A method for real-time monitoring of non-insulated rail breakage based on frequency division multiplexing

By deploying monitoring nodes on the rails and utilizing frequency division multiplexing technology and segmented relay signal transmission, the problem of the inability to monitor rail fractures in real time in existing technologies has been solved. This enables low-cost, real-time, and accurate rail breakage monitoring, which is applicable to turnout areas and can coexist with existing axle counting systems with strong anti-interference capabilities.

CN123104746APending Publication Date: 2026-11-17CHINA RAILWAY FIRST SURVEY & DESIGN INST GRP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611001252.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-07
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies cannot achieve real-time monitoring of rail breaks in long rail sections without adding insulation joints, especially in turnout areas where effective monitoring is not possible. Furthermore, existing solutions are costly, complex, or pose safety hazards.

Method used

By employing frequency division multiplexing technology and segmented relay signal transmission principle, multiple monitoring nodes are deployed on the rails, and carrier signals of different frequencies are used to achieve signal isolation and relay transmission between adjacent sections. Combined with frequency jump anti-interference and environmental adaptive adjustment, rail break monitoring and location are realized.

Benefits of technology

It enables low-cost, real-time monitoring of rail fractures without damaging the rail structure. It is suitable for turnout areas, can coexist with existing axle counting systems, has accurate positioning, strong anti-interference capabilities, and avoids false alarms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123104746A_ABST
    Figure CN123104746A_ABST
Patent Text Reader

Abstract

This invention belongs to the field of urban rail transit rail fracture monitoring technology, specifically providing a real-time monitoring method for uninsulated rail fractures based on frequency division multiplexing. This method does not install any mechanical or electrical insulation joints within the target rail section. Multiple monitoring nodes are deployed at predetermined intervals along the rail extension direction. Different frequency carrier signals are used to achieve signal isolation and relay transmission between adjacent sections. The location of the rail fracture is determined and located by signal interruption. This invention is suitable for direct installation on seamless tracks and existing axle counting lines, and coexists perfectly with the axle counting system. The frequency allocation strategy and double-rail staggered layout scheme are clearly defined, enhancing the determinism of signal isolation. Furthermore, each section is monitored independently, resulting in high accuracy in rail fracture location.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of urban rail transit rail fracture monitoring technology, specifically involving a method for real-time rail fracture monitoring and fracture location of long-distance rail sections (including turnout areas) without adding mechanical or electrical insulation joints, utilizing frequency division multiplexing technology and segmented relay signal transmission principle. Background Technology

[0002] Axle counting systems and track circuits are two mainstream technologies for monitoring track occupancy and clearance in railway signaling systems. Track circuits, based on the "current measurement loop" principle, inherently possess the capability to detect rail breaks; that is, when a rail breaks, the circuit is interrupted, which manifests as an "occupancy" state. Axle counting systems, on the other hand, use a "counting" principle and can only determine whether a section has a train; they cannot detect rail breaks within that section, thus presenting a safety vulnerability.

[0003] Currently, there are several solutions to the problem that axle counting systems cannot detect rail breaks, but all of them have limitations:

[0004] Skylight inspection plan: Real-time monitoring cannot be achieved during operating hours.

[0005] Vehicle-mounted flaw detection solution: It cannot detect the condition of the rails in front of the train, posing a safety hazard.

[0006] The combination of axle counting and track circuit is complex and costly. Furthermore, the non-insulated track circuit has strict requirements on the resistance of the track bed, resulting in a large amount of modification work.

[0007] The axle counter sensor also serves as a rail breakage detection solution: it relies on rail return flow, cannot work when there are no cars or when the train is not in operation at night, and is easily affected by unstable traction current.

[0008] Ultrasonic guided wave rail breakage detection solution: Due to the complex structure of the turnout area, the signal attenuation is severe, making it unsuitable for effective application. However, the turnout area is precisely the area with a high incidence of rail breakage.

[0009] In summary, the existing technology lacks a low-cost, real-time monitoring solution for track breakage that is applicable to turnout areas and can coexist with existing axle counting systems. Summary of the Invention

[0010] The purpose of this invention is to provide a real-time monitoring method for non-insulated rail breaks based on frequency division multiplexing, so as to solve the technical problems of existing rail break monitoring technologies, such as the need to add insulation joints, system complexity, high cost, inability to effectively monitor turnout areas, and lack of active linkage mechanism.

