmental perception data, cross-system operation data, and cross-model / cross-line data are added. A graph neural network (GNN) is used to construct a knowledge graph of "vehicle-environment-operation" association, quantify the impact weight of environmental factors (such as high temperature) on component life (such as cooling system), and improve the scenario adaptability of fault diagnosis.

[0067] The anomaly detection data knowledge base subsystem is used to effectively couple onboard monitoring data, trackside inspection data, maintenance business data, environmental perception data, cross-system operation data, and cross-model / cross-line data for data information in the operation and maintenance process of rail transit vehicles, and extract effective data to form an anomaly detection data knowledge base; at the same time, through the component health transfer learning framework, it realizes cross-model knowledge transfer, constructs a component full life cycle degradation trajectory map, introduces federated learning + self-supervised learning technology and uncertainty quantification plan, constructs cross-line and cross-model unknown fault identification, and provides prediction credibility assessment;

[0068] The vehicle key component and related component fault prediction subsystem is used to realize fault diagnosis and prediction of key components of rail transit vehicles by using intelligent diagnostic and big data analysis technology. At the same time, it introduces time sequence graph network (TGN) to construct component fault propagation network, predict secondary faults that may be caused by the main fault, generate combined maintenance solutions in advance, and provide scientific decision-making for vehicle maintenance.

[0069] The extreme working condition adaptive protection subsystem is used to build a three-dimensional mapping relationship library of "working condition-failure mode-solution" through a distributed anomaly detection framework based on federated learning, so as to ensure the real-time response of the adaptive protection system under special working conditions.

[0070] This embodiment of the rail transit vehicle fault closed-loop tracking and management system adopts a B / S architecture, including a multi-dimensional fault diagnosis subsystem, an anomaly detection data knowledge base subsystem, a fault prediction subsystem for key vehicle components and related components, and an adaptive protection subsystem for extreme operating conditions. It integrates vehicle-to-ground wireless transmission, trackside detection, and depot maintenance information transmission, deeply mining vehicle status characteristics and operating mechanisms to achieve rail transit maglev vehicle status perception and tracking, fault knowledge base fusion and upgrading, fault prediction model updates and iterations for key components and related components, and efficient response of the adaptive protection system for special operating conditions. This ensures closed-loop tracking and management of vehicle faults, forming a systematic process.

[0071] This embodiment addresses the problems of poor real-time performance, data silos, unpredictable fault risks, weak handling capabilities for unknown faults, and poor adaptability to special operating conditions in existing rail transit vehicle fault detection systems. It constructs a closed-loop management system to ensure timely and effective handling of vehicle faults. This achieves multi-dimensional improvements in the safety assurance, operation and maintenance efficiency, service quality, and management optimization of rail transit maglev vehicles. Compared with traditional vehicle fault detection technologies, the main beneficial effects of this embodiment include:

[0072] (1) Ensure safe and reliable vehicle operation. By using digital twins to map physical entities and virtual models in real time, a multi-dimensional fault diagnosis system, and an adaptive protection mechanism for extreme working conditions, the system can achieve real-time perception of vehicle status, early warning of anomalies, and rapid response, thereby reducing the risk of missed or false fault detections. Multi-layered protection is built from hardware to software to ensure the safety and stability of the vehicle under complex working conditions.

[0073] (2) Improve the efficiency and accuracy of fault handling. Based on the closed-loop fault process (fault discovery-diagnosis-repair-prevention), combined with the real-time diagnosis of digital twins, the correlation analysis of graph neural networks (GNN) and the secondary fault prediction of time sequence graph networks (TGN), the fault can be quickly located and the solution can be accurately matched, reducing vehicle downtime; at the same time, the efficiency of maintenance execution can be improved by optimizing the maintenance plan and simplifying the approval process.

[0074] (3) Optimize maintenance processes and service quality. Continuously optimize maintenance strategies by leveraging the iterative upgrades of the anomaly detection data knowledge base, cross-system knowledge transfer through federated learning, and unknown fault identification through self-supervised learning; reduce the failure rate through preventative maintenance to improve passenger travel experience and service reliability.

