A field section turnout intelligent control and information management method and system

Abstract

The application provides a field section turnout intelligent control and information management method and system, relates to the technical field of computers, and comprises the following steps: acquiring basic data sets in a target field section; performing data fusion according to the basic data sets to obtain equipment real-time state information; performing health diagnosis according to the equipment real-time state information to obtain equipment health degree evaluation results; performing intelligent decision-making according to the equipment health degree evaluation results to obtain adaptive control decision-making instructions; performing digital twin verification according to the adaptive control decision-making instructions to obtain optimized safety control instructions; and performing edge execution according to the optimized safety control instructions, transmitting the instructions to local nodes through a collaborative computing architecture to obtain a closed-loop control state of field section operation. The application realizes full-process integrated intelligent management and control of field section operation from state perception, intelligent diagnosis, active decision-making to safe execution, and significantly improves the reliability, safety and operation and maintenance efficiency of the system.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and more specifically, to a method and system for intelligent control and information management of track turnouts in a track section. Background Technology

[0002] In the field of intelligent management of rail transit depots, traditional track automation systems have long undertaken core tasks such as turnout control and signal monitoring. However, with the increasing variety of depot operations and the continuous improvement of informatization levels, these systems have gradually shown significant limitations in terms of scalability, compatibility, and stability. Existing technologies mainly rely on fixed-logic interlocking control and basic sensing equipment to achieve automated operation. Although this improves operational efficiency to some extent, it faces problems such as difficulty in compatibility with emerging electronic components, reliance on dedicated simulation panels and professional personnel for system debugging, the need to interrupt depot operations for main and backup software updates and the existence of external access security risks, and low functional integration due to limited communication interfaces between heterogeneous devices. These issues result in insufficient overall system adaptability, high maintenance costs, and poor operational stability in complex multi-tasking scenarios, making it difficult to meet the needs of modern depots for efficient, flexible, and safe intelligent operation and maintenance.

[0003] Based on the shortcomings of the existing technologies, there is an urgent need for a method and system for intelligent control and information management of track turnouts. Summary of the Invention

[0004] The purpose of this invention is to provide a method for intelligent control and information management of track turnouts in track sections, so as to improve the above-mentioned problems. To achieve the above objective, the technical solution adopted by this invention is as follows:

[0005] Firstly, this application provides a method for intelligent control and information management of track turnouts in a track section, including:

[0006] Acquire the basic dataset within the target section. The basic dataset includes current timing data during switch machine operation, signal filament status data, track circuit status data, magnetic field change data generated by vehicle electromagnetic induction, and locomotive number data based on image recognition.

[0007] Data fusion is performed based on the aforementioned basic dataset. By integrating current, state, and magnetic field data, real-time status information of the device is obtained.

[0008] Based on the real-time status information of the equipment, health diagnosis is performed, and the current time series data is dynamically analyzed through a time series model to realize turnout fault prediction and signal abnormality detection, and obtain the equipment health assessment result.

[0009] Intelligent decision-making is carried out based on the equipment health assessment results. By combining interlocking logic conditions, vehicle number consistency verification and rule engine, a route control strategy is generated to obtain adaptive control decision instructions.

[0010] Digital twin verification is performed based on the adaptive control decision command. The propagation path of the control command and the device response are deduced through a preset simulation model to obtain the optimized safety control command.

[0011] The optimized safety control instructions are executed at the edge, and the instructions are transmitted to the local node through a collaborative computing architecture to obtain the closed-loop control status of the field operation.

[0012] Secondly, this application also provides a turnout intelligent control and information management system for railway sections, including:

[0013] The acquisition module is used to acquire the basic dataset within the target field section. The basic dataset includes current timing data during the operation of the switch machine, signal filament status data, track circuit status data, magnetic field change data generated by the vehicle through electromagnetic induction, and locomotive number data based on image recognition.

[0014] The fusion module is used to perform data fusion based on the basic dataset, and obtain the real-time status information of the device by integrating the current, state and magnetic field data;

[0015] The diagnostic module is used to perform health diagnosis based on the real-time status information of the equipment, and to perform dynamic analysis of current time series data through a time series model to realize turnout fault prediction and signal abnormality detection, and obtain equipment health assessment results.

[0016] The decision module is used to make intelligent decisions based on the equipment health assessment results. By integrating interlocking logic conditions, vehicle number consistency verification and rule engine, it generates route control strategies and obtains adaptive control decision instructions.

[0017] The verification module is used to perform digital twin verification based on the adaptive control decision command, and to deduce the propagation path of the control command and the device response through a preset simulation model to obtain the optimized safety control command.

[0018] The control module is used to perform edge execution according to the optimized safety control instructions, and transmit the instructions to the local node through a collaborative computing architecture to obtain the closed-loop control status of the field operation.

