High-speed railway turnout fault multi-parameter real-time dynamic monitoring system and method
By combining distributed fiber optic vibration sensing with three-dimensional laser scanning technology, real-time multi-parameter monitoring of high-speed railway turnout faults is achieved, solving the problems of non-real-time monitoring and high false alarm rate in existing technologies, and achieving accurate fault identification and monitoring.
Patent Information
- Application Number
- CN202510861467.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-25
- Publication Date
- 2025-09-16
AI Technical Summary
Existing high-speed railway turnout detection methods are unable to achieve all-weather real-time dynamic monitoring, have slow response speeds, single monitoring parameters, high false alarm rates, and are unable to accurately identify abnormal turnout conditions.
Combining a distributed fiber optic vibration sensing system with three-dimensional laser scanning technology, it achieves real-time monitoring of multiple parameters through components such as narrow-linewidth lasers, fiber couplers, and galvanometer scanning probes. It uses phase-sensitive coherent optical time-domain reflectometry to detect abnormal vibration signals, and combines it with a three-dimensional laser scanning system for geometric parameter detection to accurately identify switch faults.
It realizes real-time dynamic monitoring of high-speed railway turnout faults, improves the sensitivity and accuracy of monitoring, reduces the false alarm rate, can accurately identify the type of turnout fault, improves the time synchronization and integration of the system, and reduces hardware costs.
Smart Images

Figure CN120651334A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of distributed optical fiber sensing technology, and in particular to a multi-parameter real-time dynamic monitoring system and method for high-speed railway turnout faults. Background Art
[0002] In recent years, my country's high-speed rail transportation industry has developed rapidly. High-speed rail has become an important part of my country's transportation system. Its safe and reliable operation is of great significance to national property protection, economic development, and the safety of people's lives. Because high-speed rail operates at speeds far higher than traditional railways, the requirements for tracks are also higher. Turnouts are the weak link in high-speed rail tracks. Due to the turning of high-speed trains, the turnout structure must withstand huge impact forces, which are prone to track wear and deformation, excessive degradation, and the inability of the point rail to fit closely with the base rail. These problems cause abnormal shaking when high-speed trains pass. However, traditional detection methods such as manual inspection and the deployment of point electrical sensors still have certain limitations. They suffer from many problems, such as the inability to provide all-weather real-time dynamic monitoring, slow response speed, single monitoring parameters, and high false alarm rates.
[0003] On high-speed railways, systems relying on a single parameter cannot accurately and effectively locate and identify abnormal turnout conditions. When a high-speed railway turnout experiences malfunctions during long-term operation, key components such as the point rail wear, deformation, and excessive degradation can generate abnormal vibration signals when a high-speed train passes through the turnout section. Therefore, a distributed fiber optic vibration sensing system combined with 3D laser scanning technology holds significant promise for high-speed railway turnout fault monitoring due to its inherent resistance to electromagnetic interference, high sensitivity, and lack of blind spots. Summary of the Invention
[0004] In order to solve the problems of poor real-time performance, single monitoring parameters, and high false alarm rate in existing track fault detection methods, this application proposes a multi-parameter real-time dynamic monitoring system and method for high-speed railway turnout faults.
[0005] The technical solution adopted in this application is: a multi-parameter real-time dynamic monitoring system for high-speed railway turnout faults, comprising a narrow linewidth laser, a first optical fiber coupler, and a second optical cable arranged on the inner side of the high-speed railway turnout point rail and along the guide rail, wing rail, and frog core direction, as well as a first optical cable and a galvanometer scanning probe arranged along the road outside the stock rail, wherein the first optical cable and the second optical cable are both connected to a distributed optical fiber acoustic wave sensing system, and abnormal vibration signals are detected by phase-sensitive coherent optical time domain reflectometry technology to restore and locate the turnout fault signal. The first optical cable and the galvanometer scanning probe are connected to a three-dimensional laser scanning system for detecting the geometric parameters of the turnout structure and accurately identifying the turnout fault; The three-dimensional laser scanning system includes a second erbium-doped fiber amplifier, a second circulator, a first photodetector, a ranging unit, an angle encoder, a main control unit, a point cloud generation module, a galvanometer scanning probe, a high-speed analog-to-digital converter and a driving module. The laser light emitted by the narrow linewidth laser is input to the first fiber coupler and divided into two paths, one of which is output to the input end of the second erbium-doped fiber amplifier, and the other is output to the distributed fiber acoustic wave sensing system. The output port of the second erbium-doped fiber amplifier is connected to port a of the second circulator; port b of the second circulator is connected to port b of the first optical cable; port c of the first optical cable is connected to port a of the galvanometer scanning probe; and the second Port c of the circulator is connected to the input end of the first photodetector; the output end of the first photodetector is connected to the input port of the ranging unit; the output port of the ranging unit is connected to the input end of the angle encoder; the output end of the angle encoder is connected to the input port c of the high-speed analog-to-digital converter; the output port f of the high-speed analog-to-digital converter is connected to the input end of the main control unit; the output end of the main control unit is connected to the input end of the point cloud generation module; the output port of the point cloud generation module is connected to the input port a of the host computer; the output port c of the host computer is connected to the input port of the drive module; and the output port of the drive module is connected to the input port b of the galvanometer scanning probe.