[0075] (4) Achieve precise vehicle management and improve efficiency and energy saving. Based on the knowledge graph of the relationship between “vehicle-environment-operation”, the impact of the environment on the lifespan of components is quantified. Combined with the degradation trajectory map of the entire life cycle of key components, the lifespan of components can be accurately predicted and maintained on demand, avoiding over-maintenance or under-maintenance. Through the virtual simulation and optimization of digital twins, the actual test cost is reduced, the energy utilization efficiency is improved, and the precise management of the entire life cycle of vehicles is achieved.

[0076] Preferably, such as Figure 2 As shown, the multi-dimensional fault diagnosis subsystem is specifically used for:

[0077] A high-fidelity digital twin model of a maglev vehicle is established, integrating a multiphysics simulation engine; real-time vehicle operation status data, trackside inspection image data, maintenance and inspection data, environmental perception data, cross-system operation data, and cross-model / cross-line data are synchronized to the digital twin model in real time to generate virtual sensor data; when the deviation between actual sensor data and virtual sensor data exceeds a threshold, deep diagnosis is triggered, and the diagnosis results are fed back to the anomaly detection data knowledge base to optimize the parameters of the digital twin model;

[0078] When the maglev vehicle is running, it transmits real-time data to the vehicle on-route status monitoring module. The vehicle on-route status monitoring module views abnormal real-time data in the digital twin model and issues an alarm, while also providing a matching solution. When the vehicle enters the depot, it transmits real-time detection data to the vehicle trackside status detection module. The detection module views abnormal detection data in the digital twin model and issues an alarm, while also matching a corresponding solution.

[0079] The vehicle on-the-way status monitoring module / vehicle trackside status detection module distributes the solution to the maglev vehicle manager and dispatch manager, and at the same time distributes fault repair information to the vehicle maintenance dispatch module of the dispatch center to ensure rapid inspection and repair of vehicles after they enter the station.

[0080] The vehicle maintenance scheduling module is triggered to initiate maintenance and repair processes periodically, reducing the possibility of failures and ensuring the healthy maintenance of all system components of the maglev vehicle.

[0081] This embodiment treats vehicle fault closed-loop tracking as a loop, monitors the health status of maglev vehicles in real time, ensures that the digital twin model of maglev vehicles can quickly identify faults and diagnose them efficiently, and continuously optimizes maintenance processes and improves service quality.

[0082] Preferably, such as Figure 3 As shown, the anomaly detection data knowledge base subsystem is specifically used for:

[0083] A fault knowledge graph is established, and historical detection data of rail transit maglev vehicles are screened. The data information is classified and collected according to relevant items to form a preliminary anomaly detection data knowledge base. The relevant items include the brand of the vehicle where the fault occurred, the vehicle model, the specific fault system, the specific fault component, the specific fault location, the fault code, the fault keywords, the fault phenomenon description, the fault cause, and the fault solution.

[0084] Based on the federated learning framework, while protecting the data privacy of each line, a basic feature space of components is established, fault features of multiple lines / multiple models are aggregated, and a fault feature transfer model is trained using few-shot transfer learning technology.

[0085] When abnormal data that does not match the knowledge base is detected, the self-supervised learning module automatically extracts the spatiotemporal features of the abnormal data and compares them with the fault feature transfer model to identify and mark suspected unknown faults. The spatiotemporal features include the vibration frequency and current fluctuation curve when the fault occurs.

[0086] By combining the diagnostic results of human experts, feature tags and solutions for new faults are automatically generated and updated to the anomaly detection data knowledge base in real time, realizing a closed-loop iteration of unknown fault-diagnosis-knowledge accumulation;

[0087] By analyzing and judging the onboard monitoring data and trackside detection data of the maglev vehicle through an anomaly detection data knowledge base, the detected fault information is matched with the corresponding fault codes. According to the different fault types, the corresponding solutions are issued and the corresponding maintenance application process is arranged.

[0088] If the fault type is emergency, a temporary solution will be sent to the maglev car driving supervisor and the dispatch room supervisor via SMS. At the same time, the corresponding fault repair application and other related application processes will be initiated in the vehicle maintenance and dispatch module.

