Information acquisition and transmission system of track control equipment

Abstract

The invention discloses an information acquisition and transmission system of track control equipment, which relates to the technical field of track traffic intelligent monitoring and comprises an intelligent sensing terminal, an edge intelligent analysis node, a safety communication module, a central control platform, a digital twin diagnosis center and a predictive maintenance engine. According to the invention, the intelligent sensing terminal integrates various sensors, adopts a permalloy shielding cavity to isolate external electromagnetic interference, is equipped with a three-stage lightning protection circuit to realize microsecond-stage lightning surge protection, can stably collect parameters of track control equipment in a severe environment, is internally provided with a lightweight LSTM model, carries out preliminary filtering analysis on original data, and is used for monitoring the parameters of the track control equipment. The device operation state can be dynamically identified, the sampling strategy can be adjusted, the basic sampling frequency is maintained in the stable state to reduce power consumption, the sampling frequency is improved by 10 times and the sensor gain is enhanced in the early warning or emergency state, all analysis decisions are completed at the local millisecond level without depending on the cloud, and the real-time performance and accuracy of data processing are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit intelligent monitoring, and in particular to an information acquisition and transmission system for rail control equipment. Background Art

[0002] The field of rail transit intelligent monitoring technology refers to a technical system that uses modern information technologies such as the Internet of Things, artificial intelligence, big data, and cloud computing to conduct real-time status perception, data collection, fault diagnosis, and early warning of key equipment such as vehicles, tracks, power supply, and signals in the rail transit system. Its core goal is to improve operation and maintenance efficiency and ensure operational safety through intelligent means, and to promote the transformation of the operation and maintenance model from "passive maintenance" to "active prevention."

[0003] Published patent: A signal acquisition and transmission system and method for section track equipment (publication number: CN117382699B). This system includes a voltage range conversion unit, a current range conversion unit, a voltage signal isolation unit, a current signal isolation unit, a voltage A / D acquisition unit, a current A / D acquisition unit, and a central processing unit. The voltage range conversion unit is connected to the voltage signal isolation unit, which is in turn connected to the voltage A / D acquisition unit. The current range conversion unit is connected to the current signal isolation unit, which is then connected to the current A / D acquisition unit. Both the voltage A / D acquisition unit and the current A / D acquisition unit are connected to the central processing unit. This system enables reliable acquisition of voltage and current data from the lightning protection simulation network disk in the ZPW-2000R non-insulated track circuit equipment, providing data support for the safe and stable operation of the equipment.

[0004] The above patent has defects: it is unable to capture transient anomalies in real time, only calculates the average value through the time series level method, relies on simple thresholds to compare grouped data, abnormal events may be missed, analysis delays cannot meet real-time response requirements, lacks the ability to classify abnormal events, cannot actively warn or locate faults, operation and maintenance rely on manual post-analysis, has a high risk of link interruption, data transmission is vulnerable to attacks, integrity and privacy cannot be guaranteed, troubleshooting is time-consuming, operation and maintenance costs are high, stability is poor in lightning, high humidity, and strong electromagnetic interference environments, and equipment is easily damaged. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an information collection and transmission system for track control equipment.

[0006] In order to achieve the above-mentioned objectives, the present invention adopts the following technical solutions: an information collection and transmission system for track control equipment, including intelligent sensing terminals, edge intelligent analysis nodes, secure communication modules, a central control platform, a digital twin diagnosis center and a predictive maintenance engine. All deployed intelligent sensing terminals, edge intelligent analysis nodes and central control platforms are connected to each other through secure communication links to form an overall link topology, and the center point of the link topology is the central control platform; the intelligent sensing terminals are deployed at the track control equipment points, and collect and locally pre-process multi-dimensional parameters through intelligent sensor components before reporting; the edge intelligent analysis nodes perform collaborative diagnosis on multiple terminal data and output alarms; the secure communication link adopts triple redundancy of main link, backup link and emergency link and combines national secret SM4 dynamic encryption and SM3 digital signature to achieve end-to-end protection; the central control platform is responsible for data aggregation, visual monitoring and remote operation and maintenance; the digital twin diagnosis center generates a three-dimensional visual model based on impedance-spectrum analysis and time domain reflection detection technology and outputs fault location; the predictive maintenance engine integrates multi-sensor health data and uses the Weibull distribution life prediction algorithm to generate component replacement recommendations.

[0007] As a further description of the above technical solution:

[0008] The intelligent sensor components of the intelligent sensing terminal include a voltage sensor, a current sensor, a contact resistance sensor, a temperature and humidity sensor, and an electromagnetic field intensity monitoring module. A Permalloy shielding cavity and a three-level lightning protection circuit are used to achieve environmental adaptive protection. The intelligent sensing terminal has an embedded lightweight LSTM model to perform preliminary analysis of real-time collected data and adjust the sampling frequency based on the operating status, increasing the sampling frequency in warning or emergency situations.

[0009] As a further description of the above technical solution:

[0010] The edge intelligent analysis node is equipped with an intelligent timing analysis engine that can locally identify 27 types of abnormal events and share model parameters through a distributed learning architecture without transmitting original data. Abnormal events include 22 types of electrical parameter anomalies and five types of environmental interference anomalies. The intelligent timing analysis engine outputs alarm instructions with event labels, locations and priorities.

[0011] As a further description of the above technical solution:

[0012] The secure communication link is synchronously constructed at the same time of message generation, with the time-sensitive optical fiber network as the main link, the anti-interference wireless frequency hopping communication as the backup link, and the power line carrier as the emergency link, and automatically switches and records the switching event when the link is changed. The secure communication link performs national secret SM4 dynamic encryption and SM3 digital signature on each frame of data at the sending end, and writes key alarm data into the blockchain evidence storage platform to form an unalterable record.

