Railway 10kV self-closing / through line early warning and fault rapid processing system

Abstract

The invention relates to a railway 10kV self-closing / through line early warning and fault rapid handling system, which comprises a sensing layer used for collecting state data and meteorological parameters of a railway 10kV self-closing / through line and executing instantaneous fault rapid removal and switching-on and switching-off; the network layer is used for converging the data acquired by the sensing layer to the platform layer in real time; and the platform layer is used for comprehensively studying, judging and confirming the fault position, immediately sending a remote control instruction to the sensing layer of the fault section to physically isolate the fault section, and collecting data through the sensing layer to carry out electrical parameter early warning and environmental state early warning. Aiming at the operation and maintenance requirements of the 10kV self-closing / through line of the railway, the intelligent, precise and efficient operation and maintenance are realized through a three-layer distributed architecture and multi-module intelligent collaborative design, the system has accuracy, timeliness, intelligence and economical efficiency, a perceptible, predictable and controllable intelligent solution is provided for a railway power supply system, and the operation and maintenance requirements of the 10kV self-closing / through line of the railway are met. And safe and stable operation of railway transportation is effectively supported.

Description

Technical Field

[0001] This invention relates to the field of railway power system operation and maintenance technology, specifically to a 10kV automatic shut-off / through-line early warning and rapid fault handling system for railways. Background Technology

[0002] Currently, the operation and maintenance of the power supply system for 10kV automatic / through railway lines under various operating conditions mainly relies on a combination of traditional relay protection devices, traveling wave fault location equipment, and regular manual inspections. While this model ensures basic operational safety to a certain extent, it still has significant structural defects that hinder the intelligent development of modern railway power supply systems.

[0003] 1. Lines without traveling wave positioning devices have weak emergency response capabilities. On such lines, once a fault occurs, the system can only rely on the "three remote" functions (telemetry, telesignaling, and remote control) of the SCADA system (Supervisory Control and Data Acquisition System) for preliminary judgment, and cannot achieve automatic identification and isolation of the faulty section. In actual handling, it is often necessary to take measures such as shutting down the entire section or manually restoring power to each section step by step to find the fault point, resulting in long-term power outages in non-faulty areas, with an average interruption time of more than 2 hours.

[0004] 2. Lines with deployed traveling wave devices suffer from intelligent closed-loop failure issues. Although some lines are equipped with traveling wave fault location systems that can provide relatively accurate fault range information within seconds, subsequent fault confirmation, switching operations, isolation execution, and power restoration still require manual on-site intervention. Each subsystem, including relay protection, traveling wave equipment, and SCADA systems, operates independently, lacking a unified data fusion platform and intelligent decision-making center, making it difficult to form a fully automated closed-loop control chain of "early warning-location-isolation-recovery."

[0005] Therefore, it is necessary to propose new measures to overcome the above-mentioned shortcomings. Summary of the Invention

[0006] The purpose of this invention is to provide a 10kV automatic / through-line early warning and rapid fault handling system for railways, in order to solve the problems of delayed response, low handling efficiency and strong reliance on manpower in the existing operation and maintenance mode.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] A system for early warning and rapid fault handling of a 10kV automatic / through railway line is provided. The system is a three-layer distributed architecture, comprising:

[0009] The sensing layer is used to collect status data and meteorological parameters of the 10kV automatic shut-off / through lines of the railway, and to perform rapid isolation and switching on / off of transient faults.

[0010] The network layer is used to aggregate data acquired by the perception layer to the platform layer in real time;

[0011] The platform layer is used to comprehensively assess and confirm the location of the fault, and then sends remote control commands to the sensing layer of the faulty section to physically isolate the faulty section. The sensing layer collects data to provide early warnings of electrical parameters and environmental conditions.

[0012] Furthermore, the perception layer includes:

[0013] An integrated intelligent device is used to monitor voltage quality, load current, and traveling wave signals in real time, perform power line switching, and collect and transmit fault data.

[0014] Micro-meteorological monitoring devices are used to collect meteorological parameters;

[0015] Tower tilt monitoring sensors are used to continuously assess the stability of tower foundations;

[0016] Image acquisition device is used to automatically identify potential hazards such as overheating of joints and hanging foreign objects.