[0089] If the fault type is general, the suggested solution will be sent to the person in charge of the dispatching room via SMS. The dispatcher will then arrange the corresponding maintenance application based on the maintenance status of the maglev vehicles in the depot.

[0090] If the fault type is minor, the dispatcher will be notified via message in the vehicle maintenance dispatch module to arrange the corresponding repair application.

[0091] After the maglev vehicle fault repair is completed, the anomaly detection data knowledge base updates the corresponding fault knowledge information. At the same time, the vehicle maintenance scheduling module arranges regular maintenance services for the maglev vehicle. During the maintenance process, the fault information and solutions of the vehicle will be updated to the anomaly detection data knowledge base in a synchronized manner.

[0092] This embodiment treats the anomaly detection data knowledge base as a loop, continuously updating the anomaly detection data knowledge base and optimizing the digital twin model parameters, thereby enhancing the applicability and reliability of the anomaly detection data knowledge base used in the maintenance process of maglev vehicles.

[0093] Preferably, such as Figure 4 As shown, the vehicle key component and related component fault prediction subsystem is specifically used for:

[0094] Based on the fault knowledge graph, the physical connections and fault propagation paths between key components are analyzed, and a component fault propagation network is constructed. On the basis of the original single component life prediction, a time series graph network (TGN) is introduced. Real-time status data of multiple components (such as current, temperature, vibration) are input to predict the secondary faults that may be caused by the main fault and generate combined maintenance plans in advance.

[0095] After the key components of the maglev vehicle are detected and analyzed through the anomaly detection data knowledge base, a life trend map of the key components and related components is generated, including the component name, type, wear value, and moving average life trend information of the key components and related components.

[0096] Historical testing data of similar key components and related components are combined to form a life trend moving average chart of key components and related components. The current key components and related components being tested display their current wear analysis and life prediction information. If the wear value and life prediction value of key components and related components exceed the threshold, a message is sent to the person in charge of the dispatch room via SMS, and the relevant maintenance process is initiated in the vehicle maintenance dispatch module.

[0097] After the key components of the maglev vehicle are inspected and maintained, the abnormal detection data knowledge base will update the corresponding fault knowledge information. At the same time, the various maintenance and inspection processes initiated by the vehicle maintenance scheduling module will update the health status of the key components of the maglev vehicle. During the maintenance process, various information values ​​of key components and related components will be updated to the key component life trend chart.

[0098] This embodiment can fully construct fault prediction for key components and related components of maglev vehicles, and iteratively update the fault prediction model to ensure the robustness and logicality of the fault prediction.

[0099] Preferably, such as Figure 5 As shown, the extreme condition adaptive protection subsystem is specifically used for:

[0100] Hardware redundancy and anti-interference upgrades enhance the ability to resist electromagnetic interference and ensure the reliability of data acquisition by acquiring data through anti-interference sensors and redundantly deployed sensors. For example, fiber optic sensing technology is used to replace traditional electromagnetic sensors at key sensing nodes to improve the ability to resist electromagnetic interference, and dual backup sensors are deployed near components that are susceptible to extreme temperatures to ensure the reliability of data acquisition.

[0101] The adaptive diagnostic model for operating conditions has a preset "operating condition-model parameter" mapping table, which adjusts the detection threshold and weight of the diagnostic model in real time based on environmental data; reinforcement learning is introduced to allow the system to autonomously optimize the diagnostic strategy based on historical processing experience under special operating conditions.

[0102] The emergency response plan database is integrated with cross-system collaboration to establish a special operating condition failure plan database, which includes emergency handling procedures for scenarios such as extreme weather and sudden vibrations; it connects with meteorological and geological monitoring departments to obtain early warning information on extreme weather and construction around the line in advance, automatically triggering preventive maintenance, and continuously updating the extreme operating condition response strategies for each vehicle through federated learning.

[0103] This embodiment can construct multi-layered protection from hardware to software for special working conditions such as strong electromagnetic interference, extreme temperature and humidity, and sudden vibration.