[0013] As a further description of the above technical solution:

[0014] The central control platform has four native functions: data centralized management platform, visual monitoring interface, remote maintenance interface and historical data backtracking analysis, and archives inter-station data for more than five years.

[0015] As a further description of the above technical solution:

[0016] The digital twin diagnosis center uses a digital twin engine to build a three-dimensional model that deforms synchronously with the site. After detecting line impedance anomalies, it triggers impedance-spectrum joint analysis and time-domain reflection detection technology to locate the cable fault location. The digital twin diagnosis center combines augmented reality technology to superimpose the fault point coordinates, line direction and burial depth information on the maintenance personnel's mobile terminal.

[0017] As a further description of the above technical solution:

[0018] The predictive maintenance engine analyzes vibration spectra, temperature rise curves, and current characteristics, uses a Weibull distribution life prediction algorithm to assess the remaining life of components, and automatically generates maintenance work orders and replacement priorities.

[0019] As a further description of the above technical solution:

[0020] The system's operating process is as follows:

[0021] S1. Intelligent sensing terminal data collection: The system first uses intelligent sensing terminals deployed on track control equipment to collect multi-dimensional data from the equipment in real time. Each intelligent sensing terminal is equipped with a voltage sensor, current sensor, contact resistance sensor, temperature and humidity sensor, and electromagnetic field intensity monitoring module to monitor the equipment's operating status and changes in the surrounding environment. The data collection process relies entirely on the terminal's embedded sensor components.

[0022] S2. Data Preprocessing and Local Decision-Making: The collected raw data undergoes preliminary filtering by the intelligent sensing terminal and is then fed into a lightweight LSTM model for analysis. The LSTM model generates data curves based on the device's historical detection data and makes a preliminary judgment on the device's operating status. The system then determines whether the device is in a normal, warning, or emergency state based on the device's status, and dynamically adjusts the data sampling frequency to ensure real-time and accuracy.

[0023] S3. Real-time monitoring and early warning processing: If the intelligent sensing terminal detects an anomaly or emergency, it immediately triggers a local alarm and increases the sampling frequency of collected data. All of this analysis and decision-making is completed locally on the intelligent sensing terminal, without relying on the cloud or edge computing nodes.

[0024] S4, Data transmission and encryption: After completing data processing, the intelligent sensing terminal transmits the processed pre-processed data, device status and alarm instructions to the edge intelligent analysis node through a secure communication link. The data is encrypted during the transmission process using the national secret SM4 dynamic encryption technology;

[0025] S5. Data analysis and anomaly detection at edge intelligent analysis nodes: After receiving data from multiple intelligent sensing terminals, edge intelligent analysis nodes perform data integration and collaborative diagnosis. Using a built-in LSTM model, edge nodes quickly scan data and identify complex events based on time series models. Using a distributed learning architecture, analysis model parameters are shared across nodes.

[0026] S6. Abnormal event processing and alarm push: Once an abnormal event is identified, the edge intelligent analysis node generates an alarm instruction based on the abnormality type, device status, and location, and pushes relevant information to the central control platform. These instructions include the time of the event, device status, and processing priority;

[0027] S7, Secure Communication and Data Backup: The secure communication module uses three redundant links during data transmission. If a link fails, another link is replaced. All transmitted data is encrypted and stored as evidence.

[0028] S8. Data reception and processing by the central control platform: The central control platform continuously receives encrypted messages from edge intelligent analysis nodes, decrypts them, and verifies their legitimacy. The verified data is stored in the database, and each record is labeled. Through the visual monitoring interface, operation and maintenance personnel can view the health status of the equipment, alarm logs, and trend charts.

[0029] S9. Maintenance Decision-Making and Execution: On the central control platform, operations and maintenance personnel gain a complete view of equipment health through real-time data monitoring and retrospective analysis of historical data. Through the remote maintenance interface, the platform can issue maintenance instructions to edge intelligent analysis nodes and push embedded software upgrades and parameter adjustments. If continuous anomalies or equipment failures are detected, the platform can remotely trigger a device reset to ensure normal system operation.

[0030] S10. Long-term data insights and optimization: All collected data and analysis results will be archived in the central database of the central control platform to form a complete data asset and stored for more than five years. Operations and maintenance personnel can conduct trend research, fault reconsideration and equipment performance optimization based on this long-term accumulated data.

[0031] The present invention has the following beneficial effects:

[0032] 1. In the present invention, the intelligent sensing terminal integrates multiple sensors, uses a Permalloy shielding cavity to isolate external electromagnetic interference, and is equipped with a three-level lightning protection circuit to achieve microsecond-level lightning surge protection. It can stably collect track control equipment parameters in harsh environments. The terminal has a built-in lightweight LSTM model to perform preliminary filtering analysis on the original data, dynamically identify the equipment operation status and adjust the sampling strategy. In the stable state, the basic sampling frequency is maintained to reduce power consumption. In the warning or emergency state, the sampling frequency is increased by 10 times and the sensor gain is enhanced. All analysis and decision-making are completed locally in milliseconds without relying on the cloud, which effectively improves the real-time and accuracy of data processing. The edge intelligent analysis node integrates multi-terminal data in the jurisdiction and accurately identifies 27 types of abnormal events within milliseconds through the built-in LSTM model. 22 categories are electrical parameter anomalies, and 5 categories are environmental interference anomalies. The system will automatically generate standardized event tags containing the anomaly type, equipment status, location, timestamp and processing priority, and simultaneously push alarm instructions to the central control platform. This multi-dimensional anomaly identification and rapid response mechanism has greatly improved the comprehensiveness and timeliness of fault diagnosis. The secure communication module has built a triple transmission mechanism of main link, backup link and emergency link. It realizes automatic switching by dynamically monitoring the link quality to ensure data transmission reliability. The data is encrypted segmented using the national secret SM4 dynamic encryption, and an additional SM3 digital signature is used to generate a hash fingerprint for tamper prevention. The key alarm data is synchronously written to the blockchain evidence storage platform. The module has a built-in clock synchronization unit to control the time error of the entire link to the millisecond level, providing precise time coordinates for event correlation.