[0017] Furthermore, the integrated smart device includes:

[0018] Current transformers are used to monitor load current;

[0019] Voltage sensors are used to monitor voltage quality;

[0020] Traveling wave acquisition module, used to monitor traveling wave signals;

[0021] Voltage vacuum circuit breakers are used to perform rapid disconnection and opening / closing of transient faults;

[0022] When a sudden increase in current, a sudden drop in voltage, or a traveling wave signal is detected, the local traveling wave location algorithm is immediately activated. After initial judgment, the results are packaged and uploaded to the platform layer. The platform layer comprehensively analyzes and confirms the fault location, and remotely controls the voltage vacuum circuit breaker to physically isolate the faulty section.

[0023] Furthermore, integrated smart devices also include:

[0024] The communication management module is used for communication connections between different devices;

[0025] The clock module is used to acquire BeiDou clock signals, providing clock signals for the integrated smart device and synchronizing the clocks of all devices.

[0026] Furthermore, the platform layer includes intelligent analysis tools and a knowledge graph engine, which integrate multi-source heterogeneous data from SCADA systems, meteorological information systems, and equipment management systems, and uses LSTM neural networks and random forests for analysis and computation.

[0027] Furthermore, the platform layer comprehensively analyzes and confirms the location of the fault, including:

[0028] The system detects the original voltage / current waveform and electrical information at the time of the fault, performs energy surge detection on the original waveform, and identifies voltage / current abrupt change points in the waveform, with the abrupt change points corresponding to the time of the fault occurrence.

[0029] The detected mutation points were marked as Wavelet transform is used to decompose waveforms containing abrupt changes, and further extract detailed features of the waveforms.

[0030] The decomposed waveform still contains information about abrupt changes. The times when the wavefront arrives at different detection points are captured and denoted as t1 and t2, respectively. The time difference between the two points is then used to... And combined with the dynamically corrected wave velocity v, through the formula Calculate the distance L from the fault point to the detection point;

[0031] Based on the calculated fault distance L, the kilometer marker of the fault point is determined.

[0032] Furthermore, the platform layer collects data from the perception layer to provide early warnings of electrical parameters, including:

[0033] Historical voltage quality data, load current data, and meteorological data are acquired, and the electrical load change trend is predicted based on the ARIMA time series algorithm.

[0034] Real-time monitoring of voltage deviation, establishment of a voltage over-limit risk prediction model based on historical data, and early identification of power quality anomalies through FFT spectrum analysis technology.

[0035] Furthermore, the platform layer collects data from the perception layer to provide early warnings of environmental conditions, including:

[0036] Real-time meteorological parameters are collected by micro-meteorological monitoring devices and combined with weather forecasts to issue severe weather warnings 48 hours in advance.

[0037] Tower tilt data is collected by tower tilt monitoring sensors, and the foundation stability risk is identified based on the trend of tilt data changes.

[0038] Furthermore, the platform layer constructs a power neural network model, with node features including state data, and uses an LSTM neural network to calculate and assess system vulnerability in real time, identifying critical nodes and weak links.

[0039] Furthermore, the platform layer employs the random forest algorithm to extract time information, state data, and meteorological parameters as key features, and outputs the probability distribution of fault risk levels.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] This invention provides a 10kV automatic shut-off / through-line early warning and rapid fault handling system for railways. Addressing the operation and maintenance needs of 10kV automatic shut-off / through-line railways, this system achieves intelligent, precise, and efficient operation and maintenance through a three-layer distributed architecture and multi-module intelligent collaborative design. It possesses core advantages such as accuracy (data accuracy > 98%), timeliness (fault warning 48 hours in advance, location isolation < 1 minute, power restoration within minutes), intelligence (multi-algorithm collaboration and knowledge graph support for full-process intelligence), and economy (fault incidence rate reduced by 40%, operation and maintenance costs reduced by 25%). It provides a perceptible, predictable, and controllable intelligent solution for railway power supply systems, effectively supporting the safe and stable operation of railway transportation. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.

[0043] Figure 1 This is a three-layer distributed architecture diagram of the railway 10kV automatic / through line early warning and rapid fault handling system provided in the embodiment of the present invention.

[0044] Figure 2 This is a functional module composition diagram of the integrated intelligent device provided in the embodiments of the present invention.

