An interactive emergency response and fault diagnosis and troubleshooting system and method

The interactive emergency response and fault diagnosis system with a four-layer hierarchical architecture enables visualized configuration and multi-mode judgment for fault diagnosis and emergency response in rail transit. It solves the problems of fixed processes and single judgment in existing systems, improves efficiency and accuracy, and reduces maintenance costs.

CN123118077APending Publication Date: 2026-11-17CRRC CHANGCHUN RAILWAY VEHICLES CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611608922.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-17
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The existing rail transit fault diagnosis and emergency response system suffers from problems such as rigid troubleshooting procedures, single judgment modes, insufficient knowledge association, and limited configuration capabilities, resulting in low efficiency and insufficient accuracy.

Method used

The interactive emergency response and fault diagnosis system adopts a four-layer architecture, including the presentation layer, application layer, data layer and interface layer. It supports visual configuration, multi-mode judgment and operation and maintenance knowledge association. It realizes flexible fault diagnosis and emergency response through process configuration module, judgment execution module and knowledge association module.

Benefits of technology

It improved troubleshooting efficiency, enhanced diagnostic accuracy, improved operational support capabilities, reduced maintenance costs, and increased system adaptability, enabling it to adapt to different fault scenarios and operational needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123118077A_ABST
    Figure CN123118077A_ABST
Patent Text Reader

Abstract

This invention belongs to the field of intelligent operation and maintenance technology for rail transit, and provides an interactive emergency response and fault diagnosis and troubleshooting system and method. The system adopts a four-layer architecture of presentation layer, application layer, data layer, and interface layer, and is configured with condition configuration, process configuration, process engine, judgment execution, knowledge association, and data acquisition modules. It supports visual drag-and-drop arrangement of troubleshooting processes, and sets start, judgment, operation, and end nodes, which can realize many-to-many mapping between faults and processes. The judgment execution module has three judgment modes: automatic, semi-automatic, and manual. It performs judgments based on the data source type and drives the process branch flow. Process nodes can be bound to operation and maintenance knowledge resources such as operation and maintenance manuals and fault cases. The method includes condition configuration, process configuration, process release, fault trigger instantiation, multi-mode judgment execution, knowledge linkage display, and process end archiving steps. This invention improves troubleshooting efficiency and judgment accuracy, lowers the operation and maintenance threshold, and enhances operation and maintenance support capabilities.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of intelligent operation and maintenance technology for rail transit, specifically relating to an interactive emergency response and fault diagnosis and troubleshooting system and method. Background Technology

[0002] With the rapid development of urban rail transit networks and the continuous expansion of operational scale, the requirements for ensuring train operation safety are becoming increasingly stringent. During train operation, various equipment failures and emergencies occur frequently. How to quickly and accurately complete fault diagnosis and emergency response is directly related to operational safety, service quality, and operational efficiency.

[0003] Currently, the main problems in fault diagnosis and emergency response in the rail transit field are as follows: Troubleshooting processes are rigid and lack flexibility: Existing troubleshooting systems mostly use preset fixed processes, which cannot flexibly adjust the troubleshooting steps and judgment logic according to the actual fault scenarios, making it difficult to adapt to complex and ever-changing fault situations.

[0004] The existing systems either rely entirely on manual judgment, which is inefficient and prone to errors, or rely entirely on automatic judgment, which makes it difficult to guarantee accuracy in complex scenarios and lacks a flexible judgment mechanism for human-machine collaboration.

[0005] Insufficient knowledge connection and weak operation and maintenance support: The troubleshooting process lacks effective connection with knowledge resources such as operation and maintenance manuals and technical documents, making it difficult for troubleshooting personnel to quickly obtain relevant technical support and operation guidance, which affects troubleshooting efficiency.

[0006] Limited configuration capabilities and poor scalability: Existing systems struggle to support visualized configuration and dynamic adjustment of troubleshooting processes. Adding new troubleshooting scenarios or modifying judgment logic often requires modifying program code, resulting in high maintenance costs and long response cycles.

[0007] Therefore, there is an urgent need to develop an interactive troubleshooting system that supports visual configuration, multi-mode judgment, and knowledge association to improve the efficiency and accuracy of fault diagnosis and emergency response in rail transit. Summary of the Invention

[0008] The core objective of this invention is to realize the visual drag-and-drop configuration of the fault diagnosis process in rail transit, the custom configuration of train status parameter judgment conditions, and to support adaptive switching between three judgment modes: automatic, semi-automatic, and manual. At the same time, it enables the precise association between fault diagnosis process nodes and operation and maintenance manuals and emergency response plans, thereby constructing a visual, configurable, interactive, and intelligent rail transit fault diagnosis and emergency response system. This will improve fault diagnosis efficiency, standardization, and emergency response accuracy, while reducing the operation and maintenance threshold and the probability of human error.

[0009] To achieve the above objectives, the present invention provides an interactive emergency response and fault diagnosis and troubleshooting system, wherein the system adopts a four-layer architecture consisting of a presentation layer, an application layer, a data layer, and an interface layer. The presentation layer carries the front-end interactive interface, and deploys the front-end interactive units of the process configuration module and the judgment and execution module, as well as the knowledge association module; The application layer carries the core business scheduling and logic processing, and deploys the judgment and calculation units of the process engine module, condition configuration module, and judgment execution module. The data layer unifies the storage of process configuration data, judgment condition data, train status parameters, operation and maintenance knowledge data, and fault mapping data; The interface layer is responsible for external data integration and deploys data acquisition modules. The condition configuration module is used to support the configuration of judgment conditions based on train status parameters, including data item selection and grouping, value retrieval strategy configuration, substitution method configuration, threshold or comparison value setting, arithmetic operation, comparison operation, logical combination and condition expression editing. The process configuration module is used to provide a visual configuration interface for troubleshooting processes, supporting drag-and-drop arrangement of process nodes, setting of connection relationships, configuration of node parameters, binding of knowledge resources, and many-to-many mapping of faults to processes. The process engine module is used for parsing, validating, instantiating, scheduling, executing, and managing the status of troubleshooting processes. The knowledge association module is used to associate, read, and display process nodes with operation and maintenance knowledge resources on the front end. The judgment execution module is used to implement the execution logic of three modes: automatic judgment, semi-automatic judgment, and manual judgment. The data acquisition module is used to collect train status parameters in real time and to perform parsing, standardization, and transmission.

[0010] Furthermore, the process node types include start node, decision node, operation node, and end node.