[0006] Furthermore, the distributed fiber optic acoustic wave sensing system includes a second fiber coupler, an acousto-optic modulator, an arbitrary waveform generator, a first erbium-doped fiber amplifier, a wavelength division multiplexer, a third fiber coupler, a first circulator, a fourth fiber coupler, a balanced photodetector, a first bandpass filter, a frequency mixing module, a sine wave generator, a low-pass filter, an orthogonal demodulator, a third circulator, a first optical cable, an optical cable connection box, a second optical cable, a pigtail terminal box, a second photodetector, a low-noise amplifier, a second bandpass filter, and an adaptive filter. Laser light emitted by a narrow linewidth laser is input to the first fiber coupler and divided into two paths, one of which is output to the input end of the second erbium-doped fiber amplifier and the other is output to input port a of the second fiber coupler. Thereafter, the laser light is divided into two paths, one of which is input as detection light from output port b to input port a of the acousto-optic modulator and the other is input as local light from output port c to input port a of the third fiber coupler. The output end of the arbitrary waveform generator is connected to input port b of the acousto-optic modulator. The output port c of the acousto-optic modulator is connected to the input end of the first erbium-doped fiber amplifier; the output end of the first erbium-doped fiber amplifier is connected to the input end of the wavelength division multiplexer; the output end of the wavelength division multiplexer is connected to the input port a of the third fiber coupler; The third fiber coupler splits the light into two paths, which are respectively injected into port a of the third circulator and port a of the first circulator through its output ports b and c; port b of the third circulator is connected to port a of the first optical cable; port b of the first circulator is connected to port a of the optical cable connection box; port b of the optical cable connection box is connected to port a of the second optical cable; port b of the second optical cable is connected to the pigtail terminal box; The output port c of the first circulator is connected to the input port a of the fourth optical fiber coupler; the output ports c and d of the fourth optical fiber coupler are connected to the input ports a and b of the balanced photodetector; the output port c of the balanced photodetector is connected to the input port a of the high-speed analog-to-digital converter; the output port d of the high-speed analog-to-digital converter is connected to the input port a of the adaptive filter; the output port c of the adaptive filter is connected to the input port of the first band-pass filter; the output port of the first band-pass filter is connected to the input port a of the mixing module; the input ports b and c of the mixing module are connected to the output ports a and b of the sine wave generator; the output port d of the mixing module is connected to the input port of the low-pass filter; the output port of the low-pass filter is connected to the input end of the orthogonal demodulator; the output end of the orthogonal demodulator is connected to the input port b of the host computer; The output port c of the third circulator is connected to the input end of the second photodetector; the output end of the second photodetector is connected to the input end of the low-noise amplifier; the output end of the low-noise amplifier is connected to the input end of the second band-pass filter; the output end of the second band-pass filter is connected to the input port b of the high-speed analog-to-digital converter; and the output port e of the high-speed analog-to-digital converter is connected to the input port b of the adaptive filter.
[0007] Furthermore, the first optical fiber coupler adopts an optical fiber coupler with a coupling ratio of 50:50.
[0008] Furthermore, the second fiber coupler uses a fiber coupler with a coupling ratio of 99:1, and 99% of the laser light is output as detection light, and 1% of the laser light is output as local light.
[0009] Furthermore, the acousto-optic modulator modulates the light into pulse light and introduces a frequency shift, and then outputs the pulse light to the input end of the first erbium-doped fiber amplifier; the arbitrary waveform generator generates an electrical pulse signal to drive the acousto-optic modulator.
[0010] Furthermore, at least one galvanometer scanning probe is provided and arranged in the area where the guide rails, wing rails and frog cores are located.