[0045] Figure 3 This is a schematic diagram illustrating the principle of wavelet transform wavefront identification in traveling wave positioning.

[0046] Figure 4 This is a flowchart of fault location and automatic isolation control provided in an embodiment of the present invention.

[0047] Figure 5 This is a flowchart of the intelligent early warning process provided in an embodiment of the present invention.

[0048] Figure 6 This is a flowchart of fault handling and closed-loop management provided in the embodiments of the present invention. Detailed Implementation

[0049] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0050] It should be noted that similar reference numerals and letters indicate similar items; therefore, once an item is defined in one embodiment, it does not need to be further defined and explained in subsequent embodiments. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0051] It should also be noted that although the order of steps is mentioned in the method description, in some cases, steps may be performed in a different order than that described here, and this should not be interpreted as a restriction on the order of steps.

[0052] This invention provides a 10kV automatic / through-line early warning and rapid fault handling system for railways. It is a comprehensive intelligent system that integrates multi-source sensing, data analysis, intelligent judgment and remote control. It can fundamentally improve the automation level of railway power supply, provide accurate early warning, rapid location, automatic isolation and intelligent recovery.

[0053] Specifically, such as Figure 1 The aforementioned early warning and rapid fault handling system is a three-layer distributed architecture, including:

[0054] (1) Sensing layer, used to collect status data and meteorological parameters of 10kV automatic blocking / through lines of railway, and to perform rapid isolation and switching of transient faults.

[0055] The perception layer includes:

[0056] The integrated intelligent device is used to monitor voltage quality, load current, and traveling wave signals in real time, perform power line switching, and collect and transmit fault data.

[0057] Micro-meteorological monitoring devices are used to collect meteorological parameters, including temperature, humidity, wind speed, wind direction, and rainfall.

[0058] Tower tilt monitoring sensors are used to continuously assess the stability of tower foundations and can prevent tower collapse accidents caused by geological disasters.

[0059] Image acquisition device is used to automatically identify potential hazards such as overheating of joints and hanging foreign objects.

[0060] Among them, such as Figure 2 The integrated smart device includes:

[0061] Current transformers are used to monitor load current.

[0062] Voltage sensors are used to monitor voltage quality.

[0063] Traveling wave acquisition module, used to monitor traveling wave signals.

[0064] A voltage vacuum circuit breaker is used to perform rapid disconnection and opening / closing of transient faults.

[0065] The communication management module is used for communication connections between different devices.

[0066] The clock module is used to acquire BeiDou clock signals, providing clock signals for the integrated smart device and synchronizing the clocks of all devices.

[0067] When a sudden increase in current, a sudden drop in voltage, or a traveling wave signal is detected, the local traveling wave location algorithm is immediately activated. After initial judgment, the results are packaged and uploaded to the platform layer. The platform layer comprehensively analyzes and confirms the fault location, and remotely controls the voltage vacuum circuit breaker to physically isolate the faulty section.

[0068] The integrated intelligent device is installed at 3-kilometer intervals. This device integrates a current transformer, voltage sensor, traveling wave acquisition module, voltage vacuum circuit breaker, communication management module, and clock module, serving not only as a data acquisition terminal but also as an edge intelligent node with autonomous decision-making capabilities. Among them:

[0069] The current transformer uses a high-frequency current transformer with a sampling rate of 2MHz;

[0070] The voltage sensor adopts a capacitive voltage divider type with an accuracy class of 0.2 and a temperature coefficient of ≤50ppm / ℃, which can realize real-time monitoring of voltage quality and power load monitoring.

[0071] The voltage vacuum circuit breaker has millisecond-level fast opening and closing capability, with an opening time of <60ms and a closing time of <80ms, meeting the requirements for rapid clearing and reclosing of transient faults;

[0072] The communication management module supports multiple protocol conversions such as IEC 61850, Modbus, and MQTT, enabling seamless connectivity with different devices;

[0073] The clock module can acquire BeiDou clock signals with a time synchronization accuracy of up to 20ns. In addition to providing clock signals for this device, all acquisition terminals are connected to ensure strict alignment of electrical quantities and environmental quantities on the time axis, laying the foundation for cross-system joint analysis.