[0011] Furthermore, the status parameters include two main categories: data collected by on-board sensors and fault data reported by the vehicle; the condition configuration module supports grouping similar status parameters into data items, and each data item can be configured with a value retrieval strategy. The value retrieval strategies include retrieval at the same location, retrieval within the same compartment, retrieval within the same unit, and retrieval of the entire vehicle, and support two substitution methods: arbitrary value substitution and substitution of all values. The arbitrary value substitution means taking any value from the set of data points determined by the value selection strategy and substituting it into the judgment condition. The judgment is valid when any value satisfies the condition. The "all substitution" refers to substituting all data point values ​​determined by the value selection strategy into the judgment condition. The judgment is valid when all values ​​meet the condition.

[0012] Furthermore, the three judgment modes in the judgment execution module are distinguished based on the data source type of the judgment conditions: the judgment conditions in the automatic judgment mode use vehicle status parameters entirely, and the system automatically obtains data and completes the judgment according to the value acquisition strategy based on the fault location; the judgment conditions in the semi-automatic judgment mode consist of vehicle status parameters and manually input data, and the manually input data is selected or entered within a preset input range. The system automatically obtains the status data and prompts the human to input before completing the judgment; the manual judgment mode allows troubleshooters to select from preset fault phenomenon description options, and the system completes the judgment based on the selection result.

[0013] Furthermore, the types of operation and maintenance knowledge resources associated with the knowledge association module include operation and maintenance manuals, fault cases, technical standards, and technical documents.

[0014] Furthermore, the conditional expressions supported by the conditional configuration module include arithmetic operations, comparison operations, and logical operations, and the logical operations include combinations of AND, OR, and NOT.

[0015] Furthermore, during the troubleshooting process, the system automatically determines the data range according to the configured value retrieval strategy based on the actual location information of the vehicle malfunction, and then completes the calculation and judgment of the judgment conditions according to the substitution method.

[0016] This invention also provides an interactive emergency response and fault diagnosis and troubleshooting method, applied to the interactive emergency response and fault diagnosis and troubleshooting system described above, comprising the following steps: Step 1: Data Item and Judgment Condition Configuration; In the condition configuration module, maintenance personnel group on-board sensors and vehicle fault parameters to construct data items, configure value retrieval strategies and substitution methods, and edit condition expressions; The system automatically pre-matches the judgment mode based on the data source within the conditions; The configured data is stored in the data layer; Step 2: Visualize the troubleshooting process configuration and associate it with faults; operations and maintenance personnel add start nodes, judgment nodes, operation nodes, and end nodes to the canvas by dragging and dropping, and connect the nodes to configure the flow relationship; judgment nodes are bound to judgment conditions or preset fault phenomenon options; operation nodes are bound to operations and maintenance knowledge resources; a many-to-many mapping relationship between faults and processes is established; configuration information is stored in the data layer. Step 3: Workflow Release; The workflow engine module performs a legality check on the workflow, including node integrity, connection logic, expression syntax, and parameter reference validity; After the check passes, the workflow is released and included in the workflow engine scheduling pool. Step 4: Fault Triggering and Process Instantiation; The data acquisition module receives fault information reported by TCMS, and the process engine module matches the fault-process mapping relationship based on the fault ID and loads the corresponding published process; The process engine module completes process instantiation, initializes the process execution status, current node, fault location parameters, binds the location information and fault parameters of this fault, and generates a unique process instance ID. Step 5: Multi-mode judgment node execution; The process engine drives the flow from the start node, and when it reaches the judgment node, it automatically reads the preset type of the node and executes the corresponding judgment logic; Step Six: Knowledge Pop-up Display; When the process flows to the judgment node or operation node that is bound to the operation and maintenance knowledge resource, the process engine module calls the knowledge association module to read the bound knowledge resource from the knowledge base, and the front-end interface provides a preview. Step 7: Process ends; the process flows to the end node, the process engine terminates the process instance; the process handling record and the execution results of each node are persistently stored to complete the troubleshooting.

[0017] Furthermore, step one specifically involves: the maintenance personnel entering train status parameters into the condition configuration interface, grouping on-board sensor data and vehicle fault data into similar parameter groups according to equipment type and carriage location, generating standardized data items, and binding them to the corresponding vehicle location tags; Configure the value retrieval strategy and substitution method for each data item, and edit custom judgment condition expressions through the system's built-in arithmetic and logical operation components to complete the threshold and logical combination configuration; The system automatically identifies the composition of the judgment condition data and completes the judgment pattern pre-matching: If the conditions only include on-board sensor data and vehicle-reported fault data, without any manually input data items or manually observed events, they will be automatically marked as automatic judgment conditions. If the condition includes vehicle data and manually entered values ​​or preset option events, it will be automatically marked as a semi-automatic judgment condition. If the conditions do not include vehicle parameters and consist only of pre-defined fault symptoms, they will be automatically marked as manual judgment conditions. The configured judgment conditions, data item rules, and value retrieval strategy parameters are synchronously stored in the data layer to form a standardized judgment condition library.

[0018] Furthermore, step two specifically involves: Operations personnel can add start nodes, decision nodes, operation nodes, and end nodes on the canvas by dragging and dropping, connect and configure the node flow relationship, and complete the basic architecture of the process. For each judgment node, select the node type: automatic, semi-automatic, or manual. When selecting automatic or semi-automatic type, the system automatically loads the data layer judgment condition library, and the operation and maintenance personnel bind the corresponding pre-configured judgment conditions to each branch. When selecting manual type, directly enter the pre-prepared fault phenomenon text and corresponding flow rules for each branch. A single process supports mixed configuration of multiple nodes and multiple judgment modes. For each operation node, link it to the operation and maintenance manual and fault case knowledge resources to complete the node knowledge binding; A fault-to-many mapping mechanism is adopted, where a single fault can be associated with multiple troubleshooting processes adapted to different scenarios, and a single process can be adapted to multiple similar faults. After the process is configured, the fault code is independently bound, a fault ID-process ID mapping relationship is established, and the data is stored in the data layer. When a fault is triggered, the corresponding process is accurately loaded through the mapping relationship. Process node configurations, node types, conditional relationships, knowledge relationships, and fault mapping relationships are all synchronously stored in the data layer.

[0019] Furthermore, step three specifically involves: The system automatically performs legality checks on the configured process, including checks on node integrity, connection logic, conditional expression syntax, and parameter reference validity. Once the checks pass, the process is published and a formal process is generated. Once published, the official process definition is synchronized to the process engine scheduling pool, awaiting a fault trigger call.