[0011] A high-speed railway turnout fault multi-parameter real-time dynamic monitoring method, using the high-speed railway turnout fault multi-parameter real-time dynamic monitoring system, includes the following steps: S1: Continuous narrow-linewidth laser light emitted by a narrow-linewidth laser is input into the first fiber coupler. The first fiber coupler splits the laser light into two parts with a 50:50 ratio. One beam of light is used for the distributed fiber acoustic wave sensing system and is injected into the second fiber coupler from the first fiber coupler. The second fiber coupler splits the laser light into two parts with a 99:1 ratio. 99% of the laser light is output as detection light and input into the acousto-optic modulator, and 1% of the laser light is output as local light and input into the fourth fiber coupler. The acousto-optic modulator modulates the light into pulsed light and introduces a frequency shift, which is then output to the first erbium-doped fiber amplifier. An arbitrary waveform generator generates an electrical pulse signal to drive the acousto-optic modulator. S2: The signal output by the acousto-optic modulator is power amplified and filtered before being input into the third fiber coupler, which is then divided into two output lights. The detection light signal of the third fiber coupler is input into port a of the third circulator and transmitted to the first optical cable through port b of the third circulator. The first optical cable collects a reference background noise signal. The collected background noise signal returns to the third circulator and is transmitted to the second photodetector through port c of the third circulator to convert the optical signal into an electrical signal. The reference signal that has completed the photoelectric conversion is pre-processed by the second photodetector through a low-noise amplifier and a second band-pass filter to suppress high-frequency electromagnetic interference. The pre-processed reference signal is transmitted to a high-speed analog-to-digital converter by the second band-pass filter. The analog signal is converted into a digital signal by the high-speed analog-to-digital converter and then transmitted to an adaptive filter. Based on the adaptive filtering algorithm, the filter parameters are adjusted in real time to minimize the interference of noise on the signal. S3: The detection light signal output by the third fiber coupler is injected into port a of the first circulator and transmitted to the optical cable connection box through port b of the first circulator. The detection light signal is injected into the second optical cable by the optical cable connection box to directly detect and transmit the vibration signal on the switch rail. The generated backward Rayleigh scattered light signal is returned to port b of the first circulator by the second optical cable through the optical cable connection box. The Rayleigh scattered signal is transmitted to the fourth fiber coupler through port c of the first circulator, and realizes coherent beat frequency with the reference light output by the second fiber coupler at the fourth fiber coupler. After the beat frequency signal is generated, it is transmitted to the balanced photodetector by the fourth fiber coupler. The beat frequency signal is converted into an electrical signal by the balanced photodetector and then output to a high-speed analog-to-digital converter to realize analog-to-digital conversion. S4: The high-speed analog-to-digital converter outputs the digital signal to the adaptive filter, which uses an adaptive filtering algorithm to filter out background noise. The digital signal is then transmitted to the first bandpass filter through the adaptive filter to filter out noise signals. The digital signal filtered by the first bandpass filter is then transmitted to the mixing module. The sine wave generator outputs standard sine and cosine wave signals and transmits them to the mixing module. The signals are orthogonally mixed in the mixing module to generate I and Q signals. The mixing module inputs the I and Q signals after orthogonal mixing into a low-pass filter for low-pass filtering. The filtered I and Q signals are then transmitted to the orthogonal demodulator, which uses a coordinate rotation digital calculation method to obtain the amplitude and phase of the vibration signal and performs phase unwrapping. The demodulated amplitude and phase signals are then sent to the host computer for preliminary fault classification processing. S5: The host computer is connected to the drive module. Once the host computer identifies and locates abnormal vibration, it will send a control command signal to the drive module. The drive module is connected to the galvanometer scanning probe, which is used to control the galvanometer scanning probe to perform three-dimensional laser scanning on the key track structure of the turnout to collect its three-dimensional geometric information. The detection laser output by the first fiber coupler is transmitted to the second erbium-doped fiber amplifier for optical power amplification. The detection light after optical power amplification is injected into port a of the second circulator and output through port b of the second circulator. It is transmitted to the galvanometer scanning probe via the first optical cable, and the laser beam is focused and the emission angle is controlled for three-dimensional scanning to obtain three-dimensional information of the turnout point rail. The reflected laser is then collected and focused to filter out ambient stray light, and then transmitted back to port b of the second circulator via the first optical cable. S6: The reflected laser signal is transmitted to the first photodetector through the port c of the second circulator to convert it into an electrical signal; the electrical signal that has completed the photoelectric conversion is transmitted to the ranging unit, and the distance is calculated by modulating the laser phase difference; the output end of the ranging unit is connected to the input end of the angle encoder, and the angle encoder records the deflection angle of the galvanometer in real time to provide spatial positioning information; the angle encoder is connected to the high-speed analog-to-digital converter to convert the analog signal into a digital signal and transmit it to the main control unit through the high-speed analog-to-digital converter, and the main control unit processes the distance and angle data in real time to generate three-dimensional coordinates; the generated three-dimensional coordinate data is transmitted by the main control unit to the point cloud generation module, and the single-point measurement data is integrated to generate a three-dimensional point cloud, and it is transmitted to the host computer to process the point cloud data for three-dimensional modeling and analysis of changes in the geometric parameters of the switch point rail, and alarm and fault identification are performed based on the standard track parameters.