[0033] 2. In the present invention, the central control platform receives decrypted data through a secure communication link. All collected data and analysis results are stored in a central database and tagged and indexed, forming a data asset for at least five years. The platform has a visual monitoring function, which can display the track status of the entire network, equipment health classification and alarm logs in real time, support multi-dimensional screening and pop-up alarms, and support remote parameter adjustment, embedded software upgrades, and equipment maintenance operations such as fault restart. It can also conduct data history backtracking by generating periodic reports and trend charts, and supports multi-condition search and data export. The digital twin diagnostic center builds a three-dimensional visual model based on the physical parameters of the equipment, dynamically maps the line impedance characteristics, and automatically initiates the diagnostic process when an impedance anomaly is detected: Abnormal sections are identified through impedance-spectrum analysis, and the fault point is estimated by combining time-domain reflectometry technology. The positioning results are then pushed to the on-site terminal in the form of AR virtual markings through augmented reality technology, guiding maintenance personnel to quickly reach the fault point and effectively improving fault repair efficiency. The predictive maintenance engine integrates multi-dimensional real-time data and historical records, and uses the Weibull distribution life prediction algorithm to calculate the component failure probability and remaining life based on shape parameters and scale parameters. Component replacement recommendations and maintenance work orders are generated. The system regularly pushes equipment degradation trend warnings to help operation and maintenance personnel intervene in advance and reduce the risk of unplanned downtime. All analysis results and suggestions are visualized through the central control platform, supporting long-term trend tracing and in-depth research. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a workflow diagram of the present invention;

[0035] Figure 2 This is a system architecture diagram of the present invention. DETAILED DESCRIPTION

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0037] Reference Figure 1-2The present invention provides an embodiment: an information collection and transmission system for track control equipment, including intelligent sensing terminals, edge intelligent analysis nodes, secure communication modules, a central control platform, a digital twin diagnosis center, and a predictive maintenance engine. All deployed intelligent sensing terminals, edge intelligent analysis nodes, and the central control platform are connected to each other through secure communication links to form an overall link topology, and the center of the link topology is the central control platform; the intelligent sensing terminals are deployed at track control equipment points, and collect and locally pre-process multi-dimensional parameters through intelligent sensor components before reporting; the edge intelligent analysis nodes perform collaborative diagnosis on multiple terminal data and output alarms; the secure communication link adopts triple redundancy of main link, backup link, and emergency link, and combines national secret SM4 dynamic encryption and SM3 digital signature to achieve end-to-end protection; the central control platform is responsible for data aggregation, visual monitoring, and remote operation and maintenance; the digital twin diagnosis center generates a three-dimensional visual model based on impedance-spectrum analysis and time-domain reflectometry detection technology and outputs fault location; the predictive maintenance engine integrates multi-sensor health data and uses the Weibull distribution life prediction algorithm to generate component replacement recommendations.

[0038] The intelligent sensor components of the intelligent sensing terminal include voltage sensors, current sensors, contact resistance sensors, temperature and humidity sensors, and electromagnetic field strength monitoring modules. Permalloy shielding cavity and three-level lightning protection circuit are used to achieve environmental adaptive protection. The intelligent sensing terminal has an embedded lightweight LSTM model to perform preliminary analysis on real-time collected data and adjust the sampling frequency according to the operating status. The sampling frequency is increased in warning or emergency situations. All deployed intelligent sensing terminals, edge intelligent analysis nodes, and each other are connected through secure communication links to form an overall link topology. The center of the link topology is the central control platform. The intelligent sensing terminal is equipped with an intelligent sensor component composed of multiple sensors. It is used to collect various parameters of track control equipment in real time. The intelligent sensing terminal only detects single-point equipment. The voltage sensor monitors the working voltage of the equipment in real time and records the voltage fluctuation range. The current sensor detects the current change during the operation of the equipment. The contact resistance sensor measures the resistance value of the equipment contact point. The temperature and humidity sensor collects the temperature and humidity data of the environment in which the equipment is located. The electromagnetic field strength monitoring module is used to monitor the electromagnetic field strength in the environment in real time and provide data support for anti-interference. The intelligent sensing terminal adopts a special design to cope with harsh environments. It uses a Permalloy shielding cavity to effectively isolate external electromagnetic interference. It is equipped with a three-level lightning protection circuit, which can achieve microsecond-level protection when lightning or surge occurs to ensure equipment stability. The intelligent sensing terminal has preliminary data analysis and decision-making capabilities, and is embedded with a lightweight LSTM model. The original data of the device status collected by the sensor component in real time is input into the LSTM model after preliminary filtering. The model generates a data curve based on the historical detection data of the device, and makes a preliminary analysis and judgment on the collected data to determine which operating state the device is in. The operating state includes a stable state with normal fluctuations, a warning state where the parameters deviate from the baseline, and an emergency state where the mutation exceeds the threshold. According to the preliminary judgment results, the sampling strategy is dynamically adjusted. If the result is a stable state, the terminal maintains the basic sampling frequency to reduce power consumption. If a warning or emergency state is detected, a local alarm instruction is immediately generated to take The sampling frequency is increased to 10 times the basic value, and the sensor gain is enhanced synchronously to ensure the capture of transient abnormal details and the real-time and accuracy of the data. All predictions and adjustments are completed locally on the terminal without relying on the cloud or edge nodes. The delay from data input to decision output is controlled at the millisecond level to meet real-time requirements. At the same time, the LSTM model will also make short-term predictions based on the changes in historical data to obtain the state of the device in the next few seconds to minutes and the probability of being in that state. The collected data and preliminary analysis results are stored locally for a short period of time for subsequent analysis. The cache time is 4 hours, and the stored data is encrypted using the national secret SM4 dynamic encryption technology to ensure data security.