[0019] The beneficial effects of this invention are as follows:

[0020] This invention acquires and fuses multi-source heterogeneous data within the field to construct unified real-time equipment status information. Based on a time-series model, it achieves intelligent diagnosis and prediction of equipment health, integrates health status, interlocking logic, and vehicle number verification for adaptive decision-making, and verifies the security of control commands through digital twin simulation. Finally, it achieves reliable execution of commands and closed-loop status feedback through edge-side collaborative computing. This enables integrated intelligent management and control of the entire field operation process, from status perception, intelligent diagnosis, proactive decision-making to safe execution, significantly improving the system's reliability, security, and operational efficiency. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of a method for intelligent control and information management of track turnouts in an embodiment of the present invention;

[0023] Figure 2 This is a schematic diagram of the structure of a turnout intelligent control and information management system in a track section, as described in an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the vehicle license plate recognition module;

[0025] Figure 4 This is a diagram illustrating the architecture of the simulated learning module.

[0026] Figure 5 This is a diagram of the cloud-edge-device collaborative processing layer architecture.

[0027] The diagram is labeled as follows: 901, Acquisition Module; 902, Fusion Module; 903, Diagnosis Module; 904, Decision Module; 905, Verification Module; 906, Control Module. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0029] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0030] Example 1:

[0031] This embodiment provides a method for intelligent control and information management of turnouts in railway sections.

[0032] See Figure 1 The figure shows that the method includes steps S100 to S600.

[0033] Step S100: Obtain the basic dataset within the target section. The basic dataset includes current timing data during the operation of the switch machine, signal filament status data, track circuit status data, magnetic field change data generated by the vehicle through electromagnetic induction, and locomotive number data based on image recognition.

[0034] The core of this step lies in building a comprehensive perception layer, aiming to systematically solve the problem of real-time capture of equipment status and vehicle dynamics within the field. Specifically, this step synchronously collects high-frequency current timing data during switch machine operation, status data of signal filament circuits, and existing status information of track circuits. Combined with RFID readers deployed at key nodes and edge computing cameras with built-in lightweight AI recognition models, multimodal perception fusion technology is used to reliably identify locomotives and vehicles. Finally, all sensor data is aggregated to the on-site edge computing node via IoT protocols. This comprehensive data collection system, from physical signals to identity information, lays a complete and reliable data foundation for subsequent intelligent processing.

[0035] Step S200: Perform data fusion based on the basic dataset. By integrating current, state, and magnetic field data, obtain real-time status information of the device.

[0036] Understandably, given the diverse types of equipment and varying data interfaces and protocols within the field, directly using raw data would lead to complex processing logic and a high risk of errors. By integrating information such as current, status, and magnetic field through data fusion technology, a unified data model is essentially constructed. This eliminates information silos and generates real-time equipment status information that accurately reflects the overall operational status of the field, providing structured input for high-level analysis and decision-making.

[0037] Step S300: Perform health diagnosis based on real-time equipment status information, dynamically analyze current time sequence data through time sequence model, realize turnout fault prediction and signal abnormality detection, and obtain equipment health assessment results.

[0038] It's important to note that in real-world scenarios, malfunctions in critical equipment such as turnouts and signals directly impact operational safety and efficiency, and traditional periodic maintenance or reactive repair methods are inherently inadequate. By using time-series models to perform in-depth analysis of dynamic data such as current, subtle signs of equipment performance degradation can be detected, allowing for prediction and early warning before malfunctions occur. Essentially, this elevates equipment maintenance strategies from reactive response to proactive prevention.

[0039] Step S400: Make intelligent decisions based on the equipment health assessment results. Generate route control strategies by integrating interlocking logic conditions, vehicle number consistency verification and rule engine, and obtain adaptive control decision instructions.

[0040] Understandably, the essence of this step is to build an intelligent decision-making engine that integrates real-time equipment health status, thereby improving the adaptability and safety of route control. Traditional interlocking logic mainly relies on fixed safety conditions (such as track occupancy and correct switch positions), lacking dynamic assessment of the equipment's own reliability. By collaboratively analyzing health assessment results with car number verification and traditional interlocking conditions, the control strategy not only meets basic safety rules but also proactively avoids potentially risky equipment or optimizes resource allocation. This generates adaptive control decisions that can adapt to the real-time state of the system, addressing the challenges of constantly changing plans and dynamic equipment status within the depot.

[0041] Step S500: Perform digital twin verification based on the adaptive control decision command, and deduce the propagation path of the control command and the device response through a preset simulation model to obtain the optimized safety control command;

[0042] It should be noted that this step introduces digital twin technology as a virtual testbed for control commands. Its core value lies in performing safety verification before actual execution. The operational environment of rail transit depots is complex, and misoperation of control commands can trigger a chain of safety issues. By simulating the entire execution process of commands in a highly realistic virtual environment, potential logical conflicts or abnormal equipment responses in the command propagation path can be detected in advance. This "verify before execution" model adds a crucial safety barrier to control commands, making it particularly suitable for rail transit depot scenarios with stringent safety and reliability requirements.