[0012] Furthermore, in step S, the upper computer obtains the specific position of the abnormal vibration by performing disturbance location through differential, cumulative averaging, and threshold disturbance location methods, and performs phase restoration on the disturbance signal according to the disturbance location to obtain the original waveform of the disturbance signal, extracts its signal features, and performs preliminary fault type identification based on the support vector machine.
[0013] Furthermore, the frequency range of the low-pass filter is 0-50 MHz, filtering out noise outside this frequency and retaining only the signal with zero intermediate frequency.
[0014] The beneficial effects of this application compared to the prior art are: First, the integration of a distributed fiber-optic acoustic wave sensing system and 3D laser scanning technology has been applied to high-speed railway turnout fault monitoring for the first time. This system uses a distributed fiber-optic acoustic wave sensing system to monitor and collect abnormal vibration signals from turnouts. Combined with 3D laser scanning technology, this system performs 3D modeling and geometric parameter analysis of the turnout structure, enabling precise identification of turnout fault types. This addresses many of the challenges of traditional high-speed railway turnout monitoring methods, including slow response, limited monitoring parameters, high false alarm rates, and the inability to provide real-time, 24 / 7 dynamic monitoring.
[0015] Second, the distributed fiber optic acoustic sensing system and the 3D laser scanning system share the same laser, but a coupler is used to split the laser light source into two paths. This ensures the alignment of the 3D laser scanning geometric data with the distributed fiber optic acoustic sensing system's vibration data in fault monitoring for high-speed railway switches, improving system time synchronization, enhancing system integration, and reducing hardware costs.
[0016] Third, a multi-area deployment approach is adopted. Optical cables are laid along the inner sides of the switch rails on both sides of the high-speed rail, along the switch rails, wing rails, and frog cores. This directly captures the contact vibration signals between the wheels and rails and monitors changes in bending stress in key switch structures, predicting wear and deformation of the switch rails and other structures. Optical cables are also laid along the road surface outside the stock rails to collect background noise. An adaptive noise suppression circuit structure is combined with a distributed fiber optic acoustic wave sensing system to dynamically adjust filtering parameters, effectively separating the target signal from the complex background noise present in the detection environment, thereby improving the sensitivity and accuracy of fault monitoring. Scanning probes are also installed to collect three-dimensional information on key switch structures for modeling analysis and precision measurement, monitoring deformation and reduction deviations of the switch rails. The system offers advantages such as high accuracy, good real-time performance, long sensing distance, and resistance to electromagnetic interference. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The present application will be further described below with reference to the accompanying drawings: Figure 1 A schematic diagram of the structure of a high-speed railway turnout fault multi-parameter real-time dynamic monitoring system provided in an embodiment of the present application; Figure 2 A schematic diagram of the specific laying of a sensing optical fiber used in a turnout according to an embodiment of the present application; In the figure: 1 is a narrow-linewidth laser; 2 is a first fiber coupler; 3 is a second fiber coupler; 4 is an acousto-optic modulator; 5 is an arbitrary waveform generator; 6 is a first erbium-doped fiber amplifier; 7 is a wavelength division multiplexer; 8 is a third fiber coupler; 9 is a first circulator; 10 is a fourth fiber coupler; 11 is a balanced photodetector; 12 is a first bandpass filter; 13 is a frequency mixing module; 14 is a sine wave generator; 15 is a low-pass filter; 16 is an orthogonal demodulator; 17 is a second erbium-doped fiber amplifier; 18 is a second circulator; 19 is a first photodetector ; 20 is the distance measuring unit; 21 is the angle encoder; 22 is the main control unit; 23 is the point cloud generation module; 24 is the third circulator; 25 is the first optical cable; 26 is the optical cable connection box; 27 is the second optical cable; 28 is the pigtail terminal box; 29 is the galvanometer scanning probe; 30 is the second photodetector; 31 is the low noise amplifier; 32 is the second bandpass filter; 33 is the high-speed analog-to-digital converter; 34 is the adaptive filter; 35 is the host computer; 36 is the drive module; 37 is the point rail; 38 is the guide rail; 39 is the wing rail; 40 is the frog core; 41 is the basic rail. DETAILED DESCRIPTION
[0018] like Figure 1 and 2 As shown, this application provides a multi-parameter real-time dynamic monitoring system for high-speed railway turnout faults. This system detects abnormal vibration signals through a distributed fiber optic acoustic wave sensing system based on phase-sensitive coherent optical time-domain reflectometry technology, enabling the restoration and location of track turnout fault signals. Furthermore, the system integrates three-dimensional laser scanning technology and fiber optic sensing technology to detect the geometric parameters of the turnout structure, detect abnormal wear and deformation of the turnout point rail, and detect excessive reduction values. This allows for accurate identification of turnout faults, enabling real-time monitoring of the high-speed railway turnout status, as well as accurate location, alarm, and classification of faults.