[0039] Example 1: Multiple intelligent sensing terminals are deployed along the tracks between one railway station and another. Each terminal is equipped with a voltage sensor, a current sensor, a contact resistance sensor, a temperature and humidity sensor, and an electromagnetic field intensity monitoring module. These sensors monitor various parameters of track control equipment in real time. The intelligent sensing terminals use their embedded lightweight LSTM models to perform preliminary analysis of the collected raw data. If the equipment's operating status is stable, the intelligent sensing terminal maintains a basic sampling frequency and performs short-term predictions based on historical data patterns to ensure a healthy assessment of the equipment's operating status. However, if the intelligent sensing terminal detects that the equipment's status deviates from the normal range and enters a warning or emergency state, the system automatically adjusts the sampling frequency and increases the sensor gain to ensure that the abnormal details are captured. At this time, the terminal generates a local alarm command and transmits the data to the edge intelligent analysis node via a secure communication link. During transmission, all data is protected using the national secret SM4 encryption technology. Data is transmitted through the main link, backup link, and emergency link to ensure data security and integrity. In this way, the intelligent sensing terminal can efficiently and in real time monitor equipment status and respond to potential failures.

[0040] The edge intelligent analysis node is equipped with an intelligent timing analysis engine, which can identify 27 types of abnormal events locally and share model parameters through a distributed learning architecture without transmitting original data. Abnormal events include 22 types of electrical parameter abnormalities and five types of environmental interference abnormalities. The intelligent timing analysis engine outputs alarm instructions with event labels, locations and priorities. The edge intelligent analysis node is deployed at the station and is equipped with an intelligent timing analysis engine for receiving pre-processed data collected in real time from all associated intelligent sensing terminals within the jurisdiction of the station, as well as the real-time operating status of the equipment and local alarm instructions. After the edge intelligent analysis node integrates the data from multiple terminals, it conducts collaborative diagnosis on the data from multiple terminals and uses the built-in LSTM model to analyze the data. It conducts rapid scanning and accurately identifies 27 types of abnormal events based on the identification of complex events based on the time series model. There are 22 types of abnormal electrical parameter events, namely voltage sag: instantaneous voltage is lower than the threshold, voltage swell: instantaneous voltage is higher than the threshold, continuous voltage fluctuation: periodic voltage fluctuation exceeds the preset frequency, current overload: continuous current exceeds the safety threshold, current harmonic distortion: harmonic content in a specific frequency band exceeds the standard, three-phase current imbalance: phase difference exceeds the allowable range, contact resistance surge: instantaneous increase in connection point resistance, grounding resistance failure: abnormal fluctuation in ground loop resistance, insulation resistance degradation: resistance value continues to decline, DC component abnormality: DC offset in AC system, leakage current exceeds the standard: equipment leakage current to ground exceeds the limit, power factor sag: reactive power instantaneous Abnormal, instantaneous power overshoot: power spike exceeds the tolerance of the equipment, frequency drift: power supply frequency deviates from the nominal value, voltage sag propagation: voltage drop chain across equipment, current inrush: equipment startup current continues to timeout, zero-sequence current abnormality: three-phase imbalance derivative fault, interharmonic interference: non-integer multiple fundamental frequency harmonics, voltage flicker: rapid voltage fluctuation visible in lighting, resonant overvoltage: LC circuit resonance causes voltage amplification, phase loss operation: single-phase power supply interruption, arc discharge: intermittent discharge current at the contact point, there are 5 types of environmental interference abnormal events, namely lightning surge conduction: microsecond current / voltage pulse, electromagnetic interference coupling: external electromagnetic field causes signal distortion, temperature sudden change alarm: local temperature rise rate exceeds the threshold, humidity saturation risk: the ambient humidity is continuously high Regarding critical values ​​and salt spray corrosion effects: high salinity environments lead to abnormal resistance. By quickly identifying these abnormal events, the system can respond within milliseconds. The analysis process does not need to rely on cloud computing power, and abnormality detection can be completed directly locally. During the data analysis process, the edge intelligent analysis node further improves the accuracy of the analysis through the self-learning optimization mechanism. The distributed learning architecture enables the nodes to share the analysis model parameters without transmitting the original data. This not only ensures the security of data privacy, but also improves the diagnostic accuracy of the entire system. When the node identifies an abnormal event, it classifies the detected anomaly and generates a standardized event label. The label includes the abnormal event, device status, location, timestamp and processing priority.Automatically generate alarm instructions and push the pre-processed data, event type, occurrence time and recommended processing priority collected by each intelligent sensing terminal to the central control platform.