[0043] Step S600: Perform edge execution according to the optimized safety control instructions, and transmit the instructions to the local node through the collaborative computing architecture to obtain the closed-loop control status of the field operation.

[0044] Understandably, this step completes the final link from intelligent decision-making to physical execution, focusing on reliably implementing instructions and managing the closed loop. If intelligent decision-making results cannot accurately and promptly drive field equipment, a value loop cannot be formed. Through edge computing capabilities, optimized safety instructions are parsed into executable drive signals for the equipment, and multiple execution units are coordinated to work together. Finally, feedback signals are collected to confirm the execution effect, forming a complete control loop. This design ensures that intelligent analysis results can accurately and reliably impact the physical world, ultimately achieving stable and efficient operation of the field.

[0045] Further, step S200 includes steps S210 to S220.

[0046] Step S210: Perform data preprocessing based on the basic dataset. By parsing the IoT communication protocol and removing abnormal data points, standardize the heterogeneous raw sensor data to obtain a regular data stream with unified spatiotemporal reference.

[0047] Step S220: Perform feature extraction based on the regular data stream. Extract turnout operation features by analyzing the amplitude and phase changes of the current time-series waveform, and extract vehicle displacement features by analyzing the intensity changes of the magnetic field induction signal, thereby obtaining a feature vector set reflecting the state of the physical object.

[0048] Step S230: Perform state fusion based on the feature vector set, bind the identity and location by associating the vehicle number recognition result with the vehicle displacement feature, and coordinate the turnout action feature and signal status to construct the interlocking logic input to obtain the real-time status information of the equipment.

[0049] Specifically, the process begins with data preprocessing, which involves parsing and cleaning the heterogeneous raw sensor data aggregated to edge computing nodes via IoT protocols. This includes parsing the communication protocol formats of different devices, removing sampling anomalies, and converting current, state, and magnetic field data into a standardized data stream with a unified spatiotemporal reference, thus resolving data format inconsistencies and noise issues. Next, feature extraction is performed. For the standardized data stream, the amplitude and phase characteristics of the switch machine's current time-series waveform at different action stages (such as start-up, switching, and locking) are analyzed to extract the switch action characteristics reflecting its mechanical performance. Simultaneously, the variation patterns of the magnetic field signal intensity collected by electromagnetic sensors are analyzed to extract the vehicle's displacement characteristics on the track, forming a feature vector set representing the key states of the physical object. Finally, state fusion is completed. Through multimodal perception fusion technology, the vehicle number recognition result is bound to the vehicle displacement characteristics. Specifically, this is achieved through locomotive number recognition devices deployed at the depot entrance and exit points. The processing flow of the vehicle number recognition device is as follows: Figure 3 As shown, cameras deployed at this location use AI video structured analysis technology and neural network learning models to analyze the relevant attributes of vehicles entering the monitoring area. The identified vehicle type and license plate information is transmitted to the license plate recognition module in binary code form. After conversion, it is displayed on the human-machine interface. At the same time, the recognition result is correlated in real time with the vehicle displacement features obtained by electromagnetic sensors during the state fusion stage, thereby accurately establishing a dynamic correspondence between vehicle identity and spatial location. On this basis, it further coordinates with multi-source information such as turnout action features and signal filament status to jointly construct complete input information that meets the interlocking logic requirements, and finally generates real-time equipment status information that can be used for intelligent decision-making.

[0050] Further, step S300 includes steps S310 to S320.

[0051] Step S310: Based on the current timing data in the real-time status information of the equipment, feature extraction is performed. By identifying the start-up, action and locking phases in the current waveform, the switch machine characteristic parameters that characterize the mechanical characteristics of the switch machine are extracted.

[0052] Step S320: Perform state deduction based on the switch machine characteristic parameters. By comparing the real-time characteristic parameters with the preset health benchmark model, identify the characteristic deviation and obtain the preliminary fault probability index.

[0053] Step S330: Based on the preliminary failure probability index, a comprehensive evaluation is conducted. By combining the continuity status of the signal filament current with historical operating data, the failure probability is weighted and corrected and trend is judged to obtain a quantitative equipment health assessment result.

[0054] Specifically, steps S310 to S330 together constitute the progressive processing flow of equipment health diagnosis. First, based on the current timing data in the real-time equipment status information, key characteristic parameters characterizing its mechanical properties (such as frictional resistance and switching time) are extracted by identifying typical phases of the current waveform during the switch machine motor's operation, such as startup, operation, and locking. Next, a state deduction is performed, comparing the extracted characteristic parameters with a preset health benchmark model (representing the parameter range under normal operating conditions). The degree of equipment performance degradation is quantified by calculating the characteristic deviation, thus obtaining a preliminary failure probability index. Finally, a comprehensive evaluation is completed. By combining the continuity status of the signal filament current (such as current interruption or fluctuation) with historical operating data (such as cumulative number of operations and maintenance records), the preliminary failure probability is weighted and corrected (e.g., assigning higher weight to frequently operated equipment) and trend judgment is performed (analyzing the parameter change rate). Finally, a quantitative equipment health assessment result is output, providing a basis for subsequent decision-making.