[0019] like Figure 1As shown, the monitoring system includes a narrow linewidth laser 1, a first fiber coupler 2, and a second fiber coupler 3, an acousto-optic modulator 4, an arbitrary waveform generator 5, a first erbium-doped fiber amplifier 6, a wavelength division multiplexer 7, a third fiber coupler 8, a first circulator 9, a fourth fiber coupler 10, a balanced photodetector 11, a first bandpass filter 12, a sine wave generator 14, a frequency mixing module 13, a low-pass filter 15, an orthogonal demodulator 16, a third circulator 24, a first optical cable 25, an optical cable connection box 26, and a second optical cable 27. 7. A distributed fiber optic acoustic wave sensing system consisting of a pigtail terminal box 28, a second photodetector 30, a low-noise amplifier 31, a second bandpass filter 32, a high-speed analog-to-digital converter 33, and an adaptive filter 34; and a three-dimensional laser scanning system consisting of a second erbium-doped fiber amplifier 17, a second circulator 18, a first photodetector 19, a ranging unit 20, an angle encoder 21, a main control unit 22, a point cloud generation module 23, a first optical cable 25, a galvanometer scanning probe 29, a high-speed analog-to-digital converter 33, and a drive module 36.
[0020] The laser light emitted by the narrow-linewidth laser 1 is input to the input port a of the first fiber coupler 2. The first fiber coupler 2 splits the laser light into two paths with a 50:50 ratio, one of which is output to the input end of the second erbium-doped fiber amplifier 17, and the other is output to the input port a of the second fiber coupler 3. The laser light is then split into two paths, one of which is input as probe light from the output port b to the input port a of the acousto-optic modulator 4, and the other is input as local light from the output port c to the input port a of the third fiber coupler 10. The output end of the arbitrary waveform generator 5 is connected to the input port b of the acousto-optic modulator 4; the output port c of the acousto-optic modulator 4 is connected to the input end of the first erbium-doped fiber amplifier 6; the output end of the first erbium-doped fiber amplifier 6 is connected to the input end of the wavelength division multiplexer 7; and the output end of the wavelength division multiplexer 7 is connected to the input port a of the third fiber coupler 8.
[0021] The third fiber coupler 8 splits the light into two paths, which are respectively emitted into port a of the third circulator 24 and port a of the first circulator 9 through its output ports b and c; port b of the third circulator 24 is connected to port a of the first optical cable 25; port b of the first circulator 9 is connected to port a of the optical cable connection box 26; port b of the optical cable connection box 26 is connected to port a of the second optical cable 27; and port b of the second optical cable 27 is connected to the pigtail terminal box 28.
[0022] The output port c of the first circulator 9 is connected to the input port a of the fourth optical fiber coupler 10; the output ports c and d of the fourth optical fiber coupler 10 are connected to the input ports a and b of the balanced photodetector 11; the output port c of the balanced photodetector 11 is connected to the input port a of the high-speed analog-to-digital converter 33; the output port d of the high-speed analog-to-digital converter 33 is connected to the input port a of the adaptive filter 34; the output port c of the adaptive filter 34 is connected to the input port of the first bandpass filter 12; the output port of the first bandpass filter 12 is connected to the input port a of the mixing module 13; the input ports b and c of the mixing module 13 are connected to the output ports a and b of the sine wave generator 14; the output port d of the mixing module 13 is connected to the input port of the low-pass filter 15; the output port of the low-pass filter 15 is connected to the input end of the orthogonal demodulator 16; and the output end of the orthogonal demodulator 16 is connected to the input port b of the host computer 35.
[0023] The output port c of the third circulator 24 is connected to the input end of the second photodetector 30; the output end of the second photodetector 30 is connected to the input end of the low-noise amplifier 31; the output end of the low-noise amplifier 31 is connected to the input end of the second band-pass filter 32; the output end of the second band-pass filter 32 is connected to the input port b of the high-speed analog-to-digital converter 33; the output port e of the high-speed analog-to-digital converter 33 is connected to the input port b of the adaptive filter 34.