[0011] To achieve the above objectives, the present invention adopts the following technical solution:

[0012] A real-time monitoring method for rail breaks in uninsulated rails based on frequency division multiplexing includes the following steps:

[0013] Step S1: Divide the entire rail into N continuous monitoring sections along the extension direction of the target rail with a preset section length L. Set up a monitoring node at the boundary point of each section. The monitoring node includes a signal transmitting unit and a signal receiving unit.

[0014] Step S2: Adjacent monitoring nodes on the same rail have different carrier frequencies, and the same mileage position on two rails has different carrier frequencies.

[0015] Step S3: The signal transmitting units of all monitoring nodes inject a characteristic current signal into one side of the monitoring section on both sides of them according to the carrier frequency allocated in step S2. The signal receiving units of the adjacent monitoring nodes located at the other end of the section receive the characteristic current signal at the same frequency.

[0016] Step S4: When a signal receiving unit of a certain monitoring node fails to receive a characteristic current signal from an adjacent monitoring node within multiple consecutive monitoring cycles, it is determined that there is a rail break in the rail section corresponding to that receiving unit.

[0017] Step S5: Send the rail break location information obtained in step S4 to the operation and maintenance terminal, and at the same time output a rail break alarm signal to the computer interlocking system, so that the computer interlocking system can perform safety protection operations.

[0018] Furthermore, the monitoring node is also allocated a spare frequency. In step S4, in addition to receiving the characteristic current signal from the adjacent monitoring node on one side at the carrier frequency allocated in step S2, the signal receiving unit of the monitoring node also listens to the signal from the adjacent node on the other side with lower sensitivity as an auxiliary criterion. If the monitoring node receives the auxiliary signal from the adjacent node on the other side but does not receive the main signal, it is determined that the section corresponding to the main signal is broken. If the signal is not received in both directions, it is determined that the monitoring node itself is faulty.

[0019] Furthermore, the method also includes a frequency switching anti-interference step: each monitoring node monitors the signal-to-noise ratio of the currently used frequency in real time. When the signal-to-noise ratio is continuously lower than a preset threshold, the monitoring node automatically switches to the backup frequency and sends a frequency switching notification to the adjacent nodes, and the adjacent nodes switch to the same backup frequency synchronously.

[0020] Furthermore, the method also includes an environmental adaptive adjustment step. The monitoring node integrates an environmental adaptive adjustment unit, which includes a temperature sensor and a humidity sensor to measure the rail temperature and track bed humidity parameters in real time. According to a preset compensation curve, the transmission power and reception sensitivity of the characteristic current signal are dynamically adjusted.

[0021] Furthermore, in step S5, after receiving the alarm signal, the computer interlocking system automatically performs at least one of the following operations: cancels the approved route through the broken rail section, prohibits the opening of signals through the broken rail section, and issues a stop command to the approaching train.

[0022] Furthermore, in step S3, the characteristic current signal is a square wave signal, a sine wave signal, or an coded pulse signal, and the signal strength is adaptively adjusted according to the section length and the rail attenuation characteristics.

[0023] Furthermore, in step S4, the continuous monitoring cycles are 3 to 5 cycles, and each monitoring cycle is 1 to 10 seconds.

[0024] Furthermore, the monitoring node is fixed to the rail web position by clamps, and its working power supply and communication link both utilize the spare cores in the existing signal cable.

[0025] The effective effects of this invention are as follows:

[0026] 1. Completely joint-free, maintaining the physical continuity of the rail.

[0027] This invention does not rely on any mechanical insulation joints (no damage to the rail structure is required) or electrical insulation joints (no 26m tuning zone is required), and achieves signal isolation between adjacent sections entirely through frequency division multiplexing technology. The rail maintains complete physical continuity, does not affect the smoothness of train operation, does not change the traction return path, and is suitable for direct installation on seamless tracks and existing axle counting tracks, coexisting perfectly with the axle counting system.