[0041] Example 2: Data from multiple intelligent sensing terminals is transmitted to an edge intelligent analysis node via a secure communication link. The edge intelligent analysis node has a built-in intelligent time series analysis engine that receives and integrates data from multiple terminals. During data processing, the edge intelligent analysis node quickly scans the data using an LSTM model and identifies 27 possible abnormal events, including electrical parameter anomalies and environmental interference anomalies. For example, the node can detect faults such as voltage sag, current overload, and grounding resistance failure, as well as environmental factors such as sudden temperature changes and electromagnetic interference. After identifying abnormal events, the edge intelligent analysis node classifies these events through time series analysis and distributed learning optimization mechanisms, and generates standardized event labels. The labels include the anomaly type, device status, occurrence time, location, and processing priority. Subsequently, the abnormal alarm information will be pushed to the central control platform for timely processing by operation and maintenance personnel. This method ensures that the edge node can complete anomaly detection and response within milliseconds, which not only improves the timeliness of equipment health monitoring, but also ensures the protection of data privacy and improves analysis accuracy.

[0042] The secure communication link is synchronously constructed at the same time of message generation, with the time-sensitive optical fiber network as the main link, the anti-interference wireless frequency hopping communication as the backup link, and the power line carrier as the emergency link, and automatically switches when the link is changed and records the switching event. The secure communication link performs national secret SM4 dynamic encryption and SM3 digital signature on each frame of data at the sending end, and writes the key alarm data into the blockchain evidence platform to form an unalterable record. The secure communication module is used for data transmission of the entire system. The intelligent sensing terminal transmits data to the edge intelligent analysis node, the edge intelligent analysis node transmits data to the intelligent sensing terminal, and the central control platform transmits data to the edge intelligent analysis node. When the edge intelligent analysis node transmits data and the edge intelligent analysis node transmits data to the central control platform, in order to resist interference and link failure, three transmission links are synchronously constructed at the same time of message generation, namely the main link, the backup link and the emergency link. The main link adopts the "time-sensitive optical fiber network" to provide low latency and high bandwidth. The backup link uses "anti-interference wireless frequency hopping communication" to automatically take over when the optical fiber is blocked. The emergency link uses the "power line carrier" to achieve physical isolation transmission, ensuring the minimum data accessibility in extreme scenarios. The module dynamically monitors the link quality indicators. When the main link shows attenuation or packet loss trend, it immediately switches to the backup link. The secure communication module uses the "data encryption and evidence storage mechanism" to perform double-layer protection on each frame of data from the first hop after the message leaves the sender. It uses the "national secret SM4 dynamic encryption" to perform segmented encryption on the data payload and uses the "SM3 digital signature" to generate a hash fingerprint for the message header and payload, which is attached to the end of the data. No node can tamper with the data without destroying the signature. The key alarm data is synchronously written to the "blockchain evidence storage platform" to form an unalterable chain record. The encryption, signature and evidence storage operations are all completed at the data sending end, without the need for external computing resources, thus avoiding the risk of encryption. To avoid the risks of key leakage and link blocking, when the message arrives at the other end, it is decrypted using the local key and compared with the SM3 signature value. If the signatures are inconsistent or the decryption fails, the data is determined to have been tampered with, the message is immediately discarded, and a link abnormality event is generated to provide a basis for subsequent tracing. To ensure cross-node data alignment, the secure communication module has a built-in clock synchronization unit, which periodically broadcasts the reference timestamp to each intelligent perception terminal and edge intelligent analysis node, and monitors the return signal to calculate the round-trip delay. When it is detected that the sub-link clock drift exceeds the allowable range, a correction instruction is immediately issued to converge the time error of the entire link to the millisecond level, providing a trusted time coordinate for subsequent event association.

[0043] The central control platform has four native functions: data centralized management platform, visual monitoring interface, remote maintenance interface and historical data retrospective analysis. It archives inter-station data for more than five years. The central control platform continuously receives encrypted messages from edge intelligent analysis nodes through a secure communication link, and completes decryption and legitimacy verification locally. The verified data is streamed into the central control platform database. The system automatically adds index tags to each record, including site, equipment, timestamp, and abnormal events, to ensure the accurate positioning of subsequent queries. The corresponding key alarm records are synchronously written into the long-term archiving area to form a complete data asset. And store it for more than five years to support long-term trend research. A display screen is set up. The platform will immediately map the newly arrived data to the display screen after decoding, forming a visual monitoring interface. Operation and maintenance personnel can view the inter-station track status, equipment health classification and alarm log on a single screen. When the edge intelligent analysis node reports "abnormal event identification and alarm" information, the monitoring interface will automatically pop up and highlight the corresponding site, give the event label, severity level and timestamp, so that personnel can locate it in the first time. The monitoring interface supports multi-dimensional screening. Operation and maintenance personnel can switch views according to lines, sites or equipment types, and intuitively grasp the operation status of the entire network. Through the remote maintenance interface, the platform The platform can send parameter adjustment instructions to the edge intelligent analysis node, and the interface supports the embedded software upgrade function. After the operation and maintenance personnel upload the new version of the program on the central control platform, the system pushes it to the target device according to the whitelist order to ensure the safety and controllability of the upgrade process. When continuous abnormalities or link loss of synchronization are detected, the operation and maintenance personnel can remotely trigger "fault restart". The platform automatically completes the device reset and records the operation log to meet the audit requirements. The platform has a built-in trend chart tool that can generate hourly, daily or weekly curves for any device parameters to help operation and maintenance personnel discover hidden degradation trends. Periodic reports are set to periodically summarize key indicators to provide operation and maintenance personnel with It provides an overview of operational health and fault distribution, supports multi-condition retrieval based on time or event tags, and quickly replays all records before and after the alarm occurs, providing a chain of evidence for fault reconsideration. All analysis results can be exported through the interface for subsequent in-depth research or third-party system calls. Through the four native functions of "data centralized management platform, visual monitoring interface, remote maintenance interface, and historical data backtracking analysis", the central control platform realizes the unified aggregation, real-time display, online operation and maintenance, and long-term data insights of inter-station equipment status data, thereby providing reliable central scheduling and decision-making support for the information collection and transmission system of the entire track control equipment.