[0024] The output port of the second erbium-doped fiber amplifier 17 is connected to port a of the second circulator 18; port b of the second circulator 18 is connected to port b of the first optical cable 25; port c of the first optical cable 25 is connected to port a of the galvanometer scanning probe 29; port c of the second circulator 18 is connected to the input end of the first photodetector 19; the output end of the first photodetector 19 is connected to the input port of the ranging unit 20; the output port of the ranging unit 20 is connected to the input end of the angle encoder 21; the output end of the angle encoder 21 is connected to the input port c of the high-speed analog-to-digital converter 33; the output port f of the high-speed analog-to-digital converter 33 is connected to the input end of the main control unit 22; the output end of the main control unit 22 is connected to the input end of the point cloud generation module 23; the output port of the point cloud generation module 23 is connected to the input port a of the host computer 35; the output port c of the host computer 35 is connected to the input port of the drive module 36; the output port of the drive module 36 is connected to the input port b of the galvanometer scanning probe 29.
[0025] The turnout is a line connection device that allows high-speed trains to switch from one track to another, and it is also one of the weak links in the track. Figure 2Figure 2 shows a schematic diagram of the high-speed railway turnout structure and optical cable layout, including a first optical cable 25, a second optical cable 27, a galvanometer scanning probe 29, a point rail 37, a guide rail 38, a wing rail 39, a frog core 40, and a stock rail 41. High-speed railway turnout failures primarily manifest as wear and deformation of the point rail 37, wing rail 39, and frog core 40, as well as excessive reduction of the point rail 37 or a lack of close contact with the stock rail 41. These failures can generate significant abnormal vibration signals and significant changes in geometric parameters when a train passes. Vibration detection can be achieved through the rational placement of optical fiber sensors, and scanning probes can be installed using 3D laser scanning technology to assist in monitoring the turnout's geometric parameters. Since the point rail 37 and the base rail 41 will come into contact when turning, the sensing fiber is not suitable for placement on the inner side of the base rail 41. Furthermore, since the point rail 37 is a key structure in the turnout and is susceptible to damage, the second optical cable 27 can be placed on the inner side of the point rail 37 along the guide rail 38, wing rail 39, and frog core 40. When a turnout malfunctions, the placed sensing fiber can detect different vibration signals. Simultaneously, the first optical cable 25 is placed along the outer side of the base rail 41 to arrange a galvanometer scanning probe 29 and collect background noise signals. The galvanometer scanning probe 29 is primarily placed in the area where the guide rail 38, wing rail 39, and frog core 40 are located. When abnormal vibration is detected, a command is issued to drive the galvanometer scanning probe 29 to scan the key areas of the turnout, collect its geometric parameter information for three-dimensional modeling, and compare and analyze it with standard parameters to accurately locate and identify the turnout fault. The multi-parameter real-time dynamic monitoring system for high-speed railway turnout faults of the present application restores and locates the abnormal vibration signals caused by the detected turnout faults in combination with the three-dimensional deformation information, thereby realizing real-time dynamic monitoring of high-speed railway turnout faults.
[0026] This application also proposes a multi-parameter real-time dynamic monitoring method for high-speed railway turnout faults, comprising the following steps: S1: A narrow-linewidth laser 1 emits a continuous narrow-linewidth laser with a central wavelength of 1550nm, which is input to the input port a of the first fiber coupler 2. The first fiber coupler 2 splits the laser into two parts with a 50:50 ratio. One beam of light is used for the distributed fiber acoustic wave sensing system and is injected from the output port c of the first fiber coupler 2 into the input port a of the second fiber coupler 3. The second fiber coupler 3 splits the laser into two parts with a 99:1 ratio. 99% of the laser light is output from the output port b of the second fiber coupler 3 as detection light, and 1% of the laser light is output from the output port c of the second fiber coupler 3 as local light and connected to the input port a of the fourth fiber coupler 10. The light from the output port b of the second fiber coupler 3 is output to the input port a of the acousto-optic modulator 4. The acousto-optic modulator 4 modulates the light into pulsed light and introduces a frequency shift of 80MHz, which is then output to the input end of the first erbium-doped fiber amplifier 6. The output end of the arbitrary waveform generator 5 is connected to the input port b of the acousto-optic modulator 4 to generate an electrical pulse signal to drive the acousto-optic modulator 4.