[0055] Further, step S400 includes steps S410 to S420.

[0056] Step S410: Determine the safety status based on the equipment health assessment results. By comparing the health status with a preset threshold, the equipment status is divided into safety, warning, or fault levels, and mapped to the corresponding interlocking logic constraints to obtain an enhanced safety constraint set.

[0057] Step S420: Perform route feasibility analysis based on the enhanced safety constraint set and vehicle number identification data. Verify the consistency between the vehicle number and the planned route. Under the premise of satisfying the vehicle number verification, filter and modify the traditional interlocking conditions based on the safety constraint set to obtain a set of feasible routes.

[0058] Step S430: Generate and optimize strategies based on the set of feasible routes. Evaluate the efficiency, priority and equipment health trend of each route through the rule engine, select the optimal route and generate a control strategy containing adaptive adjustment instructions to obtain adaptive control decision instructions.

[0059] Specifically, firstly, based on the equipment health assessment results, the equipment status is classified into different levels such as safe, warning, or fault by comparing them with preset thresholds. These levels are then mapped to specific interlocking logic constraints, forming an enhanced set of safety constraints. Next, route feasibility analysis is performed. Based on the enhanced set of safety constraints and vehicle number identification data, the consistency between the vehicle number and the planned route is verified. Under the premise of satisfying the vehicle number verification, traditional interlocking conditions are filtered and modified to obtain a set of feasible routes. Finally, strategy generation and optimization are completed. The rule engine evaluates the operating efficiency, operation priority, and equipment health trends of each feasible route, selects the optimal route, and generates a control strategy containing adaptive adjustment instructions. Ultimately, adaptive control decision instructions that can adapt to the real-time system status are output.

[0060] Further, step S500 includes steps S510 to S520.

[0061] Step S510: Map the virtual environment according to the adaptive control decision command. By synchronously inputting the control command and the real-time status information of the equipment into the preset simulation model, a dynamic virtual scene consistent with the physical field is constructed to obtain the initial state of the twin for simulation.

[0062] Step S520: Based on the initial state of the twin, perform deduction, execution and monitoring, calculate the chain logic reaction and equipment action sequence triggered by the control command in the virtual scene, and monitor the state changes of the virtual signal machine, switch machine and track circuit in real time to obtain the complete system response trajectory and potential conflict points;

[0063] Step S530: Perform safety verification and optimization based on the system response trajectory and potential conflict points. By analyzing whether there are any violations of interlocking rules, equipment over-limit operation, or route conflicts in the trajectory, modify the original control commands or generate alternative commands to obtain optimized safety control commands.

[0064] Understandably, steps S510 to S530 together constitute the digital twin verification process before the control command is actually executed. First, based on the adaptive control decision command, a dynamic virtual scenario is constructed by synchronously inputting the command and the current real-time equipment status information into a preset high-fidelity simulation model. This constructs a scenario completely consistent with the physical section in terms of equipment layout, interlocking relationships, and real-time status, thus obtaining a digital twin with accurate initial conditions for subsequent simulations. Next, simulation execution and monitoring are performed. Based on the initialized twin state, the simulation model calculates a series of chain logic reactions triggered by the control command in the virtual scenario (e.g., operating a turnout will trigger changes in the display of related signals and lock the corresponding track section) and the standard action sequence of the equipment. During this process, the status changes of the virtual signals, switch machines, and track circuits are monitored in real time, resulting in a complete virtual system response trajectory containing a time series, and identifying potential conflict points (such as contradictory signal displays, route setting conflicts, etc.). Finally, safety verification and optimization are completed. Through in-depth analysis of the obtained system response trajectory and potential conflict points, it is checked whether there are violations of established interlocking rules, equipment actions exceeding limits (such as abnormal switch machine switching time), or route resource conflicts. Based on this, the original control commands are logically modified, timing adjusted, or new alternative commands are generated. Finally, the safety control commands that have been virtually verified and optimized are output to ensure the safety and reliability during actual execution to the greatest extent.