[0044] Example 3: The central control platform is responsible for receiving encrypted messages from edge intelligent analysis nodes, decrypting and verifying their legitimacy. The verified data is written to the central control platform database, and the device status, health level and fault alarm are displayed in real time through the visual monitoring interface. When the edge intelligent analysis node reports an abnormal event, the central control platform will automatically pop up a window on the interface and highlight the location of the faulty device, the severity of the event and the timestamp to help the operation and maintenance personnel respond in the first time. When the operation and maintenance personnel confirm the fault, the central control platform can send parameter adjustment instructions to the edge intelligent analysis node through the remote maintenance interface, or perform equipment maintenance through the system embedded software upgrade function. When persistent abnormalities or link loss of synchronization are detected, the platform can remotely trigger the device reset function, automatically restore the normal operation of the device and record the operation log. All operation processes will form detailed fault records and analysis reports for subsequent data backtracking and fault review.

[0045] The digital twin diagnosis center uses the digital twin engine to build a three-dimensional model that deforms synchronously with the site, and triggers impedance-spectrum joint analysis and time-domain reflection detection technology after detecting line impedance anomalies to locate the cable fault location. The digital twin diagnosis center combines augmented reality technology to superimpose the fault point coordinates, line direction and burial depth information on the maintenance personnel's mobile terminal. The digital twin diagnosis center receives pre-processed data collected by the intelligent sensing terminal, and by calling the digital twin engine, it builds a three-dimensional visualization model for each track control device based on the physical parameters of the equipment, accurately restoring the spatial structure and electrical connection relationship of the track control equipment. The model maps continuous data streams into dynamic geometric bodies. The model will deform synchronously with changes in equipment parameters, thereby dynamically mapping the line impedance characteristics of the real equipment, so that the virtual scene is always consistent with the on-site status, providing a virtual reference benchmark for fault location. The model resolution is sufficient to present details such as terminal blocks and grounding loops, facilitating subsequent fine diagnosis. When abnormal fluctuations in line impedance are detected, the diagnostic process is automatically activated, combined with impedance-spectrum analysis technology, to scan the impedance frequency response characteristics, identify impedance mismatch points, and highlight them in the model. The corresponding section is identified and the amplitude, frequency, and duration of the anomaly are recorded. For the marked anomaly section, time-domain reflectometry technology is applied. An equivalent excitation pulse is injected into the model and the reflected waveform is calculated. The distance from the fault point to the reference end is inferred based on the reflection time difference and propagation velocity. This distance is marked on the model with a digital ruler, resulting in a cable fault location with an accuracy of ±15 meters. Once located, the center uses augmented reality technology to push the fault coordinates, line direction, and burial depth information to the on-site maintenance personnel's mobile device in real time. An AR virtual marking is superimposed on the mobile device screen to guide maintenance personnel along the optimal path to the fault point. Maintenance personnel use their mobile device or AR glasses to point to on-site facilities and see three-dimensional navigation arrows and fault markers aligned with the real environment, reducing manual search time. The fault type is determined based on impedance characteristics, spectral morphology, and historical maintenance records, and diagnostic recommendations are automatically generated. The report includes the fault nature, precise distance, possible impact range, recommended replacement parts, and repair priority ranking. All information is pushed to the maintenance personnel's mobile device in structured text format and archived in the central database for subsequent tracing.

[0046] The predictive maintenance engine analyzes the vibration spectrum, temperature rise curve and current characteristics, uses the Weibull distribution life prediction algorithm to evaluate the remaining life of the components, and automatically generates maintenance work orders and replacement priorities. The predictive maintenance engine first collects data from different sensors from the intelligent sensing terminal. These data streams are integrated and transmitted to the predictive maintenance engine to provide necessary input for subsequent analysis. The collected multi-dimensional data is combined with the historical data of the equipment to establish an equipment health assessment model. This model integrates the characteristics of multiple sensors, and uses this information to evaluate the current health status of the equipment and identify potential hidden faults. Based on the collaborative analysis of temperature rise curves and current characteristics, the predictive maintenance engine can accurately identify hidden faults in the equipment. These faults can be It can cause equipment failure in the future. The predictive maintenance engine uses the Weibull distribution life prediction algorithm. The Weibull distribution life model predicts the future failure probability and remaining life of the equipment by analyzing the historical data and current operating status of the equipment. Based on the actual historical life data, the least squares method estimates the shape parameter and scale parameter of the Weibull distribution by fitting the log-log graph of the data. The scale parameter (η) determines the scale of the life and represents the typical life of the equipment or component. The shape parameter (β) determines the shape of the data distribution and affects the failure trend. The Weibull distribution formula is used to predict the remaining life of equipment components. The cumulative distribution function of the Weibull distribution can calculate the probability of equipment failure at a specific time point, thereby providing a life prediction for the equipment. The expected life is calculated by the formula: , the calculation formula is: , the remaining life calculation formula is: , t is the life of the equipment or component, β is the shape parameter, η is the scale parameter, Γ is the gamma function, and the expected life can help predict whether the equipment has the risk of failure in the future. F(t) represents the probability of the equipment failing before time t. The component failure probability distribution curve is constructed based on historical failure data, and the life of each component of the equipment is evaluated in combination with the real-time health score. Based on the evaluation results, the system regularly generates component replacement suggestions to ensure that necessary preventive maintenance is carried out before the failure occurs and reduce the risk of sudden failure. When a potential failure is predicted, the predictive maintenance engine will automatically generate a maintenance work order, recommend which components to replace, and arrange the maintenance priority. These suggestions will be pushed to the mobile terminal of the operation and maintenance personnel to help them deal with it in a timely manner according to the severity and impact range of the fault. Based on the real-time health score and the component failure probability distribution curve, equipment degradation trend warnings are regularly pushed to reduce unplanned downtime.