[0027] S2: The first erbium-doped fiber amplifier 6 amplifies the optical power and outputs the light to the input end of the wavelength division multiplexer 7; the wavelength division multiplexer 7 filters the input light and inputs it from the output end to the input port a of the third optical fiber coupler 8, which is divided into two output lights by the third optical fiber coupler 8, wherein the detection light signal of the output port b of the third optical fiber coupler 8 is input to the port a of the third circulator 24, and is transmitted to the port a of the first optical cable 25 through the port b of the third circulator 24. The first optical cable 25 is laid along the road outside the turnout basic track 41 to collect a reference background noise signal; the collected background noise signal returns to the third circulator 24 and passes through the third circulator 24. Port c is transmitted to the second photodetector 30 to convert the optical signal into an electrical signal; the reference signal that has completed the photoelectric conversion is preprocessed by the second photodetector 30 through a low-noise amplifier 31 and a second band-pass filter 32 to suppress high-frequency electromagnetic interference; the preprocessed reference signal is transmitted by the second band-pass filter 32 to the input port b of the high-speed analog-to-digital converter 33, and the analog signal is converted into a digital signal through the high-speed analog-to-digital converter 33, and then transmitted to the b input end of the adaptive filter 34 through the output port e of the high-speed analog-to-digital converter 33. Based on the adaptive filtering algorithm, the filter parameters are adjusted in real time to minimize the interference of noise on the signal.
[0028] S3: The detection light signal outputted from the output port c of the third optical fiber coupler 8 is injected into the port a of the first circulator 9 and transmitted to the port a of the optical cable connection box 26 through the port b of the first circulator 9. The detection light signal is injected from the port b of the optical cable connection box 26 into the port a of the second optical cable 27. The optical cable is laid along the switch rail and is used to directly detect and transmit the vibration signal on the switch rail. The port b of the second optical cable 27 is connected to the pigtail terminal box 28 to realize the end processing of the optical cable. The generated backward Rayleigh scattered light signal is returned from the second optical cable 27 to the port b of the first circulator 9 through the optical cable connection box 26. Rayleigh scattering The signal is transmitted to the input port b of the fourth fiber coupler 10 through the port c of the first circulator 9, and realizes coherent beat frequency with the reference light output from the output port c of the second fiber coupler 3 at the fourth fiber coupler 10. After the beat frequency signal is generated, it is transmitted from the output ports c and d of the fourth fiber coupler 10 to the input ports a and b of the balanced photodetector 11 respectively; the beat frequency signal is converted into an electrical signal by the balanced photodetector 11 and output through the output port c to the input port a of the high-speed analog-to-digital converter 33 to realize the conversion of the analog signal to the digital signal, which facilitates the subsequent processing of the digital signal using FPGA technology.
[0029] S4: The high-speed analog-to-digital converter 33 outputs the digital signal from the output port d to the input port a of the adaptive filter 34, and filters out the background noise through the adaptive filtering algorithm; the digital signal is transmitted to the input port of the first bandpass filter 12 through the output port c of the adaptive filter 34. The center frequency of the first bandpass filter 12 is 20MHz, which is used to filter out the noise signal; the digital signal after filtering by the first bandpass filter 12 is transmitted to the input port a of the mixing module 13, and the sine wave generator 14 outputs standard sine wave and cosine wave signals with a frequency of 20MHz and transmits them to the input ports b and c of the mixing module 13. The signals are orthogonally mixed in the mixing module 13 to generate two signals, I and Q.
[0030] The mixing module 13 inputs the I and Q signals after orthogonal mixing through the output port d to the input port of the low-pass filter 15 for low-pass filtering. The frequency range of the low-pass filter 15 is 0-50MHz, which filters out noise outside this frequency and retains only the signal with zero intermediate frequency; the filtered I and Q signals are transmitted to the input end of the orthogonal demodulator 16 by the low-pass filter 15, and the amplitude and phase of the vibration signal are obtained by the coordinate rotation digital calculation method, and phase unwrapping is performed; the demodulated amplitude signal and phase signal are transmitted to the input port b of the host computer 35, and the disturbance location is performed by the difference, cumulative average, and threshold disturbance location method to obtain the specific position of the abnormal vibration, and the phase of the disturbance signal is restored according to the disturbance location to obtain the original waveform of the disturbance signal, and its signal features are extracted and preliminary fault type identification is performed based on the support vector machine.