[0074] The main control unit of the integrated intelligent device is a processor core module, adopting a dual-core ARM Cortex-A53 architecture with a main frequency of 1.2GHz and equipped with an independent floating-point arithmetic unit. It is configured with 2GB of DDR4 high-speed memory with a bandwidth of 3200MT / s. The storage system is 32GB of eMMC embedded storage, supporting wear leveling algorithms. It possesses edge computing capabilities, completing fault type classification within 10ms, performing preliminary fault location, and transmitting fault data to the system master station within 1 minute.

[0075] The micro-meteorological monitoring device is set up at a distance of 5 kilometers, integrating sensors for six elements: temperature, humidity, air pressure, wind speed, wind direction, and rainfall, with a data refresh rate of 1Hz.

[0076] The tower tilt monitoring sensors are installed at 3-kilometer intervals and directly connected to the integrated intelligent device. They have dual-axis tilt angles with an accuracy of ±0.1°, continuously assessing the stability of the tower foundation and preventing tower collapse accidents caused by geological disasters.

[0077] The image acquisition device is installed at outdoor equipment such as cable terminals and intermediate joints, conductor connection points, transformers and integrated intelligent devices in mixed power lines. It is equipped with a 2-megapixel infrared thermal imager and supports AI image recognition algorithms to automatically identify potential hazards such as overheated joints and hanging foreign objects.

[0078] (2) Network layer, used to aggregate data acquired by the sensing layer to the platform layer in real time. Through wireless private network (such as LTE-4G, 5G) or fiber optic communication channel, data from various distributed sensors are aggregated to the platform layer in real time. It supports two-way communication mechanism to ensure the safe, reliable and low-latency issuance of remote control commands and meet the millisecond-level action requirements.

[0079] (3) Platform layer, used to comprehensively analyze and confirm the location of the fault, and then send remote control commands to the sensing layer of the fault section to physically isolate the fault section. The sensing layer collects data to provide early warning of electrical parameters and environmental status.

[0080] In this invention, the platform layer adopts a collaborative computing architecture, integrating multi-source heterogeneous data from SCADA systems, meteorological information systems, equipment management systems, etc., and integrating various intelligent analysis tools such as LSTM neural networks, random forests, graph theory algorithms, and knowledge graph engines to construct equipment relationship networks, fault mode libraries, and handling experience libraries. It also constructs a 1:1 three-dimensional model of the line to achieve intelligent operation and maintenance with virtual-real interaction.

[0081] In this invention, the information acquired by the traveling wave acquisition module is used for traveling wave positioning. Wavelet transform technology is employed to extract the arrival time of the traveling wave front, significantly improving the accuracy of wave front identification. The system can dynamically adjust the line impedance and propagation velocity parameters according to changes in ambient temperature, further enhancing positioning accuracy. For cable-overhead mixed lines, the system sets calculation parameters for each section to ensure positioning reliability under complex topologies. Simultaneously, combined with waveform feature analysis technology, it effectively distinguishes between phase-to-phase short circuits and single-phase grounding faults. Figure 3 The platform layer can comprehensively analyze and confirm the fault location based on traveling wave positioning, including:

[0082] S11: By detecting the original voltage / current waveforms and electrical information at the time of fault occurrence in the power system, energy surge detection is performed on the original waveforms to identify voltage / current abrupt change points in the waveforms. These abrupt change points usually correspond to the time when the fault occurs.

[0083] S12: Detected mutation points are marked as Wavelet transform is used to decompose waveforms containing abrupt changes, and further extract detailed features of the waveforms to more accurately capture wavefront moments.

[0084] S13: The decomposed waveform still contains Information on abrupt change points is obtained, and the arrival times of the wavefront at different detection points are accurately captured, denoted as t1 and t2 respectively. The time difference between the two points is then utilized. And combined with the dynamically corrected wave velocity v, through the formula The distance L from the fault point to the detection point is calculated. Based on the calculated fault distance L, the kilometer marker of the fault point is determined with an accuracy within ±100 meters. Considering the characteristics of mixed cable-overhead lines, the system dynamically corrects the line wave impedance and propagation velocity parameters according to changes in ambient temperature. Calculation parameters are set separately for different sections (cable sections and overhead sections) to achieve dynamic correction of wave velocity, thereby improving the accuracy of fault location.