[0028] 2. Frequency division multiplexing provides natural isolation and strong anti-interference capability.

[0029] By assigning different frequencies to adjacent sections (100Hz, 300Hz, 500Hz... for odd-numbered nodes, and 200Hz, 400Hz, 600Hz... for even-numbered nodes), natural isolation of signals between adjacent sections is achieved, fundamentally avoiding signal crosstalk. The staggered frequency layout of the two rails (one rail using an odd-numbered sequence and the other an even-numbered sequence at the same mileage) further reduces crosstalk between the two rails.

[0030] 3. Relay transmission covers long distances with precise positioning.

[0031] Each monitoring node acts as both a signal receiver for the previous segment and a signal transmitter for the next segment, forming a relay link. Compared to existing technologies using carrier communication, this invention further clarifies the frequency allocation strategy and the dual-track staggered layout scheme, enhancing the determinism of signal isolation. Furthermore, independent monitoring of each segment ensures high accuracy in locating track breaks.

[0032] 4. Frequency hopping technology enhances resistance to traction current.

[0033] When a monitoring node detects strong electromagnetic interference such as traction current harmonics on its current operating frequency, it automatically switches to a backup frequency (e.g., 100Hz → 110Hz), and adjacent nodes switch synchronously to maintain communication continuity. This mechanism effectively solves the problem of strong interference from traction return current to monitoring signals in electrified railway rails, realizing a differentiated technical approach of active anti-interference rather than passive filtering.

[0034] 5. Dual-signal auxiliary criteria to distinguish between rail breakage and equipment failure.

[0035] The receiving unit simultaneously listens for the main signal at the primary frequency and for the auxiliary signal on the other side at a lower sensitivity. This design allows the system to distinguish between "signal loss due to a rail break" and "inability to receive signals due to equipment malfunction." If the auxiliary signal is received but the main signal is not, it indicates a rail break; if neither signal is received, it indicates a node malfunction. This mechanism effectively avoids false alarms caused by equipment failure.

[0036] 6. Also applicable to turnout areas

[0037] This invention does not rely on detection principles that require continuous propagation media, such as ultrasonic waves, and is not affected by rail gaps, switch rail and main rail overlap structures. It can be deployed and operated normally in turnout areas, filling the gap in existing technologies that cannot effectively monitor turnout areas. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the monitoring node deployment and frequency allocation of the present invention.

[0039] Figure 2 This is a schematic diagram illustrating the principle of track breakage monitoring and determination in this invention.

[0040] Figure 3 This is a schematic diagram of the frequency hopping anti-interference mechanism of the present invention.

[0041] Figure 4 This is a schematic diagram of the system architecture of the present invention.

[0042] Explanation of the labels in the diagram:

[0043] 1-First rail; 2-Second rail; 3-Monitoring node (including transmitting and receiving units); 4-Characteristic current signal; 5-Main monitoring signal transmission path; 6-Auxiliary monitoring signal transmission path; 7-Broken rail point; 8-Broken rail monitoring host; 9-Signal centralized monitoring system; 10-Computer interlocking system; 11-Signal ATS subsystem. Detailed Implementation

[0044] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0045] Part 1: System Composition and Deployment

[0046] This invention does not install any mechanical or electrical insulation joints within the target rail section. Instead, it deploys multiple monitoring nodes at predetermined intervals along the rail extension direction and uses carrier signals of different frequencies to achieve signal isolation and relay transmission between adjacent sections. The location of the rail break is determined and located by signal interruption.

[0047] See Figures 1 to 4 This embodiment uses a 10-kilometer double-track railway line as an example for illustration. The axle counting system on this line serves as a means of detecting track occupancy, but it does not have any rail breakage monitoring function. The monitoring system of this invention includes: a monitoring node 3 deployed every 1 kilometer along the extension direction of the rails, a rail breakage monitoring host 8 installed indoors, a centralized signal monitoring system 9, a computer interlocking system 10, and a signal ATS subsystem 11.