[0065] Preferably, the digital twin upon which the above verification mechanism relies is constructed based on cloud-based big data and real-time data streams, possessing high-fidelity simulation capabilities. The system's operating mechanism is as follows: Figure 4The architecture illustrates this: to start the simulation system, the simulation unit must first be activated. At this point, the actual operating system automatically shuts down and disconnects from the physical PLC. Subsequently, the status data generated by virtual control units such as the signal control unit, switch machine control unit, and electromagnetic sensor control unit are processed by the input / output analysis unit and the analog quantity processing unit. Finally, the status of each virtual device is dynamically displayed on the operating console, thereby simulating the on-site operation situation with high precision. In this virtual environment, operations such as route setting, single-operation of turnouts, control of signals, input of car numbers, and occupation of track circuits can be safely performed (for example, reverse commands can be executed to verify logic when setting a route), effectively saving on-site debugging costs. This twin supports multiple modes at the application layer: In training mode, instructors can inject specific faults into the twin, and the system can deduce the propagation path and impact range of the fault in real time. It can also be combined with VR technology to provide trainees with an immersive emergency response training environment, greatly improving the learning experience and training effect. In the operation and maintenance mode, operation and maintenance personnel can trace back the historical status change sequence of the equipment on the digital twin and intuitively view key indicators such as the remaining useful life (RUL) predicted by the AI ​​model. This provides intuitive data support for the formulation of accurate and forward-looking maintenance plans, thereby expanding the value of digital twin technology beyond security assurance to simulation debugging, teaching and training, and intelligent operation and maintenance.

[0066] Further, step S600 includes steps S610 to S620.

[0067] Step S610: Based on the optimized safety control instructions, perform instruction parsing and driving. Through the edge computing node, the high-order control strategy is parsed into specific device driving commands and converted into control signals of the corresponding actuators to obtain the underlying instruction set that can directly drive the device.

[0068] Step S620: Perform distributed execution and synchronization based on the underlying instruction set. Send the instruction set in parallel to local execution units such as the switch machine intelligent controller and signal intelligent controller via the fieldbus to coordinate the execution of actions by each device according to the preset timing sequence and obtain the collaborative action status of the device group.

[0069] Step S630: Perform status feedback and closed-loop confirmation based on the coordinated action status. By collecting the turnout position, signal display and track circuit status after execution in real time, and comparing them with the expected status, verify the completion of the control command and obtain the closed-loop control status.

[0070] It should be noted that the above process begins with instruction parsing and driving. Edge computing nodes receive safety control instructions (typically high-level route control strategies) optimized through digital twin verification. By parsing their logical meaning, they convert these instructions into specific, executable drive commands. Furthermore, they generate level or digital signals conforming to the interface specifications of actuators such as switch machine intelligent controllers and signal intelligent controllers, forming a low-level instruction set that can directly drive the equipment. Next, distributed execution and synchronization are performed. The low-level instruction set is reliably distributed in parallel to local execution units distributed throughout the field section via a fieldbus network. During this process, the system strictly coordinates the timing of actions of each execution unit to ensure that the switch machine, signal, and other equipment groups can collaboratively and orderly complete the actions required by the control commands, thereby obtaining the collaborative action status of the equipment group. Finally, status feedback and closed-loop confirmation are completed. Real-time data is collected on the actual status feedback of key equipment after execution (such as the actual position of switches collected by sensors, the actual display color of signals, and the occupancy / idle status of track circuits), and these actual states are accurately compared with the expected states of the control commands. If the status is consistent, the command is confirmed to have been executed successfully, and the system enters a new stable operating state. If a deviation occurs, a fault alarm and handling mechanism is triggered. Ultimately, this feedback comparison mechanism verifies the completion of the control command, realizes closed-loop control of the entire field operation status, and ensures that the intelligent decision results are accurately and reliably executed in the physical world.

[0071] The reliability and intelligence of the aforementioned edge execution process benefit from, for example, Figure 5 The cloud-edge-device collaborative processing layer architecture is shown. In this architecture, the cloud (centralized and intelligent) undertakes the functions of centralized optimization and continuous learning: the cloud receives anonymized data uploaded from edge nodes across all segments of the network, stores it in a big data platform for aggregation, and then uses a deep learning engine to analyze the massive amount of aggregated historical data to train more accurate fault prediction and health management models. Through model management functions, the optimized models are regularly upgraded to all edge nodes via OTA, enabling the lightweight AI diagnostic models deployed on the edge to continuously evolve, thereby improving the fault prediction accuracy and adaptability of the entire system. In addition, the hash values ​​of all key operation commands and alarm logs generated on the edge are uploaded to the cloud for notarization via blockchain technology, thus establishing an immutable security audit trace chain. This cloud-edge collaborative mechanism not only ensures the real-time performance and reliability of edge control, but also endows the system with the ability for global optimization and continuous evolution.

[0072] Example 2:

[0073] like Figure 2 As shown in the figure, this embodiment provides a smart control and information management system for track turnouts in a track section. The system includes:

[0074] The acquisition module 901 is used to acquire the basic dataset within the target field section. The basic dataset includes current timing data during the operation of the switch machine, signal filament status data, track circuit status data, magnetic field change data generated by the vehicle through electromagnetic induction, and locomotive number data based on image recognition.

[0075] The fusion module 902 is used to perform data fusion based on the basic dataset. By integrating current, state, and magnetic field data, it obtains real-time status information of the device.

[0076] The diagnostic module 903 is used to perform health diagnosis based on the real-time status information of the equipment. It performs dynamic analysis of current time sequence data through a time sequence model to realize turnout fault prediction and signal abnormality detection, and obtain equipment health assessment results.