[0104] In summary, the rail transit vehicle fault closed-loop tracking and management system of this application has the following core features:

[0105] Key points:

[0106] (1) A fault diagnosis system integrating multiple technologies. A digital twin of the maglev vehicle is constructed based on multi-physics field coupling simulation technology to realize real-time mapping between physical entities and virtual models, supporting virtual sensor data generation and deviation-triggered deep diagnosis; a fault diagnosis decision engine integrating deep reinforcement learning is used to construct a knowledge graph of "vehicle-environment-operation" in combination with graph neural network (GNN) to quantify the influence weight of environmental factors on component life and improve the scenario adaptability of diagnosis.

[0107] (2) Cross-domain collaborative anomaly detection data knowledge base. A federated learning framework is adopted to protect data privacy, aggregate fault features of multiple lines / multi-models, and realize cross-model knowledge transfer through few-shot transfer learning; combined with self-supervised learning technology, the spatiotemporal features of unknown faults are automatically extracted to realize a closed-loop iteration of "unknown fault-diagnosis-knowledge accumulation" and continuously upgrade the knowledge base.

[0108] (3) Fault prediction mechanism for key components and related components. A time-series graph network (TGN) is introduced to construct a "component fault propagation network" to predict the "secondary faults" that may be caused by the "primary fault" and generate combined maintenance solutions to avoid fault propagation; a full life cycle degradation trajectory map of key components is constructed, and a life trend moving average map is formed by combining real-time data and historical data to achieve accurate life prediction and threshold early warning.

[0109] (4) Adaptive protection system under extreme operating conditions. Develop a distributed anomaly detection framework based on federated learning and construct a three-dimensional mapping relationship library of "operating condition-failure mode-solution"; implement multi-layer protection from hardware to software, and support autonomous optimization and cross-system collaborative response under special operating conditions.

[0110] (5) Full-process closed-loop management mechanism. Integrate the entire process of fault discovery and reporting, confirmation and registration, diagnosis and assessment, maintenance planning and execution, prevention and maintenance. Through the integrated transmission of vehicle-to-ground wireless transmission, trackside detection and depot maintenance information transmission, realize closed-loop tracking of faults from discovery to resolution, and ensure the systematicness and efficiency of the process.

[0111] Furthermore, the functions and steps of the rail transit vehicle fault closed-loop tracking management system provided in this application can be implemented by software. Therefore, another preferred embodiment of this application provides a rail transit vehicle fault closed-loop tracking management method, including the following steps:

[0112] S1. Used to introduce multi-physics coupling simulation technology, construct a digital twin of maglev vehicles, realize real-time mapping between physical entities and virtual models, and develop a fault diagnosis decision engine based on deep reinforcement learning; on the basis of dynamic collection of existing real-time vehicle operation status data, trackside inspection image data, and maintenance inspection data, add environmental perception data, cross-system operation data, and cross-model / cross-line data, and use graph neural networks to construct a knowledge graph of "vehicle-environment-operation" association, quantify the impact weight of environmental factors on component life, and improve the scenario adaptability of fault diagnosis;

[0113] S2. For data information in the operation and maintenance process of rail transit vehicles, effectively couple on-board monitoring data, trackside inspection data, maintenance business data, environmental perception data, cross-system operation data, and cross-model / cross-line data to extract effective data and form an anomaly detection data knowledge base; at the same time, through the component health transfer learning framework, cross-model knowledge transfer is realized, a component full life cycle degradation trajectory map is constructed, federated learning + self-supervised learning technology and uncertainty quantification plan are introduced to construct cross-line and cross-model unknown fault identification and provide prediction credibility assessment;

[0114] S3. Utilize intelligent diagnostic and big data analysis technologies to achieve fault diagnosis and prediction of key components of rail transit vehicles. At the same time, introduce time sequence graph networks to construct component fault propagation networks, predict secondary faults that may be caused by primary faults, generate combined maintenance plans in advance, and provide scientific decision-making for vehicle maintenance.

[0115] S4. By constructing a three-dimensional mapping relationship library of "operating condition-failure mode-solution" through a distributed anomaly detection framework based on federated learning, the adaptive protection system can respond in real time under special operating conditions.