[0048] Monitoring node installation

[0049] Each monitoring node is secured to the rail web using clamps, eliminating the need for drilling or welding. The node's housing is an IP67-rated metal shielded box, internally integrating a signal transmitting unit, a signal receiving unit, a microprocessor, a frequency hopping control module, a temperature and humidity sensor, and a power management module. The node's power supply is drawn from a spare conductor in the existing trackside signal cable. Wired communication between the node and the indoor control unit is also achieved via a spare conductor in the existing signal cable.

[0050] Frequency planning

[0051] Taking the first rail 1 as an example, node A is set at kilometer 0, with a main frequency of 100Hz and a backup frequency of 110Hz; node B is set at kilometer 1, with a main frequency of 300Hz and a backup frequency of 310Hz; node C is set at kilometer 2, with a main frequency of 500Hz and a backup frequency of 510Hz; node D is set at kilometer 3, with a main frequency of 700Hz and a backup frequency of 710Hz... and so on. The frequencies of adjacent nodes are different, with a main frequency interval of 200Hz and a backup frequency interval of 200Hz.

[0052] The frequency allocation of the second rail 2 is staggered from that of the first rail: at the same mileage position, the nodes on the second rail are assigned a number sequence frequency (200Hz, 400Hz, 600Hz, 800Hz...), forming a frequency staggered layout with the first rail, with spare frequencies of 210Hz, 410Hz, 610Hz, 810Hz... respectively.

[0053] Auxiliary monitoring configuration

[0054] Each monitoring node not only receives signals from adjacent nodes on one side at its primary listening frequency, but also listens for signals from adjacent nodes on the other side at a lower sensitivity (sensitivity reduced by approximately 20dB). Taking node B as an example: it primarily listens for the 100Hz signal from node A (strong listening) and secondarily listens for the 500Hz signal from node C (weak listening). During normal operation, node B simultaneously receives the 100Hz signal (strong) from node A and the 500Hz signal (weak, due to attenuation from traversing 1 km of rail) from node C. If node B receives the 500Hz auxiliary signal but not the 100Hz primary signal, it is determined that there is a rail break between node A and node B; if neither signal is received, it is determined that node B itself may have a fault.

[0055] Part Two: Signal Transmission and Relay Principles

[0056] See Figure 2 Node A's transmitting unit injects a characteristic current signal 4 into the rail section to the right (1 km direction) at a frequency of 100 Hz. After the signal propagates 1 km along the rail, Node B's receiving unit receives the signal at the same frequency (100 Hz), completing the monitoring of the first section. Simultaneously, Node B's transmitting unit injects a characteristic current signal into the rail section to the right (2 km direction) at its allocated main frequency (300 Hz). Node C's receiving unit receives the signal at a frequency of 300 Hz. This process continues, forming a relay-style signal transmission link—each node acts as both a signal receiver for the previous section and a signal transmitter for the next section.

[0057] Each node's signal transmitting unit is driven by a constant current source to ensure that the output signal strength is unaffected by changes in rail impedance. The receiving unit uses a narrowband filter and a high-gain amplifier to effectively suppress out-of-band interference. The signal monitoring period is 5 seconds, with a characteristic current signal transmitted once per period. If no signal is received for three consecutive periods, the alarm logic is triggered.

[0058] Part Three: Rail Breakage Detection and Alarm

[0059] See Figure 2 When the rail break point 7 occurs in the section between node A and node B, the 100Hz main signal cannot be transmitted from node A to node B. The receiving unit of node B does not receive the 100Hz main signal for three consecutive monitoring cycles (5 seconds per cycle, 15 seconds in total), but continuously receives a weak 500Hz auxiliary signal from node C. Based on this, it is determined that the rail break is in the section between node A and node B.

[0060] Node B immediately sends the rail break alarm information (including Node B number, location of the rail break section (0-1 km), time of rail break occurrence, and auxiliary signal reception status) to the indoor host 8 via the communication link. After comprehensively analyzing the alarm information, the indoor host sends the rail break location information and a red alarm signal to the centralized signal monitoring system 9, and simultaneously outputs a relay dry contact closing signal to the computer interlocking system 10.

[0061] Upon receiving a rail break alarm signal, the computer interlocking system immediately and automatically performs the following protective operations: cancels all approved routes passing through the rail break section; prohibits the opening of any signals passing through the rail break section; forwards the rail break section closure information to the Automatic Train Monitoring System 11, which then issues voice prompts and emergency stop commands to approaching trains.