[0020] Furthermore, step four specifically involves: During train operation, the TCMS system reports fault data including fault codes, fault locations, and fault times. The data acquisition module receives vehicle fault data in real time, parses and standardizes it, and then transmits it to the process engine module. The process engine matches the fault-process mapping relationship in the data layer according to the fault ID and retrieves the corresponding published troubleshooting process. The process engine completes process instantiation, initializes process execution status, current node, and fault location parameters, binds the location information and fault parameters of this fault, and generates a unique process instance ID; Original fault data, standardized data, and process instance information are used for subsequent node judgment and scheduling.

[0021] Furthermore, the specific judgment logic in step five is as follows: Automatic Judgment Mode: The system automatically retrieves data acquisition modules, obtains vehicle status data within the corresponding range based on the fault location and the data item value retrieval strategy bound to the node, substitutes the data into the judgment condition expression according to the preset substitution method, and completes automatic calculation through the system's built-in logical operation algorithm and numerical comparison algorithm, outputs a yes / no judgment result, and automatically flows to the corresponding branch node based on the result; Semi-automatic judgment mode: The system automatically instantiates the vehicle parameter part in the judgment condition, completes vehicle data collection and calculation; at the same time, according to the configured manual input items, the system loads standardized preset input components on the front-end interface. The operation and maintenance personnel can only complete the selection or numerical input within the preset range. After the input is completed, the system automatically merges the manual data and vehicle data, substitutes them into the conditional expression to complete the unified calculation, outputs a unique judgment result and flows to the branch. Manual judgment mode: The system does not need to instantiate vehicle parameters and fault data. The front-end interface displays a pre-made, fixed and unmodifiable list of fault phenomena. Maintenance personnel select the corresponding option based on the actual observation results. The system automatically matches the judgment result and the corresponding flow branch according to the preset option-result mapping rules.

[0022] Furthermore, step six specifically involves: when the process flows to the operation node or judgment node, the process engine module actively calls the knowledge association module to match the bound operation and maintenance knowledge resources according to the current node ID. On the front-end troubleshooting interface, users can click to view the corresponding operation and maintenance manual, fault cases, and technical standards, which support online preview, retrieval, and zoom. Knowledge resource data is retrieved from the data layer knowledge base and pushed to the front-end interface for display in real time.

[0023] Furthermore, step seven specifically includes: Once all nodes have been executed, the process flows to the end node, and the process engine automatically terminates the process instance. The final process handling results and node execution results are persistently stored in the data layer, completing the entire process of troubleshooting.

[0024] Compared with the prior art, the present invention has the following beneficial effects: 1. Improve troubleshooting efficiency: Through visual configuration, troubleshooting processes can be quickly built and adjusted without modifying program code, significantly shortening the deployment cycle of troubleshooting processes; 2. Enhance judgment accuracy: The multi-mode judgment mechanism fully leverages the efficiency of automatic judgment and the accuracy of human judgment, improving judgment accuracy while ensuring efficiency; 3. Enhance operation and maintenance support capabilities: Through knowledge association, provide troubleshooting personnel with timely and accurate technical support, reduce the time spent looking up information, and improve the quality of troubleshooting; 4. Enhance system adaptability: Support dynamic process configuration and multi-mode judgment, enabling the system to adapt to changes in different fault scenarios and operational needs; 5. Reduce maintenance costs: Visual configuration capabilities enable operations and maintenance personnel to maintain troubleshooting processes independently, reducing reliance on developers and lowering system maintenance costs. Attached Figure Description

[0025] To more clearly illustrate the technical solution and advantages of the present invention, a description in conjunction with the following drawings will make it clearer and easier to understand. It should be noted that the drawings shown herein are only partial schematic diagrams related to this application, and not all of the drawings; those skilled in the art can still obtain other related drawings based on these drawings without any inventive effort.

[0026] Figure 1 This is a business process diagram of the interactive troubleshooting system for rail transit according to the present invention; Figure 2 This is a schematic diagram of the overall architecture of the interactive troubleshooting system for rail transit according to the present invention; Figure 3 This is a schematic diagram of the configuration interface for the judgment condition related data of the present invention, used to illustrate the configuration method of the value retrieval strategy and substitution method; Figure 4 This is a schematic diagram of the troubleshooting process configuration interface of the present invention, used to demonstrate the method of configuring process nodes and connection relationships in the process configuration canvas; Figure 5 This is a schematic diagram of the interface for configuring the branch binding judgment conditions of the present invention, used to demonstrate the automatic and semi-automatic methods for judging branch binding judgment conditions; Figure 6 This is a schematic diagram of the interface for linking knowledge data in this invention, used to demonstrate the operation nodes and the method for determining the knowledge data linked to the nodes; Figure 7 This is a schematic diagram of the interactive troubleshooting - manual / semi-automatic judgment interface of the present invention, used to demonstrate the method of manually selecting the judgment result during manual or semi-automatic judgment; Figure 8 This is a schematic diagram illustrating the execution logic of the automatic judgment of the present invention, used to show the execution logic of various types of judgments; Figure 9 This is a schematic diagram of the interactive troubleshooting-displaying associated knowledge file interface of the present invention, used to demonstrate the system's display method when associating knowledge files with process nodes; Figure 10 This is a schematic diagram of the interactive troubleshooting and knowledge file viewing interface of the present invention, showing the user a method for viewing the content of knowledge files. Detailed Implementation

[0027] This detailed implementation, in conjunction with the accompanying drawings, fully elucidates the system setup and deployment process, the complete timeline of the interactive troubleshooting method, the technical implementation methods for each stage, and the data flow logic. It clarifies the implementation methods of each technical feature, enabling those skilled in the art to completely replicate the system's technical solution and achieve the corresponding technical effects based on this content. It also fully explains the module linkage logic, judgment rules, and configuration rules.

[0028] To address the problems of fixed troubleshooting procedures, single judgment modes, insufficient knowledge association, and limited configuration capabilities in existing rail transit fault diagnosis and emergency response systems, this invention provides an interactive emergency response and fault diagnosis troubleshooting system and method that supports visualized configuration of troubleshooting procedures, flexible judgment in multiple modes, and association of operation and maintenance knowledge.

[0029] The interactive emergency response and fault diagnosis system described in this invention adopts a four-layer hierarchical architecture plus an independent interface layer design, combined with Figure 2 The overall system architecture diagram illustrates the completion of the system's overall setup and implementation. Details of the implementation at each level are as follows: Presentation Layer: Based on Vue + Ant Design UI, a web-based visual interface is built, including a deployment process configuration canvas, a troubleshooting execution interface, a knowledge preview interface, and a manual data entry interface. It carries all the front-end interactive functions of the process configuration module, the judgment execution module, and the knowledge association module. Interface operation commands are transmitted to the application layer via the HTTP protocol.