[0085] The platform layer can collect data from the perception layer to provide early warnings of electrical parameters. The specific process is as follows:

[0086] S21: Acquire historical voltage quality data, load current data, and meteorological data, and predict electrical load change trends based on the ARIMA time series algorithm;

[0087] S22: Real-time monitoring of voltage deviation, establishment of a voltage over-limit risk prediction model based on historical data, and early identification of power quality anomalies through FFT spectrum analysis technology.

[0088] Simultaneously, the platform layer can also collect data from the perception layer to provide early warnings of environmental conditions. The specific process is as follows:

[0089] S31: Collect real-time meteorological parameters through micro-meteorological monitoring devices, and issue severe weather warnings 48 hours in advance in conjunction with weather forecasts;

[0090] S32: Collect tower tilt data through tower tilt monitoring sensors, and identify geological risks to foundation stability such as foundation settlement and landslides based on the trend of tilt data changes.

[0091] In addition, a power neural network model is constructed at the platform layer. The node features include state data. The LSTM neural network algorithm is used. The input feature vector includes 128-dimensional parameters (an array of 128 values, each value representing a standardized key parameter) such as historical load sequences, real-time meteorological data, and date type features. The model is automatically trained every hour, dynamically adjusting the prediction parameters to adapt to changes in operating conditions, assessing system vulnerability in real time, and identifying key nodes and weak links.

[0092] Furthermore, the platform layer employs a random forest algorithm to extract time information, status data, and meteorological parameters as key features, outputting a probability distribution of fault risk levels. Based on combinations of parameters such as temperature, humidity, and wind speed, an icing growth prediction model is established, issuing icing warnings 48 hours in advance to assist maintenance departments in developing de-icing plans or arranging special inspections, effectively preventing large-scale power outages caused by natural disasters.

[0093] like Figure 4 When a line fault occurs, this system, based on the fault information (including traveling wave data, protection action signals, etc.) uploaded by the integrated intelligent devices at both ends of the fault point, confirms the fault location through comprehensive analysis at the platform level, and then sends remote control commands to the intelligent terminals on both sides of the faulty section. Upon receiving the command, the terminal automatically disconnects the built-in voltage vacuum circuit breaker, achieving millisecond-level physical isolation of the faulty section. After isolation is completed, the terminal feeds back the status to the platform level, which then generates the optimal closing strategy and remotely controls adjacent switches to restore power to the non-faulty section. The entire process can be completed within 2 minutes, significantly reducing the scope and duration of the power outage.

[0094] like Figure 6 When a line fault occurs, this system automatically integrates all information, including waveform recordings, environmental data, operation logs, location results, and switch action records, to generate a standardized fault file and permanently archive it. Simultaneously, the system automatically generates an electronic maintenance work order, pushes it to the corresponding responsible work team, and includes on-site troubleshooting suggestions and repair plan guidelines, promoting the shift from "fault response" to "closed-loop management" and forming a closed loop.

[0095] like Figure 5The early warning and rapid fault handling system provided by this invention enables comprehensive and multi-dimensional real-time monitoring of the operating status of 10kV automatic / through railway lines and their surrounding environment, breaking through the limitations of monitoring only electrical quantities. Before a fault occurs, an intelligent early warning model is established based on status data and meteorological parameters to identify potential risks in advance, transforming passive repair into proactive defense. Furthermore, after a fault occurs, high-precision and rapid fault location can be achieved, and the isolation of the faulty section can be automatically completed, minimizing the scope and duration of power outages. This early warning and rapid fault handling system is a highly integrated and intelligent system architecture that combines sensing, communication, analysis, and control, breaking down information silos between subsystems, achieving autonomous decision-making and collaborative linkage across the entire chain, and comprehensively improving power supply reliability and operation and maintenance efficiency.

[0096] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A railway 10kV automatic gate / through line early warning and rapid fault handling system, characterized in that: The early warning and rapid fault handling system is a three-layer distributed architecture, including: The sensing layer is used to collect status data and meteorological parameters of the 10kV automatic shut-off / through lines of the railway, and to perform rapid isolation and switching on / off of transient faults. The network layer is used to aggregate data acquired by the perception layer to the platform layer in real time; The platform layer is used to comprehensively assess and confirm the location of the fault, and then sends remote control commands to the sensing layer of the faulty section to physically isolate the faulty section. The sensing layer collects data to provide early warnings of electrical parameters and environmental conditions.