[0047] The system's operating process is as follows:

[0048] S1. Intelligent sensing terminal data collection: The system first uses intelligent sensing terminals deployed on track control equipment to collect multi-dimensional data from the equipment in real time. Each intelligent sensing terminal is equipped with a voltage sensor, current sensor, contact resistance sensor, temperature and humidity sensor, and electromagnetic field intensity monitoring module to monitor the equipment's operating status and changes in the surrounding environment. The data collection process relies entirely on the terminal's embedded sensor components.

[0049] S2. Data Preprocessing and Local Decision-Making: The collected raw data undergoes preliminary filtering by the intelligent sensing terminal and is then fed into a lightweight LSTM model for analysis. The LSTM model generates data curves based on the device's historical detection data and makes a preliminary judgment on the device's operating status. The system then determines whether the device is in a normal, warning, or emergency state based on the device's status, and dynamically adjusts the data sampling frequency to ensure real-time and accuracy.

[0050] S3. Real-time monitoring and early warning processing: If the intelligent sensing terminal detects an anomaly or emergency, it immediately triggers a local alarm and increases the sampling frequency of collected data. All of this analysis and decision-making is completed locally on the intelligent sensing terminal, without relying on the cloud or edge computing nodes.

[0051] S4, Data transmission and encryption: After completing data processing, the intelligent sensing terminal transmits the processed pre-processed data, device status and alarm instructions to the edge intelligent analysis node through a secure communication link. The data is encrypted during the transmission process using the national secret SM4 dynamic encryption technology;

[0052] S5. Data analysis and anomaly detection at edge intelligent analysis nodes: After receiving data from multiple intelligent sensing terminals, edge intelligent analysis nodes perform data integration and collaborative diagnosis. Using a built-in LSTM model, edge nodes quickly scan data and identify complex events based on time series models. Using a distributed learning architecture, analysis model parameters are shared across nodes.

[0053] S6. Abnormal event processing and alarm push: Once an abnormal event is identified, the edge intelligent analysis node generates an alarm instruction based on the abnormality type, device status, and location, and pushes relevant information to the central control platform. These instructions include the time of the event, device status, and processing priority;

[0054] S7, Secure Communication and Data Backup: The secure communication module uses three redundant links during data transmission. If a link fails, another link is replaced. All transmitted data is encrypted and stored as evidence.

[0055] S8. Data reception and processing by the central control platform: The central control platform continuously receives encrypted messages from edge intelligent analysis nodes, decrypts them, and verifies their legitimacy. The verified data is stored in the database, and each record is labeled. Through the visual monitoring interface, operation and maintenance personnel can view the health status of the equipment, alarm logs, and trend charts.

[0056] S9. Maintenance Decision-Making and Execution: On the central control platform, operations and maintenance personnel gain a complete view of equipment health through real-time data monitoring and retrospective analysis of historical data. Through the remote maintenance interface, the platform can issue maintenance instructions to edge intelligent analysis nodes and push embedded software upgrades and parameter adjustments. If continuous anomalies or equipment failures are detected, the platform can remotely trigger a device reset to ensure normal system operation.

[0057] S10. Long-term data insights and optimization: All collected data and analysis results will be archived in the central database of the central control platform to form a complete data asset and stored for more than five years. Operations and maintenance personnel can conduct trend research, fault reconsideration and equipment performance optimization based on this long-term accumulated data.

[0058] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An information acquisition and transmission system for track control equipment, characterized by: It includes intelligent sensing terminals, edge intelligent analysis nodes, secure communication modules, a central control platform, a digital twin diagnosis center, and a predictive maintenance engine. All deployed intelligent sensing terminals, edge intelligent analysis nodes, and the central control platform are connected to each other through secure communication links to form an overall link topology, with the central control platform as the center of the link topology. Intelligent sensing terminals are deployed at track control equipment points, and collect and pre-process multi-dimensional parameters locally through intelligent sensor components before reporting. Edge intelligent analysis nodes conduct collaborative diagnosis of data from multiple terminals and output alarms. The secure communication link uses triple redundancy of primary, backup, and emergency links, and combines national SM4 dynamic encryption and SM3 digital signatures to achieve end-to-end protection. The central control platform is responsible for data aggregation, visual monitoring, and remote operation and maintenance. The digital twin diagnostic center generates a three-dimensional visual model and outputs fault location based on impedance-spectrum analysis and time-domain reflectometry detection technology; the predictive maintenance engine integrates multi-sensor health data and uses the Weibull distribution life prediction algorithm to generate component replacement recommendations.