[0116] The other sub-steps of this method correspond one-to-one with the functions of each component in the system, and will not be elaborated here. It can solve the technical problems of poor real-time performance, data silos, unpredictable fault risks, weak handling capabilities for unknown faults, and poor adaptability to special operating conditions in existing technologies. Compared with existing technologies, the beneficial effects of the rail transit vehicle fault closed-loop tracking management method provided in this application are the same as those of the rail transit vehicle fault closed-loop tracking management system provided in the above embodiments, and other technical features in the rail transit vehicle fault closed-loop tracking management device are the same as those disclosed in the methods of the above embodiments, and will not be elaborated here.

[0117] like Figure 6 As shown, a preferred embodiment of this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the closed-loop tracking and management method for rail transit vehicle faults in the above embodiments.

[0118] The electronic device provided in this application employs the closed-loop tracking and management method for rail transit vehicle faults described in the above embodiments, which can solve the technical problems of poor real-time performance, data silos, unpredictable fault risks, weak handling capabilities for unknown faults, and poor adaptability to special operating conditions in existing technologies. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the closed-loop tracking and management system for rail transit vehicle faults provided in the above embodiments, and other technical features of the electronic device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0119] like Figure 7 As shown, a preferred embodiment of this application also provides a computer device, which may be a terminal or a liveness detection server, and its internal structure diagram may be as follows. Figure 7 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The network interface is used to communicate with other external computer devices via a network connection. When the computer program is executed by the processor, it implements the steps of the aforementioned closed-loop tracking management method for rail transit vehicle faults.

[0120] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0121] The computer equipment provided in this application, employing the rail transit vehicle fault closed-loop tracking and management system described in the above embodiments, can solve the technical problems of poor real-time performance, data silos, unpredictable fault risks, weak handling capabilities for unknown faults, and poor adaptability to special operating conditions in existing technologies. Compared with the prior art, the beneficial effects of the computer equipment provided in this application are the same as those of the rail transit vehicle fault closed-loop tracking and management system provided in the above embodiments, and other technical features in the electronic equipment are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0122] A preferred embodiment of this application also provides a storage medium, the storage medium including a stored program, which, when the program is executed, controls the device where the storage medium is located to perform the steps of the rail transit vehicle fault closed-loop tracking management method in the above embodiments.

[0123] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0124] If the functions described in this embodiment are implemented as software functional units and sold or used as independent products, they can be stored in one or more computing device-readable storage media. Based on this understanding, the parts of this application's embodiments that contribute to the prior art or the technical solutions can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computing device (which may be a personal computer, server, mobile computing device, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage media include: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0125] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented using various computer languages, such as the object-oriented programming language C++ and the embedded programming language C.

[0126] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0127] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0128] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0129] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described closed-loop tracking management method for rail transit vehicle faults.

[0130] The computer program product provided in this application can solve the technical problems of poor real-time performance, data silos, unpredictable fault risks, weak handling capabilities for unknown faults, and poor adaptability to special working conditions in the prior art. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the rail transit vehicle fault closed-loop tracking management system provided in the above embodiments, and will not be repeated here.

[0131] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0132] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A closed-loop tracking and management system for faults in rail transit vehicles, characterized in that, include: A multi-dimensional fault diagnosis subsystem is used to introduce multi-physics field coupling simulation technology, construct a digital twin of the maglev vehicle, realize real-time mapping between physical entities and virtual models, and develop a fault diagnosis decision engine based on deep reinforcement learning. Based on the dynamic collection of existing real-time vehicle operation status data, trackside inspection image data, and maintenance inspection data, environmental perception data, cross-system operation data, and cross-model / cross-line data are added. A graph neural network is used to construct a knowledge graph of the relationship between "vehicle-environment-operation" to quantify the impact weight of environmental factors on component life and improve the scenario adaptability of fault diagnosis. The anomaly detection data knowledge base subsystem is used to effectively couple on-board monitoring data, trackside detection data, maintenance business data, environmental perception data, cross-system operation data, and cross-model / cross-line data for data information in the operation and maintenance process of rail transit vehicles, and extract effective data to form an anomaly detection data knowledge base. Simultaneously, through the component health transfer learning framework, cross-vehicle knowledge transfer is realized, a component full life cycle degradation trajectory map is constructed, federated learning + self-supervised learning technology and uncertainty quantification plan are introduced to construct cross-line and cross-vehicle unknown fault identification and provide prediction credibility assessment. The vehicle key component and related component fault prediction subsystem is used to realize fault diagnosis and prediction of key components of rail transit vehicles by using intelligent diagnosis and big data analysis technology. At the same time, it introduces time sequence graph network to construct component fault transmission network, predict secondary faults that may be caused by the main fault, generate combined maintenance solutions in advance, and provide scientific decision-making for vehicle maintenance. The extreme working condition adaptive protection subsystem is used to build a three-dimensional mapping relationship library of "working condition-failure mode-solution" through a distributed anomaly detection framework based on federated learning, so as to ensure the real-time response of the adaptive protection system under special working conditions.