2. The railway 10kV automatic gate / through line early warning and rapid fault handling system according to claim 1, characterized in that: The perception layer includes: An integrated intelligent device is used to monitor voltage quality, load current, and traveling wave signals in real time, perform power line switching, and collect and transmit fault data. Micro-meteorological monitoring devices are used to collect meteorological parameters; Tower tilt monitoring sensors are used to continuously assess the stability of tower foundations; Image acquisition device is used to automatically identify potential hazards such as overheating of joints and hanging foreign objects.

3. The railway 10kV automatic gate / through line early warning and rapid fault handling system according to claim 2, characterized in that: The integrated smart device includes: Current transformers are used to monitor load current; Voltage sensors are used to monitor voltage quality; Traveling wave acquisition module, used to monitor traveling wave signals; Voltage vacuum circuit breakers are used to perform rapid disconnection and opening / closing of transient faults; When a sudden increase in current, a sudden drop in voltage, or a traveling wave signal is detected, the local traveling wave location algorithm is immediately activated. After initial judgment, the results are packaged and uploaded to the platform layer. The platform layer comprehensively analyzes and confirms the fault location, and remotely controls the voltage vacuum circuit breaker to physically isolate the faulty section.

4. The railway 10kV automatic gate / through line early warning and rapid fault handling system according to claim 3, characterized in that: The integrated smart device also includes: The communication management module is used for communication connections between different devices; The clock module is used to acquire BeiDou clock signals, providing clock signals for the integrated smart device and synchronizing the clocks of all devices.

5. The railway 10kV automatic gate / through line early warning and rapid fault handling system according to claim 4, characterized in that: The platform layer includes intelligent analysis tools and a knowledge graph engine, which integrate multi-source heterogeneous data from SCADA systems, meteorological information systems, and equipment management systems, and uses LSTM neural networks and random forests for analysis and computation.

6. The railway 10kV automatic gate / through line early warning and rapid fault handling system according to claim 5, characterized in that: The platform layer comprehensively analyzes and confirms the location of the fault, including: The system detects the original voltage / current waveform and electrical information at the time of the fault, performs energy surge detection on the original waveform, identifies voltage / current abrupt change points in the waveform, and the abrupt change points correspond to the time of the fault occurrence. The detected mutation point is marked as The waveforms containing the mutation points are decomposed by using wavelet transform, and the detail features of the waveforms are further extracted; The decomposed waveforms still contain mutation point information, capturing the time when the wave head reaches different detection points, respectively denoted as t1 and t2, and the distance L from the fault point to the detection point is calculated by the formula combined with the dynamically corrected wave speed v. ​ Based on the calculated fault distance L, the kilometer marker of the fault point is determined.

7. The railway 10kV automatic gate / through line early warning and rapid fault handling system according to claim 6, characterized in that: The platform layer collects data from the perception layer to provide early warnings of electrical parameters, including: Historical voltage quality data, load current data, and meteorological data are acquired, and the electrical load change trend is predicted based on the ARIMA time series algorithm. Real-time monitoring of voltage deviation, establishment of a voltage over-limit risk prediction model based on historical data, and early identification of power quality anomalies through FFT spectrum analysis technology.

8. The railway 10kV automatic gate / through line early warning and rapid fault handling system according to claim 7, characterized in that: The platform layer collects data from the perception layer to provide early warnings of environmental conditions, including: Real-time meteorological parameters are collected by micro-meteorological monitoring devices and combined with weather forecasts to issue severe weather warnings 48 hours in advance. Tower tilt data is collected by tower tilt monitoring sensors, and the foundation stability risk is identified based on the trend of tilt data changes.

9. The railway 10kV automatic gate / through line early warning and rapid fault handling system according to claim 8, characterized in that: The platform layer constructs a power neural network model, with node features including state data. It uses an LSTM neural network to calculate the data, assess system vulnerability in real time, and identify critical nodes and weak links.

10. The railway 10kV automatic gate / through line early warning and rapid fault handling system according to claim 9, characterized in that: The platform layer uses the random forest algorithm to extract time information, state data and meteorological parameters as key features, and outputs the probability distribution of fault risk level.

Images