[0030] Application Layer: Based on the Spring Boot microservice architecture, a core business service cluster is built, including process configuration service, conditional compilation service, judgment execution service, and process scheduling service. It implements all the backend logic of the conditional configuration module and the process engine module, and is responsible for all business rule calculations, node scheduling, pattern matching, and branch determination.

[0031] Data Layer: A combination of MySQL and Influx DB is used for storage. MySQL persistently stores process definition data, judgment condition data, fault process mapping data, and operation and maintenance knowledge data; Influx DB stores real-time train status data and fault data, ensuring data read and write efficiency and providing data support for all upper-layer modules.

[0032] Interface layer: Based on TCP / IP and MVB protocols, a dedicated data interface is developed, and a data acquisition module is deployed to achieve data interoperability with the train TCMS control system and on-board sensor array, and to complete the real-time access, parsing and standardized processing of on-board data and fault data.

[0033] The process configuration module, judgment execution module, knowledge association module, process engine module, condition configuration module, and data acquisition module run according to the specified execution order and execution logic, specifically: Configuration phase: The condition configuration module completes data item management and judgment condition configuration, and the process configuration module completes basic information configuration, troubleshooting process configuration, binding judgment nodes with judgment conditions, binding operation nodes with operation and maintenance manuals, and associating processes with faults. Release phase: The process engine module verifies the validity of the configured process and releases it, then incorporates the released process into the process engine scheduling pool; Operation phase: When a fault occurs, the process engine module automatically obtains and instantiates the troubleshooting process associated with the fault, and completes the scheduling and execution of the start node, end node, judgment node, and operation node during the user's interactive troubleshooting operation. When a judgment node is executed, the judgment execution module performs a judgment based on the judgment type: In automatic judgment, it calls the data acquisition module to obtain vehicle data and automatically executes the judgment logic; in semi-automatic judgment, it automatically obtains vehicle status parameters and receives manual input or selection within a preset range, and completes the judgment after fusion; in manual judgment, it displays preset fault phenomenon options and completes the judgment based on the user's selection; after the judgment is completed, it automatically flows to the corresponding subsequent node. When a node is associated with knowledge-based files such as operation and maintenance manuals, the knowledge association module is invoked to read and display the knowledge information.

[0034] The following details the steps for implementing the full-time interactive emergency response and fault diagnosis and troubleshooting method: This invention covers the entire timeline from initial configuration, process deployment, fault triggering, node execution to final archiving. Each step clearly defines the corresponding interface, backend module, data flow, judgment rules, and implementation techniques, achieving complete implementation of technical features. The business process is detailed below. Figure 1 .

[0035] Step 1: Configure data items and judgment conditions.

[0036] 1.1 User Interface: Visual configuration interface for the condition configuration module (presentation layer); 1.2 Execution Module: Condition Configuration Module (Application Layer); 1.3 Implementation Process: First, maintenance personnel enter train status parameters into the condition configuration interface, group the on-board sensor data and vehicle fault data according to equipment type and carriage location, generate standardized data items, and bind the corresponding vehicle location tags (carriage number, unit number, equipment tag number). Secondly, configure the value retrieval strategy (same location / same carriage / same unit / whole vehicle) and substitution method (arbitrary value / all substitution) for each data item. Through the system's built-in arithmetic and logical operation components, edit custom judgment condition expressions to complete the threshold and logical combination configuration, such as... Figure 3 ; 1.4 Core Judgment Rules: The system automatically identifies the composition of judgment condition data and completes pre-matching of judgment patterns. Judgment Algorithm: Traverses the condition configuration parameter pool. If the condition only contains vehicle sensor data and vehicle-reported fault data, without any manually input data items or manually observed events, it is automatically marked as an "automatic judgment condition." If the condition contains vehicle data + manually input values / pre-set option events, it is automatically marked as a "semi-automatic judgment condition." The system background automatically completes the pattern classification and database entry, and the rules are fixed and reusable. 1.5 Data Flow: The configured judgment conditions, data item rules, and value retrieval strategy parameters are synchronously stored in the data layer condition configuration database to form a standardized judgment condition library.

[0037] Step 2: Configure and associate the visual troubleshooting process with faults.

[0038] 2.1 User Interface: Visual canvas interface for process configuration module (presentation layer); 2.2 Execution Modules: Process Configuration Module (Presentation Layer), Condition Configuration Module (Application Layer); 2.3 Implementation Process: Operations personnel add start nodes, decision nodes, operation nodes, and end nodes to the canvas by dragging and dropping, and connect the nodes to configure the flow relationships, thus completing the basic architecture of the process. Figure 4 ; For the decision node, manually select the node type (automatic / semi-automatic / manual): When selecting automatic / semi-automatic type, the system automatically loads the data layer decision condition library, and the operations and maintenance personnel bind the corresponding pre-configured decision conditions to each branch, such as... Figure 5 When selecting the manual type, there is no need to associate with the condition library; simply enter the pre-built fault phenomenon text and corresponding flow rules for each branch in the manual deployment. A single process supports mixed configuration of multiple nodes and multiple judgment modes. For each operational node, knowledge resources such as operation and maintenance manuals and fault cases are associated to complete node knowledge binding, such as... Figure 6 ; 2.4 Fault Process Association Rules: A many-to-many mapping mechanism between faults and processes is adopted. A single fault can be associated with multiple troubleshooting processes adapted to different scenarios, and a single process can be adapted to multiple similar faults. After the process is configured, a fault code is independently bound, a fault ID-process ID mapping relationship is established, and stored in the data layer mapping database. When a fault is triggered, the corresponding process is accurately loaded through the mapping relationship, rather than the fault being bound within the process itself. 2.5 Data Flow: Process node configuration, node type, condition association, knowledge association, and fault mapping are all synchronously stored in the data layer process configuration database.

[0039] Step 3: Process Release.

[0040] 3.1 User Interface: Process Configuration Interface Publishing Function Module (Presentation Layer); 3.2 Execution Module: Process Engine Module (Application Layer); 3.3 Implementation Process: The system automatically performs legality checks on the configured process, including checks on node integrity, connection logic, conditional expression syntax, and parameter reference validity. Once the checks pass, the process is published and a formal process is generated. 3.4 Data Flow: After publication, the official process definition is synchronized to the process engine scheduling pool, waiting for a fault to trigger the call.

[0041] Step 4: Fault Triggering and Process Instantiation (corresponding to) Figure 8 (Execution flow diagram).