2. The rail transit vehicle fault closed-loop tracking and management system according to claim 1, characterized in that, The multi-dimensional fault diagnosis subsystem is specifically used for: A high-fidelity digital twin model of a maglev vehicle is established, integrating a multiphysics simulation engine; real-time vehicle operation status data, trackside inspection image data, maintenance and inspection data, environmental perception data, cross-system operation data, and cross-model / cross-line data are synchronized to the digital twin model in real time to generate virtual sensor data; when the deviation between actual sensor data and virtual sensor data exceeds a threshold, deep diagnosis is triggered, and the diagnosis results are fed back to the anomaly detection data knowledge base to optimize the parameters of the digital twin model; When the maglev vehicle is running, it transmits real-time data to the vehicle on-route status monitoring module. The vehicle on-route status monitoring module views abnormal real-time data in the digital twin model and issues an alarm, while also providing a matching solution. When the vehicle enters the depot, it transmits real-time detection data to the vehicle trackside status detection module. The detection module views abnormal detection data in the digital twin model and issues an alarm, while also matching a corresponding solution. The vehicle en route status monitoring module / vehicle trackside status detection module distributes the solution to the maglev vehicle manager and dispatch manager, and at the same time distributes fault repair information to the vehicle maintenance dispatch module in the dispatch center to ensure rapid inspection and repair of vehicles after they enter the station.

3. The rail transit vehicle fault closed-loop tracking and management system according to claim 2, characterized in that, The multi-dimensional fault diagnosis subsystem is also specifically used for: The vehicle maintenance scheduling module is triggered to initiate maintenance and repair processes periodically, reducing the possibility of failures and ensuring the healthy maintenance of all system components of the maglev vehicle.

4. The rail transit vehicle fault closed-loop tracking and management system according to claim 1, characterized in that, The anomaly detection data knowledge base subsystem is specifically used for: Establish a fault knowledge graph, screen historical inspection data of rail transit maglev vehicles, classify and collect data information according to relevant projects, and form a preliminary abnormal detection data knowledge base; Based on the federated learning framework, while protecting the data privacy of each line, a basic feature space of components is established, fault features of multiple lines / multiple models are aggregated, and a fault feature transfer model is trained using few-shot transfer learning technology. When abnormal data that does not match the knowledge base is detected, the self-supervised learning module automatically extracts the spatiotemporal features of the abnormal data and compares them with the fault feature transfer model to identify and mark suspected unknown faults. The spatiotemporal features include the vibration frequency and current fluctuation curve when the fault occurs. By combining the diagnostic results of human experts, feature tags and solutions for new faults are automatically generated and updated to the anomaly detection data knowledge base in real time, realizing a closed-loop iteration of unknown fault-diagnosis-knowledge accumulation; By analyzing and judging the on-board monitoring data and trackside detection data of maglev vehicles through the anomaly detection data knowledge base, the detected fault information is matched with the corresponding fault code. According to the different fault types, the corresponding solutions are issued and the corresponding maintenance application process is arranged. After the maglev vehicle fault repair is completed, the anomaly detection data knowledge base updates the corresponding fault knowledge information. At the same time, the vehicle maintenance scheduling module arranges regular maintenance services for the maglev vehicle. During the maintenance process, the fault information and solutions of the vehicle will be updated to the anomaly detection data knowledge base in a synchronized manner.