[0062] Part Four: Frequency Hopping Anti-interference Mechanism

[0063] See Figure 3 Each monitoring node continuously monitors the signal-to-noise ratio (SNR) of the currently used frequency. Taking node A as an example, when it detects that the SNR of the 100Hz main frequency is consistently below 20dB (indicating that the frequency is subject to strong electromagnetic interference, such as traction current harmonics or co-frequency interference generated by external equipment), node A's frequency hopping control module initiates the frequency hopping process:

[0064] Node A switches to the backup frequency of 110Hz and generates a frequency switching notification message, which is then sent to the adjacent node B via rail carrier communication.

[0065] After receiving the frequency switching notification, the receiver of node B synchronously switches its receiving frequency from 100Hz to 110Hz.

[0066] The transmitting end of node A continues to inject the characteristic current signal at a frequency of 110Hz, and the receiving end of node B receives it at a frequency of 110Hz, achieving seamless switching and uninterrupted communication.

[0067] Meanwhile, node A reports the frequency switching event to the indoor host and records it in the operation and maintenance log.

[0068] This frequency hopping mechanism differs from the passive filtering schemes in existing technologies, achieving active anti-interference and is particularly suitable for subway lines with abundant traction current harmonics and complex electromagnetic environments.

[0069] Part 5: Environmental Adaptive Adjustment

[0070] The monitoring nodes have built-in temperature and humidity sensors that collect rail temperature and track bed humidity data once per minute. When the system detects changes in environmental parameters, it performs the following adaptive adjustments:

[0071] Temperature compensation: With 20℃ as the reference temperature, when the rail temperature drops by 10℃, the rail resistivity increases by about 4%, and the system automatically increases the signal transmission power by 4%~6% to compensate for the attenuation; when the temperature rises, the transmission power is reduced accordingly to save energy.

[0072] Humidity compensation: When the track bed humidity rises from 50%RH to 90%RH, the leakage of the rail to the ground increases, and the signal strength at the receiving end may decrease. The system automatically increases the receiving sensitivity threshold to avoid false judgments triggered by normal humidity changes.

[0073] Compensation curve: Based on the temperature-humidity-signal attenuation mathematical model established by field calibration data, continuous linear adjustment of transmit power and receive sensitivity is achieved.

[0074] The environmental adaptive adjustment mechanism ensures that the accuracy and false alarm rate of the track break detection remain stable under the changing temperature and humidity conditions of the four seasons, and are not affected by the external environment.

[0075] Part Six: Special Handling of Switch Area

[0076] In the turnout section, due to the overlapping structure of the switch rail and stock rail, the electrical continuity of the rails is maintained (the switch rail and stock rail are electrically connected through limit switches and spacers), and the signal transmission of this scheme is unaffected. The monitoring node is installed at the web position of the stock rail, avoiding the movable area of ​​the switch rail, to prevent interference with turnout switching.

[0077] For the rail gap location where the fishplate connects within the turnout area, if the fishplate becomes loose, causing an abnormal increase in contact resistance, the characteristic current signal will be significantly attenuated when passing through this location, resulting in a decrease in signal strength at the receiving end. By analyzing the trend of signal strength changes in adjacent monitoring cycles (e.g., a signal strength decrease of more than 30% in several consecutive cycles), the indoor control unit can provide early warning of fishplate loosening, enabling preventative maintenance rather than just post-traumatic rail breakage alarms.

[0078] Part 7: Coexistence with Existing Axle Counting Systems

[0079] The monitoring nodes in this solution are installed on the web of the rail, offset by more than 1 meter from the wheel axle sensors of the axle counting system. Due to their different installation locations, there is no spatial conflict. The monitoring signal frequency (100Hz and above) and the operating frequency of the axle counting sensors (typically several kilohertz to tens of kilohertz) are not in the same frequency band; physical isolation is achieved through frequency division, preventing mutual interference. The two systems have independent power supply, independent communication, and independent judgment, without affecting each other.