[0077] The decision module 904 is used to make intelligent decisions based on the equipment health assessment results. By integrating interlocking logic conditions, vehicle number consistency verification and rule engine, it generates route control strategies and obtains adaptive control decision instructions.

[0078] The verification module 905 is used to perform digital twin verification based on the adaptive control decision command. It uses a preset simulation model to deduce the propagation path of the control command and the device response to obtain the optimized safety control command.

[0079] The control module 906 is used to perform edge execution based on the optimized safety control instructions, and transmit the instructions to the local node through a collaborative computing architecture to obtain the closed-loop control status of the field operation.

[0080] In one specific embodiment of this application, the fusion module 902 includes:

[0081] The first fusion unit is used to perform data preprocessing based on the basic dataset. By parsing the IoT communication protocol and removing abnormal data points, it standardizes the heterogeneous raw sensor data to obtain a regular data stream with unified spatiotemporal reference.

[0082] The second fusion unit is used to extract features based on the regular data stream. It extracts turnout operation features by analyzing the amplitude and phase changes of the current timing waveform and extracts vehicle displacement features by analyzing the intensity changes of the magnetic field induction signal, thus obtaining a feature vector set that reflects the state of the physical object.

[0083] The third fusion unit is used to perform state fusion based on the feature vector set. It binds the identity and location by associating the vehicle number recognition result with the vehicle displacement feature, and coordinates the turnout action feature and signal status to construct the interlocking logic input to obtain the real-time status information of the equipment.

[0084] In one specific embodiment of this application, the diagnostic module 903 includes:

[0085] The first diagnostic unit is used to extract features based on the current timing data in the real-time status information of the equipment. By identifying the start-up, action and locking phases in the current waveform, it extracts the switch machine characteristic parameters that characterize the mechanical characteristics of the switch machine.

[0086] The second diagnostic unit is used to perform state deduction based on the switch machine's characteristic parameters. By comparing the real-time characteristic parameters with the preset health benchmark model, it identifies the degree of characteristic deviation and obtains preliminary fault probability indicators.

[0087] The third diagnostic unit is used to conduct a comprehensive assessment based on preliminary failure probability indicators. By combining the continuity status of the signal filament current with historical operating data, it performs weighted correction and trend judgment on the failure probability to obtain a quantitative equipment health assessment result.

[0088] In one specific embodiment of this application, the decision module 904 includes:

[0089] The first decision unit is used to determine the safety status based on the equipment health assessment results. By comparing the health status with a preset threshold, the equipment status is divided into safety, warning or fault levels, and mapped to the corresponding interlocking logic constraints to obtain an enhanced safety constraint set.

[0090] The second decision-making unit is used to perform route feasibility analysis based on the enhanced safety constraint set and vehicle number identification data. By verifying the consistency between the vehicle number and the planned route, and under the premise of satisfying the vehicle number verification, it filters and corrects the traditional interlocking conditions based on the safety constraint set to obtain a set of feasible routes.

[0091] The third decision unit is used to generate and optimize strategies based on a set of feasible routes. It evaluates the efficiency, priority, and equipment health trends of each route through a rule engine, selects the optimal route, and generates a control strategy containing adaptive adjustment instructions to obtain adaptive control decision instructions.

[0092] In one specific embodiment of this application, the verification module 905 includes:

[0093] The first verification unit is used to map the virtual environment according to the adaptive control decision command. By synchronously inputting the control command and the real-time status information of the equipment into the preset simulation model, a dynamic virtual scene consistent with the physical field is constructed to obtain the initial state of the twin for deduction.

[0094] The second verification unit is used to perform deduction, execution and monitoring based on the initial state of the twin. It calculates the chain logic reaction and equipment action sequence triggered by the control command in the virtual scene, and monitors the state changes of the virtual signal machine, switch machine and track circuit in real time to obtain the complete system response trajectory and potential conflict points.

[0095] The third verification unit is used to perform safety verification and optimization based on the system response trajectory and potential conflict points. By analyzing whether there are any violations of interlocking rules, equipment over-limit operation, or route conflicts in the trajectory, the original control commands are modified or alternative commands are generated to obtain optimized safety control commands.

[0096] In one specific embodiment of this application, the control module 906 includes:

[0097] The first control unit is used to parse and drive the optimized safety control instructions. It parses the high-order control strategy into specific device drive commands through the edge computing node and converts them into control signals for the corresponding actuators to obtain the underlying instruction set that can directly drive the device.

[0098] The second control unit is used for distributed execution and synchronization based on the underlying instruction set. It sends the instruction set in parallel to local execution units such as the switch machine intelligent controller and the signal intelligent controller via the fieldbus, and coordinates each device to perform actions according to the preset timing sequence to obtain the coordinated action status of the device group.

[0099] The third control unit is used to provide status feedback and closed-loop confirmation based on the coordinated action status. It collects the turnout position, signal display and track circuit status after execution in real time, compares them with the expected status, verifies the completion of the control command, and obtains the closed-loop control status.