[0042] 4.1 Triggering conditions: During train operation, the TCMS system reports fault information such as fault code, fault location, and fault time; 4.2 Execution Modules: Data Acquisition Module (Interface Layer), Process Engine Module (Application Layer); 4.3 Implementation Process: The data acquisition module receives vehicle fault data in real time, parses and standardizes it, and then transmits it to the process engine module. The process engine matches the fault-process mapping relationship in the data layer according to the fault ID and retrieves the corresponding published troubleshooting process. The process engine completes process instantiation, initializes process execution status, current node, and fault location parameters, binds the location information and fault parameters of this fault, and generates a unique process instance ID; 4.4 Data flow: original fault data, standardized data, and process instance information are used for subsequent node judgment and scheduling.

[0043] Step 5: Execute the multi-mode judgment node.

[0044] The process engine module drives the process from the start node to the judgment node. When it reaches the judgment node, it automatically reads the preset type of the node and executes the corresponding judgment logic. All judgment logic is calculated by the system's preset program without human subjective intervention. 5.1 The automatic judgment mode is implemented and executed. The system automatically calls the data acquisition module, accurately obtains the vehicle status data within the corresponding range according to the fault location and the data item value retrieval strategy bound to the node, substitutes it into the judgment condition expression according to the preset substitution method, and completes automatic calculation through the system's built-in logical operation algorithm (AND / OR / NOT) and numerical comparison algorithm, outputs the yes / no judgment result, and automatically flows to the corresponding branch node according to the result, without human intervention throughout the process; 5.2 The semi-automatic judgment mode is implemented. The system automatically instantiates the vehicle parameter part of the judgment conditions and completes vehicle data collection and calculation. At the same time, according to the configured manual input items, standardized preset input components (fixed options / numerical input boxes, the content of which are preset and cannot be customized) are loaded on the front-end interface. Maintenance personnel can only complete the selection or numerical input within the preset range. After the input is completed, the system automatically merges the manual data and vehicle data, substitutes them into the conditional expression to complete the unified calculation, outputs a unique judgment result and flows to the branch. The human only undertakes the input action and does not participate in the judgment logic. 5.3 In the manual judgment mode, the system does not need to instantiate vehicle parameters and fault data. The front-end interface displays a pre-defined, fixed, and unmodifiable list of fault phenomena (options are based on standard maintenance settings and have no custom input). Maintenance personnel select the corresponding option based on actual observations. The system automatically matches the judgment result with the corresponding workflow branch according to the preset option-result mapping rules. The result is unique and controllable, with no subjective judgment bias. Figure 7 .

[0045] Step 6: Knowledge pop-up window interactive display.

[0046] 6.1 Execution Modules: Knowledge Association Module (Presentation Layer), Process Engine Module (Application Layer); 6.2 Implementation Process: When the workflow reaches the operation node or judgment node, the workflow engine actively calls the knowledge association module to match the bound operation and maintenance knowledge resources based on the current node ID. On the front-end troubleshooting interface, users can click to view the corresponding operation and maintenance manuals, fault cases, and technical standards. Online previewing, searching, and zooming are supported without needing to consult across systems. Displaying and opening associated operation and maintenance manuals is as follows: Figure 9 , Figure 10 ; 6.3 Data flow: Knowledge resource data is retrieved from the knowledge base in the data layer and pushed to the front-end interface for display in real time.

[0047] Step 7: Process ends.

[0048] 7.1 Implementation Rules: After all nodes have been executed, the process flows to the end node, and the process engine automatically terminates the process instance; 7.2 Data flow: The final process handling results and node execution results are persistently stored in the data layer, completing the entire process of troubleshooting.

[0049] As a specific implementation method, the process visualization configuration is as follows: The process visualization configuration module is developed based on an open-source process editor to achieve the following functionalities: Node panel: The left side displays four types of nodes, which can be dragged and dropped to the canvas with one click. Node properties are associated with basic parameters by default. Canvas area: Supports node drag positioning, multi-point connection, batch editing, undo and redo, automatically verifies connection logic, and eliminates isolated nodes and invalid branches; Attribute panel: The right side displays the configuration parameters of the selected node in real time, and supports modifying the node name, judgment type, association conditions, and knowledge resources.

[0050] As a specific implementation method, the judgment condition configuration is as follows: 1. Status parameter types and data item grouping: The system builds a standardized train status parameter library, uniformly collecting two types of core parameters. The parameters are classified and marked by data tags. A structured grouping algorithm is used to aggregate parameters of the same equipment, type, and dimension into a single data item and bind it to the location identifier. This provides a data foundation for subsequent value retrieval strategies. Those skilled in the art can directly use this grouping rule to complete parameter configuration. 2. Data item value selection strategies and substitution methods are implemented. The system has built-in fixed filtering algorithms for four value selection strategies and logical operation algorithms for two substitution methods. The algorithms are fixed and cannot be modified. After a fault is triggered, the system automatically parses the fault location label, matches the corresponding value selection strategy to the data range to be filtered, and then performs OR / AND logical operations according to the substitution method. The entire process is automated, requiring no manual intervention, and the judgment result is unique. 3. The implementation of conditional expressions uses the open-source expression parsing engine (QLExpress) to implement custom conditional operations. It supports regular comparison operations and logical operations, and can nest and combine complex conditional expressions. The engine automatically completes syntax verification, operation execution, and result output, ensuring the accuracy and stability of condition judgment.

[0051] As a specific implementation method, the standardization of multi-mode judgment is implemented as follows: The core of the three judgment modes in this invention is standardized prefabrication, system judgment, and manual input only, which completely avoids the interference of human factors on the technical results: 1. All manually entered content and options are preset fixed content by the system, with no custom editing or subjective judgment permissions; 2. All decision-making logic, data operations, and branching processes are executed by the system's preset algorithms and expression rules; 3. Under the same fault scenario and the same parameter configuration, the system output results are completely consistent, demonstrating a high degree of reproducibility; 4. Manual operation is only an auxiliary information collection action, does not participate in technical judgment and logical operation, and does not affect the final technical effect of the system.

[0052] As a specific implementation method, knowledge association is as follows: 1. Knowledge resource management is implemented by building an operation and maintenance knowledge base that supports batch uploading, version updates, management and operation and maintenance manuals, fault cases, technical standards, equipment drawings and other resources, which are stored uniformly in the data layer and resource index tags are established to achieve rapid matching; 2. Knowledge association and display implementation adopts a node-knowledge one-to-many binding mechanism. The association and binding are completed in the configuration stage, and the knowledge loading is triggered by the process engine in the process execution stage. The front end displays the knowledge in a pop-up window in real time, supporting retrieval, preview and download. There is no need to jump across systems, realizing the synchronous linkage between troubleshooting and knowledge support.