[0080] When this solution detects a rail break, the route is blocked via the interlocking system, while the axle counting system continues to operate normally (it is normal for trains to be unable to enter the rail break section). The two systems complement each other rather than replace each other. This solution fills the safety gap in the axle counting system, which "can only determine whether there is a train, but cannot determine whether the track is broken."

Claims

1. A method for real-time monitoring of rail breaks in non-insulated steel rails based on frequency division multiplexing, characterized in that, Includes the following steps: Step S1: Divide the entire rail into N continuous monitoring sections along the extension direction of the target rail with a preset section length L. Set up a monitoring node at the boundary point of each section. The monitoring node includes a signal transmitting unit and a signal receiving unit. Step S2: Adjacent monitoring nodes on the same rail have different carrier frequencies, and the same mileage position on two rails has different carrier frequencies. Step S3: The signal transmitting units of all monitoring nodes inject a characteristic current signal into one side of the monitoring section on both sides of them according to the carrier frequency allocated in step S2. The signal receiving units of the adjacent monitoring nodes located at the other end of the section receive the characteristic current signal at the same frequency. Step S4: When a signal receiving unit of a certain monitoring node fails to receive a characteristic current signal from an adjacent monitoring node within multiple consecutive monitoring cycles, it is determined that there is a rail break in the rail section corresponding to that receiving unit. Step S5: Send the rail break location information obtained in step S4 to the operation and maintenance terminal, and at the same time output a rail break alarm signal to the computer interlocking system, so that the computer interlocking system can perform safety protection operations.

2. The method for real-time monitoring of rail breaks in non-insulated rails based on frequency division multiplexing according to claim 1, characterized in that, The monitoring node is also allocated a spare frequency. In step S4, in addition to receiving the characteristic current signal from the adjacent monitoring node on one side at the carrier frequency allocated in step S2, the signal receiving unit of the monitoring node also listens to the signal from the adjacent node on the other side with lower sensitivity as an auxiliary criterion. If the monitoring node receives the auxiliary signal from the adjacent node on the other side but does not receive the main signal, it is determined that the section corresponding to the main signal is broken. If the signal is not received in both directions, it is determined that the monitoring node itself is faulty.

3. The method for real-time monitoring of rail breaks in non-insulated rails based on frequency division multiplexing according to claim 2, characterized in that, The method also includes a frequency switching anti-interference step: each monitoring node monitors the signal-to-noise ratio of the currently used frequency in real time. When the signal-to-noise ratio is continuously lower than a preset threshold, the monitoring node automatically switches to the backup frequency and sends a frequency switching notification to the adjacent nodes. The adjacent nodes then switch to the same backup frequency synchronously.

4. The method for real-time monitoring of rail breaks in non-insulated rails based on frequency division multiplexing according to claim 3, characterized in that, The method also includes an environmental adaptive adjustment step. The monitoring node integrates an environmental adaptive adjustment unit, which includes a temperature sensor and a humidity sensor to measure the rail temperature and track bed humidity parameters in real time. According to a preset compensation curve, the transmission power and reception sensitivity of the characteristic current signal are dynamically adjusted.

5. The method for real-time monitoring of rail breaks in non-insulated rails based on frequency division multiplexing according to claim 4, characterized in that, In step S5, after receiving the alarm signal, the computer interlocking system automatically performs at least one of the following operations: cancels the approved route through the broken rail section, prohibits the opening of signals through the broken rail section, and issues a stop command to the approaching train.

6. The method for real-time monitoring of rail breaks in non-insulated rails based on frequency division multiplexing according to claim 1, characterized in that, In step S3, the characteristic current signal is a square wave signal, a sine wave signal, or an coded pulse signal, and the signal strength is adaptively adjusted according to the section length and the rail attenuation characteristics.

7. The method for real-time monitoring of rail breaks in uninsulated rails based on frequency division multiplexing according to claim 1, characterized in that, In step S4, there are 3 to 5 consecutive monitoring cycles, and each monitoring cycle lasts from 1 to 10 seconds.

8. The method for real-time monitoring of rail breaks in non-insulated rails based on frequency division multiplexing according to claim 1, characterized in that, The monitoring node is fixed to the rail web position by clamps, and its working power and communication link both utilize the spare cores in the existing signal cable.