[0100] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent control and information management of track turnouts in a track section, characterized in that, include: Acquire the basic dataset within the target section. The basic dataset includes current timing data during switch machine operation, signal filament status data, track circuit status data, magnetic field change data generated by vehicle electromagnetic induction, and locomotive number data based on image recognition. Data fusion is performed based on the aforementioned basic dataset. By integrating current, state, and magnetic field data, real-time status information of the device is obtained. Based on the real-time status information of the equipment, health diagnosis is performed, and the current time series data is dynamically analyzed through a time series model to realize turnout fault prediction and signal abnormality detection, and obtain the equipment health assessment result. Intelligent decision-making is carried out based on the equipment health assessment results. By combining interlocking logic conditions, vehicle number consistency verification and rule engine, a route control strategy is generated to obtain adaptive control decision instructions. Digital twin verification is performed based on the adaptive control decision command. The propagation path of the control command and the device response are deduced through a preset simulation model to obtain the optimized safety control command. The optimized safety control instructions are executed at the edge, and the instructions are transmitted to the local node through a collaborative computing architecture to obtain the closed-loop control status of the field operation.

2. The intelligent control and information management method for track turnouts according to claim 1, characterized in that, Data fusion is performed based on the aforementioned basic dataset. By integrating current, state, and magnetic field data, real-time device status information is obtained, including: Data preprocessing is performed based on the aforementioned basic dataset. By parsing the IoT communication protocol and removing outlier data points, the heterogeneous raw sensor data is standardized to obtain a regular data stream with a unified spatiotemporal reference. Feature extraction is performed based on the regularized data stream. The turnout operation features are extracted by analyzing the amplitude and phase changes of the current timing waveform, and the vehicle displacement features are extracted by analyzing the intensity changes of the magnetic field induction signal, thus obtaining a feature vector set that reflects the state of the physical object. State fusion is performed based on the feature vector set. The vehicle identification result is associated with the vehicle displacement feature to bind the identity and location. The turnout action feature and signal status are coordinated to construct the interlocking logic input to obtain the real-time status information of the equipment.

3. The intelligent control and information management method for track turnouts according to claim 1, characterized in that, Based on the real-time status information of the equipment, health diagnosis is performed. Dynamic analysis of current time-series data is conducted using a time-series model to predict turnout faults and detect signal anomalies, resulting in an equipment health assessment, including: Feature extraction is performed based on the current timing data in the real-time status information of the equipment. By identifying the start-up, action, and locking phases in the current waveform, characteristic parameters of the switch machine that characterize its mechanical properties are extracted. Based on the switch machine's characteristic parameters, state deduction is performed. By comparing the real-time characteristic parameters with a preset health benchmark model, the degree of characteristic deviation is identified, and a preliminary fault probability index is obtained. Based on the preliminary failure probability index, a comprehensive evaluation is conducted. By combining the continuity status of the signal filament current with historical operating data, the failure probability is weighted and corrected, and a trend judgment is made to obtain a quantitative equipment health assessment result.

4. The intelligent control and information management method for track turnouts according to claim 1, characterized in that, Intelligent decision-making is performed based on the equipment health assessment results. By comprehensively considering interlocking logic conditions, vehicle number consistency verification, and a rule engine, a route control strategy is generated, resulting in adaptive control decision instructions, including: Based on the equipment health assessment results, the safety status is determined. By comparing the health status with a preset threshold, the equipment status is divided into safety, warning, or fault levels, and mapped to the corresponding interlocking logic constraints to obtain an enhanced safety constraint set. Based on the enhanced safety constraint set and vehicle number identification data, a route feasibility analysis is performed. By verifying the consistency between the vehicle number and the planned route, and under the premise of satisfying the vehicle number verification, the traditional interlocking conditions are screened and modified based on the safety constraint set to obtain a set of feasible routes. Based on the set of feasible routes, strategies are generated and optimized. The efficiency, priority, and device health trend of each route are evaluated through a rule engine. The optimal route is selected and a control strategy containing adaptive adjustment instructions is generated to obtain adaptive control decision instructions.

5. The intelligent control and information management method for track turnouts according to claim 1, characterized in that, Digital twin verification is performed based on the adaptive control decision command. The propagation path of the control command and the device response are deduced through a preset simulation model to obtain optimized safety control commands, including: The virtual environment is mapped according to the adaptive control decision command. By synchronously inputting the control command and the real-time status information of the equipment into the preset simulation model, a dynamic virtual scene consistent with the physical field is constructed to obtain the initial state of the twin for deduction. Based on the initial state of the twin, the system is deduced, executed, and monitored. By calculating the chain logic reactions and equipment action sequences triggered by control commands in the virtual scene, and by monitoring the state changes of the virtual signal machine, switch machine, and track circuit in real time, a complete system response trajectory and potential conflict points are obtained. Based on the system response trajectory and the potential conflict points, safety verification and optimization are performed. By analyzing whether there are any violations of interlocking rules, equipment over-limit operation, or route conflicts in the trajectory, the original control commands are modified or alternative commands are generated to obtain optimized safety control commands.