[0053] As a specific implementation method, the working process of the data acquisition module is as follows: 1. Data access and communication implementation: The system connects to the train's TCMS system via MVB bus and onboard Ethernet dual communication protocols to receive onboard sensor data and fault data in real time, ensuring the real-time performance and integrity of data transmission. 2. Data parsing and standardization implementation: Develop multi-protocol parsing adapters to adapt to data formats of different vehicle models and subsystems, perform standardized parsing according to vehicle data protocols, and transform raw data into a unified standard data format for the system, providing accurate data support for judgment and calculation.

[0054] As a specific implementation method, the workflow engine module works as follows: 1. Process parsing and instantiation: After the process engine loads the troubleshooting process corresponding to the fault, it automatically completes node parsing, relationship verification, parameter initialization, binds the location, type, and parameter information of this fault, generates an independent process instance, isolates the execution process of different faults, and avoids data interference. 2. Process scheduling and execution are implemented using an event-driven scheduling mechanism. Each node is executed sequentially according to its topological relationship. The system automatically matches branch flows based on the system calculation results of the judgment node. After the operation node is completed, manual confirmation is required to proceed to the next node. The entire scheduling logic is solidified and orderly, with no chaotic flow issues.

[0055] The core of this invention is to avoid the problem of technical deviations caused by human factors. All manual operations are standardized and pre-configured input selections, with no subjective judgment or logical intervention. The system's final judgment results, process flow, and data calculations are all automatically executed by preset program rules, conditional expressions, and value retrieval strategies. Humans only perform information input actions and do not change the system's inherent technical logic or output results. This strictly conforms to natural laws and preset technical rules, ensuring that the technical solution's output results are unique, controllable, and reproducible, meeting the core requirement of patented technical solutions that conform to natural laws.

[0056] The core innovations of this invention include: 1. Innovation Point 1: Visual configuration of troubleshooting process.

[0057] A graphical workflow editor is used, allowing for the visual configuration of troubleshooting workflows through node drag-and-drop and connection arrangement. Workflow node types include: (1) Start node: Defines the entry point of the process; (2) Decision node: Implements condition judgment and branch flow, and supports multi-mode judgment; (3) Operation node: Define the specific troubleshooting operation steps; (4) End node: Defines the process exit.

[0058] 2. Innovation Point Two: Judgment Condition Configuration and Data Item Value Retrieval Strategy Based on State Parameters.

[0059] (1) The state parameters described in this invention include two main types of data sources: Onboard sensor data acquisition: Analog and digital data collected in real time by sensors of various subsystems of the train, including temperature, pressure, voltage, current, speed, displacement, switch status, etc., reflecting the real-time operating status of various equipment in the vehicle; Fault data reported by the vehicle: Fault information automatically diagnosed and reported by the train's onboard control system (such as TCMS), including fault code, fault level, fault occurrence time, fault subsystem and location, etc., reflecting the abnormal and fault status identified by the vehicle.

[0060] (2) The judgment condition configuration supports the following features: Data Item Grouping: Similar state parameters are structured and grouped into data items. Each data item can contain one or more sensor acquisition points or fault data points of the same type. For example, the brake cylinder pressure sensors of all compartments in the vehicle are grouped into the "Brake Cylinder Pressure" data item, and the status signals of each door are grouped into the "Door Status" data item. Value retrieval strategy configuration: When configuring judgment conditions for each data item, the value retrieval strategy can be selected, including four strategies: value retrieval at the same location, value retrieval in the same carriage, value retrieval in the same unit, and value retrieval for the whole vehicle. It also supports selecting any value to substitute or substituting all values. Logical Combinations: Supports AND, OR, and NOT logical combinations of multiple conditions to construct complex judgment logic; Expression editing: Supports editing and validating custom conditional expressions.

[0061] (3) The detailed explanation of the value selection strategy is shown in Table 1 below: Table 1. Value Selection Strategy Value retrieval strategy Data range illustrate Values ​​at the same position The corresponding data point at the location where the fault occurred. It only acquires data from sensors or devices at the same location where the fault occurred, making it suitable for scenarios requiring precise fault location. Value taken from the same carriage All data points of the carriage where the fault occurred. It can acquire all similar data points within the carriage where the fault is located, and is suitable for carriage-level fault analysis and judgment. Values ​​within the same unit All data points of the train unit where the fault occurred. It can acquire all similar data points within the grouping unit where the fault is located, and is suitable for unit-level fault analysis and judgment. Whole vehicle value All data points for the entire vehicle. It acquires all similar data points across the entire vehicle, making it suitable for comprehensive judgment and comparative analysis at the vehicle level. Under any of the above value selection strategies, the system further supports two substitution methods: Arbitrary value substitution: Randomly select a value from the set of data points determined by the value selection strategy and substitute it into the judgment condition for calculation. When any value meets the condition, it is determined to be true (logical OR). It is suitable for scenarios that quickly screen for the existence of anomalies. Substitute all: Substitute all data point values ​​determined by the value retrieval strategy into the judgment condition one by one or in batches for calculation. The condition is only considered true when all values ​​meet the condition (logical AND). This is suitable for scenarios with strict verification. When executing the troubleshooting process, the system automatically obtains the vehicle status data within the corresponding range according to the configured value retrieval strategy and substitution method based on the actual location information of the vehicle fault, substitutes it into the judgment conditions to complete the calculation and judgment, and achieves accurate matching between fault location and judgment conditions.

[0062] 3. Innovation Point Three: Multi-mode Judgment Mechanism.