2. The information collection and transmission system for track control equipment according to claim 1, characterized in that: The intelligent sensor components of the intelligent sensing terminal include a voltage sensor, a current sensor, a contact resistance sensor, a temperature and humidity sensor, and an electromagnetic field intensity monitoring module. A Permalloy shielding cavity and a three-level lightning protection circuit are used to achieve environmental adaptive protection. The intelligent sensing terminal has an embedded lightweight LSTM model to perform preliminary analysis of real-time collected data and adjust the sampling frequency based on the operating status, increasing the sampling frequency in warning or emergency situations.

3. The information collection and transmission system for track control equipment according to claim 1, characterized in that: The edge intelligent analysis node is equipped with an intelligent timing analysis engine that can locally identify 27 types of abnormal events and share model parameters through a distributed learning architecture without transmitting original data. Abnormal events include 22 types of electrical parameter anomalies and five types of environmental interference anomalies. The intelligent timing analysis engine outputs alarm instructions with event labels, locations and priorities.

4. The information collection and transmission system for track control equipment according to claim 1, characterized in that: The secure communication link is synchronously constructed at the same time of message generation, with the time-sensitive optical fiber network as the main link, the anti-interference wireless frequency hopping communication as the backup link, and the power line carrier as the emergency link, and automatically switches and records the switching event when the link is changed. The secure communication link performs national secret SM4 dynamic encryption and SM3 digital signature on each frame of data at the sending end, and writes key alarm data into the blockchain evidence storage platform to form an unalterable record.

5. The information collection and transmission system for track control equipment according to claim 1, characterized in that: The central control platform has four native functions: data centralized management platform, visual monitoring interface, remote maintenance interface and historical data backtracking analysis, and archives inter-station data for more than five years.

6. The information collection and transmission system for track control equipment according to claim 1, characterized in that: The digital twin diagnosis center uses a digital twin engine to build a three-dimensional model that deforms synchronously with the site. After detecting line impedance anomalies, it triggers impedance-spectrum joint analysis and time-domain reflection detection technology to locate the cable fault location. The digital twin diagnosis center combines augmented reality technology to superimpose the fault point coordinates, line direction and burial depth information on the maintenance personnel's mobile terminal.

7. The information collection and transmission system for track control equipment according to claim 1, characterized in that: The predictive maintenance engine analyzes vibration spectra, temperature rise curves, and current characteristics, uses a Weibull distribution life prediction algorithm to assess the remaining life of components, and automatically generates maintenance work orders and replacement priorities.

8. The information collection and transmission system for track control equipment according to claim 1, characterized in that: The system's operating process is as follows: S1. Intelligent sensing terminal data collection: The system first uses intelligent sensing terminals deployed on track control equipment to collect multi-dimensional data from the equipment in real time. Each intelligent sensing terminal is equipped with a voltage sensor, current sensor, contact resistance sensor, temperature and humidity sensor, and electromagnetic field intensity monitoring module to monitor the equipment's operating status and changes in the surrounding environment. The data collection process relies entirely on the terminal's embedded sensor components. S2. Data Preprocessing and Local Decision-Making: The collected raw data undergoes preliminary filtering by the intelligent sensing terminal and is then fed into a lightweight LSTM model for analysis. The LSTM model generates data curves based on the device's historical detection data and makes a preliminary judgment on the device's operating status. The system then determines whether the device is in a normal, warning, or emergency state based on the device's status, and dynamically adjusts the data sampling frequency to ensure real-time and accuracy. S3. Real-time monitoring and early warning processing: If the intelligent sensing terminal detects an anomaly or emergency, it immediately triggers a local alarm and increases the sampling frequency of collected data. All of this analysis and decision-making is completed locally on the intelligent sensing terminal, without relying on the cloud or edge computing nodes. S4, Data transmission and encryption: After completing data processing, the intelligent sensing terminal transmits the processed pre-processed data, device status and alarm instructions to the edge intelligent analysis node through a secure communication link. The data is encrypted during the transmission process using the national secret SM4 dynamic encryption technology; S5. Data analysis and anomaly detection at edge intelligent analysis nodes: After receiving data from multiple intelligent sensing terminals, edge intelligent analysis nodes perform data integration and collaborative diagnosis. Using a built-in LSTM model, edge nodes quickly scan data and identify complex events based on time series models. Using a distributed learning architecture, analysis model parameters are shared across nodes. S6. Abnormal event processing and alarm push: Once an abnormal event is identified, the edge intelligent analysis node generates an alarm instruction based on the abnormality type, device status, and location, and pushes relevant information to the central control platform. These instructions include the time of the event, device status, and processing priority; S7, Secure Communication and Data Backup: The secure communication module uses three redundant links during data transmission. If a link fails, another link is replaced. All transmitted data is encrypted and stored as evidence. S8. Data reception and processing by the central control platform: The central control platform continuously receives encrypted messages from edge intelligent analysis nodes, decrypts them, and verifies their legitimacy. The verified data is stored in the database, and each record is labeled. Through the visual monitoring interface, operation and maintenance personnel can view the health status of the equipment, alarm logs, and trend charts. S9. Maintenance Decision-Making and Execution: On the central control platform, operations and maintenance personnel gain a complete view of equipment health through real-time data monitoring and retrospective analysis of historical data. Through the remote maintenance interface, the platform can issue maintenance instructions to edge intelligent analysis nodes and push embedded software upgrades and parameter adjustments. If continuous anomalies or equipment failures are detected, the platform can remotely trigger a device reset to ensure normal system operation. S10. Long-term data insights and optimization: All collected data and analysis results will be archived in the central database of the central control platform to form a complete data asset and stored for more than five years. Operations and maintenance personnel can conduct trend research, fault reconsideration and equipment performance optimization based on this long-term accumulated data.

Images