6. A smart control and information management system for track turnouts in a track section, characterized in that, include: The acquisition module is used to acquire the basic dataset within the target field section. The basic dataset includes current timing data during the operation of the switch machine, signal filament status data, track circuit status data, magnetic field change data generated by the vehicle through electromagnetic induction, and locomotive number data based on image recognition. The fusion module is used to perform data fusion based on the basic dataset, and obtain the real-time status information of the device by integrating the current, state and magnetic field data; The diagnostic module is used to perform health diagnosis based on the real-time status information of the equipment, and to perform dynamic analysis of current time series data through a time series model to realize turnout fault prediction and signal abnormality detection, and obtain equipment health assessment results. The decision module is used to make intelligent decisions based on the equipment health assessment results. By integrating interlocking logic conditions, vehicle number consistency verification and rule engine, it generates route control strategies and obtains adaptive control decision instructions. The verification module is used to perform digital twin verification based on the adaptive control decision command, and to deduce the propagation path of the control command and the device response through a preset simulation model to obtain the optimized safety control command. The control module is used to perform edge execution according to the optimized safety control instructions, and transmit the instructions to the local node through a collaborative computing architecture to obtain the closed-loop control status of the field operation.

7. The intelligent control and information management system for track turnouts according to claim 6, characterized in that, The fusion module includes: The first fusion unit is used to perform data preprocessing based on the basic dataset. By parsing the Internet of Things communication protocol and removing abnormal data points, the heterogeneous raw sensor data is standardized to obtain a regular data stream with unified spatiotemporal reference. The second fusion unit is used to extract features based on the regularized data stream. It extracts turnout operation features by analyzing the amplitude and phase changes of the current timing waveform and extracts vehicle displacement features by analyzing the intensity changes of the magnetic field induction signal, thereby obtaining a feature vector set that reflects the state of the physical object. The third fusion unit is used to perform state fusion based on the feature vector set, bind the identity and position by associating the vehicle number recognition result with the vehicle displacement feature, and coordinate the turnout action feature and signal status to construct the interlocking logic input to obtain the real-time status information of the equipment.

8. The intelligent control and information management system for track turnouts according to claim 6, characterized in that, The diagnostic module includes: The first diagnostic unit is used to extract features based on the current timing data in the real-time status information of the equipment, and extract switch machine characteristic parameters that characterize the mechanical characteristics of the switch machine by identifying the start-up, action and locking phases in the current waveform. The second diagnostic unit is used to perform state deduction based on the switch machine characteristic parameters. By comparing the real-time characteristic parameters with the preset health benchmark model, it identifies the characteristic deviation and obtains a preliminary fault probability index. The third diagnostic unit is used to conduct a comprehensive evaluation based on the preliminary fault probability index. By combining the continuity status of the signal filament current with historical operating data, the fault probability is weighted and corrected and trend is judged to obtain a quantitative equipment health assessment result.

9. The intelligent control and information management system for track turnouts according to claim 6, characterized in that, The decision-making module includes: The first decision unit is used to determine the safety status based on the equipment health assessment results. By comparing the health status with a preset threshold, the equipment status is divided into safety, warning or fault levels, and mapped to the corresponding interlocking logic constraints to obtain an enhanced safety constraint set. The second decision unit is used to perform route feasibility analysis based on the enhanced safety constraint set and vehicle number identification data. By verifying the consistency between the vehicle number and the planned route, and under the premise of satisfying the vehicle number verification, the traditional interlocking conditions are screened and modified based on the safety constraint set to obtain a set of feasible routes. The third decision unit is used to generate and optimize strategies based on the set of feasible routes. It evaluates the efficiency, priority and equipment health trend of each route through a rule engine, selects the optimal route and generates a control strategy containing adaptive adjustment instructions, thereby obtaining adaptive control decision instructions.

10. The intelligent control and information management system for track turnouts according to claim 6, characterized in that, The verification module includes: The first verification unit is used to perform virtual environment mapping according to the adaptive control decision command. By synchronously inputting the control command and real-time status information of the device into the preset simulation model, a dynamic virtual scene consistent with the physical field is constructed to obtain the initial state of the twin for deduction. The second verification unit is used to perform deduction, execution and monitoring based on the initial state of the twin. It calculates the chain logic reaction and equipment action sequence triggered by the control command in the virtual scene, and monitors the state changes of the virtual signal machine, switch machine and track circuit in real time to obtain the complete system response trajectory and potential conflict points. The third verification unit is used to perform safety verification and optimization based on the system response trajectory and the potential conflict points. By analyzing whether there are any violations of interlocking rules, equipment over-limit operation, or route conflicts in the trajectory, the original control commands are modified or alternative commands are generated to obtain optimized safety control commands.

Images