[0063] The decision node supports three decision modes, which are distinguished according to the data source type used in the decision conditions, as shown in Table 2 below: Table 2. Differentiation of Judgment Modes Judgment Pattern Data composition of judgment conditions Implementation mechanism Automatic judgment The judgment criteria are based entirely on vehicle status parameters (sensor-collected data and fault data reported by the vehicle), and do not include any data items that require manual input. The system automatically obtains vehicle status data according to the fault location and value retrieval strategy, substitutes it into the judgment conditions to complete the calculation and judgment, and can determine the process flow direction without manual intervention. Semi-automatic judgment The judgment criteria consist of vehicle status parameters and manually input data such as observation and inspection results that cannot be automatically collected from the vehicle. The system automatically acquires the collectable status data and prompts troubleshooters to input observation or inspection results that cannot be automatically acquired. It then combines the two sets of data to automatically calculate the judgment criteria. Manual judgment The judgment conditions cannot be automatically calculated from vehicle status data and must rely on the observation of fault phenomena (such as foreign objects on the surface) by troubleshooting personnel. The system provides a pre-defined list of fault description options. Troubleshooters manually select the corresponding option based on the observed phenomena. The system then determines the judgment result based on the user's selection and completes the subsequent process flow. The core difference between the three modes is as follows: Automatic judgment can be entirely automated, requiring no human intervention; semi-automatic judgment requires manual supplementation of data that cannot be automatically collected before the system completes the calculations; manual judgment relies entirely on manual selection of pre-defined phenomenon description options to drive the judgment. These three modes can be flexibly configured according to actual fault scenarios, enabling efficient troubleshooting through human-machine collaboration. To avoid interference from human factors in the technical results, all manual inputs and option selections in this invention are standardized pre-configured, with no subjective judgment authority. The final results of semi-automatic and manual judgments are uniquely determined by the system's preset rules, conditional expressions, and option mapping logic. Humans only perform information entry operations and do not participate in the judgment logic calculations, ensuring that the output results of the technical solution are stable, reproducible, and fully comply with natural laws and technical rule constraints.

[0064] 4. Innovation Point Four: Connection of Operation and Maintenance Knowledge.

[0065] Each process node can be linked to relevant operation and maintenance knowledge resources to provide immediate technical support for troubleshooting personnel. Operation and Maintenance Manual Association: Associates with relevant equipment operation and maintenance manual chapters to provide standard operating instructions; Fault Case Association: Associates similar historical fault cases to provide reference solutions; Technical document association: Associate related technical standards, drawings, manuals, and other documents.

Claims

1. An interactive emergency response and fault diagnosis and troubleshooting system, characterized in that, The system adopts a four-layer architecture consisting of a presentation layer, an application layer, a data layer, and an interface layer. The presentation layer deployment process configuration module, the front-end interaction unit of the judgment and execution module, and the knowledge association module; The judgment and calculation unit of the application layer deployment process engine module, condition configuration module, and judgment execution module; The data layer uniformly stores process configuration data, judgment condition data, train status parameters, operation and maintenance knowledge data, and fault mapping data. The interface layer deploys a data acquisition module; The condition configuration module is used to support the configuration of judgment conditions based on train status parameters, including data item selection and grouping, value retrieval strategy configuration, substitution method configuration, threshold or comparison value setting, arithmetic operation, comparison operation, logical combination and condition expression editing. The process configuration module is used to provide a visual configuration interface for troubleshooting processes, supporting drag-and-drop arrangement of process nodes, setting of connection relationships, configuration of node parameters, binding of knowledge resources, and many-to-many mapping of faults to processes. The process engine module is used for parsing, validating, instantiating, scheduling, executing, and managing the status of troubleshooting processes. The knowledge association module is used to associate, read, and display process nodes with operation and maintenance knowledge resources on the front end. The judgment execution module is used to implement the execution logic of three modes: automatic judgment, semi-automatic judgment, and manual judgment. The data acquisition module is used to collect train status parameters in real time and to perform parsing, standardization, and transmission.

2. The interactive emergency response and fault diagnosis system according to claim 1, characterized in that, The process nodes include start node, decision node, operation node, and end node.

3. The interactive emergency response and fault diagnosis system according to claim 1, characterized in that, The status parameters include two main categories: data collected by on-board sensors and fault data reported by the vehicle. The condition configuration module supports grouping similar status parameters into data items. Each data item can be configured with a value selection strategy, which includes value selection at the same location, value selection in the same compartment, value selection in the same unit, and value selection for the whole vehicle. It also supports two substitution methods: arbitrary value substitution and all value substitution. The arbitrary value substitution means taking any value from the set of data points determined by the value selection strategy and substituting it into the judgment condition. The judgment is valid when any value satisfies the condition. The "all substitution" refers to substituting all data point values ​​determined by the value selection strategy into the judgment condition. The judgment is valid when all values ​​meet the condition.

4. The interactive emergency response and fault diagnosis system according to claim 1, characterized in that, The distinction between the three judgment modes in the judgment execution module is based on the data source type of the judgment conditions: the judgment conditions of the automatic judgment mode use vehicle status parameters entirely, and the system automatically obtains data and completes the judgment according to the value acquisition strategy based on the fault location. The judgment conditions in the semi-automatic judgment mode consist of vehicle status parameters and manually input data. The manually input data is selected or entered within the preset input range. The system automatically obtains the status parameters and prompts the human to input them before completing the judgment. In manual diagnostic mode, troubleshooters select from preset fault description options, and the system makes a judgment based on the selection.

5. The interactive emergency response and fault diagnosis system according to claim 1, characterized in that, The knowledge association module associates operation and maintenance knowledge resources including operation and maintenance manuals, fault cases, technical standards, and technical documents.

6. The interactive emergency response and fault diagnosis system according to claim 1, characterized in that, The conditional expressions supported by the conditional configuration module include arithmetic operations, comparison operations, and logical operations, and the logical operations include combinations of AND, OR, and NOT.

7. The interactive emergency response and fault diagnosis system according to claim 3, characterized in that, When executing the troubleshooting process, the system automatically determines the data range according to the configured value retrieval strategy based on the actual location information of the vehicle malfunction, and then completes the calculation and judgment of the judgment conditions according to the substitution method.

8. An interactive emergency response and fault diagnosis and troubleshooting method, characterized in that, An interactive emergency response and fault diagnosis troubleshooting system as described in any one of claims 1 to 7, comprising the following steps: Step 1: Data Item and Judgment Condition Configuration; In the condition configuration module, maintenance personnel group on-board sensors and vehicle fault parameters to construct data items, configure value retrieval strategies and substitution methods, and edit condition expressions; The system automatically pre-matches the judgment mode based on the data source within the conditions; The configured data is stored in the data layer; Step 2: Configure and associate the visual troubleshooting process with faults; Operation and maintenance personnel can add start nodes, judgment nodes, operation nodes, and end nodes on the canvas by dragging and dropping, and connect the nodes to configure the flow relationship; judgment nodes are bound to judgment conditions or preset fault phenomenon options; Operation nodes are bound to maintenance knowledge resources; a many-to-many mapping relationship between faults and processes is established; configuration information is stored in the data layer. Step 3: Workflow Release; The workflow engine module performs a legality check on the workflow, including node integrity, connection logic, expression syntax, and parameter reference validity; After the check passes, the workflow is released and included in the workflow engine scheduling pool. Step 4: Fault Triggering and Process Instantiation; The data acquisition module receives fault information reported by TCMS, and the process engine module matches the fault-process mapping relationship based on the fault ID and loads the corresponding published process. The process engine module completes process instantiation, initializes process execution status, current node, and fault location parameters, binds the location information and fault parameters of this fault, and generates a unique process instance ID. Step 5: Multi-mode judgment node execution; The process engine drives the flow from the start node, and when it reaches the judgment node, it automatically reads the preset type of the node and executes the corresponding judgment logic; Step Six: A pop-up window displays related knowledge information. When the process flows to the judgment node or operation node bound to the operation and maintenance knowledge resource, the process engine module calls the knowledge association module to read the bound knowledge resource from the knowledge base, and the front-end interface provides a preview. Step 7: Process complete; When the process reaches the end node, the process engine terminates the process instance. Persistently store process handling records and execution results of each node to complete troubleshooting.