5. The rail transit vehicle fault closed-loop tracking and management system according to claim 4, characterized in that, The relevant items include the vehicle brand, vehicle model, specific faulty system, specific faulty component, specific fault location, fault code, fault keywords, description of fault symptoms, fault cause, and fault solution.

6. The rail transit vehicle fault closed-loop tracking and management system according to claim 4, characterized in that, Depending on the type of fault, solutions will be issued accordingly, and corresponding procedures will be arranged for maintenance applications. Specific procedures include: If the fault type is emergency, a temporary solution will be sent to the maglev car driving supervisor and the dispatch room supervisor via SMS. At the same time, the corresponding fault repair application and other related application processes will be initiated in the vehicle maintenance and dispatch module. If the fault type is general, the suggested solution will be sent to the person in charge of the dispatching room via SMS. The dispatcher will then arrange the corresponding maintenance application based on the maintenance status of the maglev vehicles in the depot. If the fault type is minor, the dispatcher will be notified via message in the vehicle maintenance dispatch module to arrange the corresponding repair application.

7. The rail transit vehicle fault closed-loop tracking and management system according to claim 1, characterized in that, The vehicle key component and related component fault prediction subsystem is specifically used for: Based on the fault knowledge graph, the physical connections and fault propagation paths between key components are analyzed, and a component fault propagation network is constructed. On the basis of the original single component life prediction, a time series graph network is introduced, and real-time status data of multiple components are input to predict the secondary faults that may be caused by the main fault and generate combined maintenance plans in advance. After the key components of the maglev vehicle are detected and analyzed through the anomaly detection data knowledge base, a life trend map of the key components and related components is generated, including the component name, type, wear value, and life trend moving average information of the key components and related components; Historical testing data of similar key components and related components are combined to form a life trend moving average chart of key components and related components. The current key components and related components being tested display their current wear analysis and life prediction information. If the wear value and life prediction value of key components and related components exceed the threshold, a message is sent to the person in charge of the dispatch room via SMS, and the relevant maintenance process is initiated in the vehicle maintenance dispatch module.

8. The rail transit vehicle fault closed-loop tracking and management system according to claim 7, characterized in that, The vehicle key component and related component fault prediction subsystem is also specifically used for: After the key components of the maglev vehicle are inspected and maintained, the abnormal detection data knowledge base will update the corresponding fault knowledge information. At the same time, the various maintenance and inspection processes initiated by the vehicle maintenance scheduling module will update the health status of the key components of the maglev vehicle. During the maintenance process, various information values ​​of key components and related components will be updated to the key component life trend chart.

9. The rail transit vehicle fault closed-loop tracking and management system according to claim 1, characterized in that, The extreme condition adaptive protection subsystem is specifically used for: Hardware redundancy and anti-interference upgrades enhance electromagnetic interference resistance and ensure data acquisition reliability by using anti-interference sensors and redundantly deployed sensors. The adaptive diagnostic model for operating conditions has a preset "operating condition-model parameter" mapping table, and adjusts the detection threshold and weight of the diagnostic model in real time according to environmental data; reinforcement learning is introduced to enable the system to autonomously optimize the diagnostic strategy based on historical processing experience under special operating conditions. The emergency response plan database is integrated with cross-system collaboration to establish a special working condition fault response plan database, which includes emergency handling procedures for extreme weather and sudden vibration scenarios; By coordinating with meteorological and geological monitoring departments, we can obtain early warning information on extreme weather and construction around the line, automatically trigger preventive maintenance, and continuously update the extreme condition response strategies for each vehicle through federated learning.

10. The rail transit vehicle fault closed-loop tracking and management system according to claim 9, characterized in that, Data acquisition through anti-interference sensors and redundantly deployed sensors specifically includes: Fiber optic sensing technology is used to replace traditional electromagnetic sensors at key sensing nodes to improve the resistance to electromagnetic interference. Dual backup sensors are deployed near components that are susceptible to extreme temperatures to ensure the reliability of data acquisition.

Images