9. The interactive emergency response and fault diagnosis and troubleshooting method according to claim 8, characterized in that, Step one specifically involves: the maintenance personnel entering train status parameters into the condition configuration interface, grouping on-board sensor data and vehicle fault data into similar parameter groups according to equipment type and carriage location, generating standardized data items, and binding them to the corresponding vehicle location tags. Configure the value retrieval strategy and substitution method for each data item, and edit custom judgment condition expressions through the system's built-in arithmetic and logical operation components to complete the threshold and logical combination configuration; The system automatically identifies the composition of the judgment condition data and completes the judgment pattern pre-matching: If the conditions only include on-board sensor data and vehicle-reported fault data, without any manually input data items or manually observed events, they will be automatically marked as automatic judgment conditions. If the condition includes vehicle data and manually entered values ​​or preset option events, it will be automatically marked as a semi-automatic judgment condition. If the conditions do not include vehicle parameters and consist only of pre-defined fault symptoms, they will be automatically marked as manual judgment conditions. The configured judgment conditions, data item rules, and value retrieval strategy parameters are synchronously stored in the data layer to form a standardized judgment condition library.

10. The interactive emergency response and fault diagnosis and troubleshooting method according to claim 8, characterized in that, Step two specifically involves: Operations personnel can add start nodes, decision nodes, operation nodes, and end nodes on the canvas by dragging and dropping, connect and configure the node flow relationship, and complete the basic architecture of the process. For each judgment node, select the node type: automatic, semi-automatic, or manual. When selecting automatic or semi-automatic type, the system automatically loads the data layer judgment condition library, and the operation and maintenance personnel bind the corresponding pre-configured judgment conditions to each branch. When selecting manual type, directly enter the pre-prepared fault phenomenon text and corresponding flow rules for each branch. A single process supports mixed configuration of multiple nodes and multiple judgment modes. For each operation node, link it to the operation and maintenance manual and fault case knowledge resources to complete the node knowledge binding; A fault-to-many mapping mechanism is adopted, where a single fault can be associated with multiple troubleshooting processes adapted to different scenarios, and a single process can be adapted to multiple faults of the same type. After the process is configured, the fault code is independently bound, a fault ID-process ID mapping relationship is established and stored in the data layer, and the corresponding process is accurately loaded through the mapping relationship when a fault is triggered. Process node configurations, node types, conditional relationships, knowledge relationships, and fault mapping relationships are all synchronously stored in the data layer.

11. The interactive emergency response and fault diagnosis and troubleshooting method according to claim 8, characterized in that, Step three specifically involves: The system automatically performs legality checks on the configured process, including checks on node integrity, connection logic, conditional expression syntax, and parameter reference validity. Once the checks pass, the process is published and a formal process is generated. Once published, the official process definition is synchronized to the process engine scheduling pool, awaiting a fault trigger call.

12. The interactive emergency response and fault diagnosis and troubleshooting method according to claim 8, characterized in that, Step four specifically involves: During train operation, the TCMS system reports fault data including fault codes, fault locations, and fault times. The data acquisition module receives vehicle fault data in real time, parses and standardizes it, and then transmits it to the process engine module. The process engine matches the fault-process mapping relationship in the data layer according to the fault ID and retrieves the corresponding published troubleshooting process. The process engine completes process instantiation, initializes process execution status, current node, and fault location parameters, binds the location information and fault parameters of this fault, and generates a unique process instance ID; Original fault data, standardized data, and process instance information are used for subsequent node judgment and scheduling.

13. The interactive emergency response and fault diagnosis and troubleshooting method according to claim 8, characterized in that, The specific judgment logic in step five is as follows: Automatic Judgment Mode: The system automatically retrieves data acquisition modules, obtains vehicle status data within the corresponding range based on the fault location and the data item value retrieval strategy bound to the node, substitutes the data into the judgment condition expression according to the preset substitution method, and completes automatic calculation through the system's built-in logical operation algorithm and numerical comparison algorithm, outputs a yes / no judgment result, and automatically flows to the corresponding branch node based on the result; Semi-automatic judgment mode: The system automatically instantiates the vehicle parameter part in the judgment condition, completes vehicle data collection and calculation; at the same time, according to the configured manual input items, the system loads standardized preset input components on the front-end interface. The operation and maintenance personnel can only complete the selection or numerical input within the preset range. After the input is completed, the system automatically merges the manual data and vehicle data, substitutes them into the conditional expression to complete the unified calculation, outputs a unique judgment result and flows to the branch. Manual judgment mode: The system does not need to instantiate vehicle parameters and fault data. The front-end interface displays a pre-made, fixed and unmodifiable list of fault phenomena. Maintenance personnel select the corresponding option based on the actual observation results. The system automatically matches the judgment result and the corresponding flow branch according to the preset option-result mapping rules.

14. The interactive emergency response and fault diagnosis and troubleshooting method according to claim 8, characterized in that, Step six specifically involves the following: When the process flows to the operation node or judgment node, the process engine module actively calls the knowledge association module to match the bound operation and maintenance knowledge resources according to the current node ID. The corresponding operation and maintenance manual, fault cases, and technical standards can be viewed on the front-end troubleshooting interface, and online preview, retrieval, and zoom are supported. Knowledge resource data is retrieved from the data layer knowledge base and pushed to the front-end interface for display in real time.

15. The interactive emergency response and fault diagnosis and troubleshooting method according to claim 8, characterized in that, Step seven specifically involves: Once all nodes have been executed, the process flows to the end node, and the process engine automatically terminates the process instance. The final process handling results and node execution results are persistently stored in the data layer, completing the entire process of troubleshooting.