[0031] S5: The output port c of the host computer 35 is connected to the driving module 36. Once the host computer 35 identifies and locates abnormal vibration, it will send a control command signal to the driving module 36; the driving module 36 is connected to the input port b of the galvanometer scanning probe 29, which is used to control the galvanometer scanning probe 29 to perform three-dimensional laser scanning on the key track structure of the turnout to collect its three-dimensional geometric information; the detection laser from the output port b of the first optical fiber coupler 2 is transmitted to the input end of the second erbium-doped fiber amplifier 17 for optical power amplification; the detection light after optical power amplification is injected into the port a of the second circulator 18 and output through the port b of the second circulator 18 and transmitted to the port a of the galvanometer scanning probe 29 through the first optical cable 25, focusing the laser beam and controlling the emission angle for three-dimensional scanning to obtain three-dimensional information of the turnout point rail, and then collecting the reflected laser and focusing to filter out ambient stray light, and transmitting it back to the port b of the second circulator 18 through the first optical cable 25.
[0032] S6: The reflected laser signal is transmitted to the input end of the first photodetector 19 through the port c of the second circulator 18 and converted into an electrical signal; the electrical signal that has completed the photoelectric conversion is transmitted to the ranging unit 20, and the distance is calculated by modulating the laser phase difference; the output end of the ranging unit 20 is connected to the input end of the angle encoder 21, and the angle encoder 21 can record the deflection angle of the galvanometer in real time to provide spatial positioning information; the angle encoder 21 is connected to the input port c of the high-speed analog-to-digital converter 33, converts the analog signal into a digital signal and transmits it to the main control unit 22 through the output port f of the high-speed analog-to-digital converter 33, and the main control unit 22 processes the distance and angle data in real time to generate three-dimensional coordinates; the generated three-dimensional coordinate data is transmitted by the main control unit 22 to the point cloud generation module 23, and the single-point measurement data is integrated to generate a three-dimensional point cloud, and the point cloud is transmitted to the input port a of the host computer 35 to process the point cloud data for three-dimensional modeling and analysis of changes in the geometric parameters of the switch point rail, and alarm and fault identification are performed based on the track standard parameters.
[0033] This application is a multi-parameter collaborative monitoring solution that integrates distributed fiber-optic vibration sensing and three-dimensional laser scanning technology. Based on phase-sensitive coherent optical time-domain reflectometry (PCOTR), the system dynamically filters out background noise interference through an adaptive noise suppression circuit, improving the sensitivity and positioning accuracy of vibration signal detection. Combined with three-dimensional laser scanning technology, it performs three-dimensional modeling and data analysis on the turnout track structure, enabling visualization and classification of turnout faults. The placement of optical cable 33 on the point rail allows for precise monitoring of high-risk turnout track structures and track health monitoring in high-risk turnout areas. FPGA technology is used to accelerate the processing of laser scanning and vibration data, extract vibration features, and perform three-dimensional modeling analysis. Fault classification and identification of vibration signals is performed based on a support vector machine (SVM). Changes in the geometric parameters of key turnout components are measured and analyzed, and combined with the three-dimensional model, precise identification and visualization of turnout fault points are achieved. This solution addresses the slow response, limited transmission, and inaccurate single-parameter judgment criteria of traditional monitoring systems, enabling real-time monitoring of high-speed rail turnout vibration anomalies, precise location, and intelligent fault type identification.
[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.
Claims
1. A multi-parameter real-time dynamic monitoring system for high-speed railway turnout faults, characterized by: The invention comprises a narrow linewidth laser (1), a first optical fiber coupler (2), a second optical cable (27) arranged on the inner side of the high-speed railway switch point rail (37) and along the guide rail (38), the wing rail (39), and the frog core (40), and a first optical cable (25) and a galvanometer scanning probe (29) arranged along the outer road of the basic rail (41), wherein the first optical cable (25) and the second optical cable (27) are both connected to a distributed optical fiber acoustic wave sensing system, and abnormal vibration signals are detected based on phase-sensitive coherent optical time domain reflection technology to achieve restoration and positioning of switch fault signals, and the first optical cable (25) and the galvanometer scanning probe (29) are connected to a three-dimensional laser scanning system for detecting geometric parameters of the switch structure and accurately identifying switch faults; The three-dimensional laser scanning system comprises a second erbium-doped fiber amplifier (17), a second circulator (18), a first photodetector (19), a distance measuring unit (20), an angle encoder (21), a main control unit (22), a point cloud generation module (23), a galvanometer scanning probe (29), a high-speed analog-to-digital converter (33) and a driving module (36). The laser light emitted by the narrow linewidth laser (1) is input to the first fiber coupler (2) and is divided into two paths, one of which is output to the input end of the second erbium-doped fiber amplifier (17), and the other is output to the distributed fiber acoustic wave sensing system. The output port of the second erbium-doped fiber amplifier (17) is connected to the port a of the second circulator (18); the port b of the second circulator (18) is connected to the port b of the first optical cable (25); the port c of the first optical cable (25) is connected to the port a of the galvanometer scanning probe (29); and the output port of the second erbium-doped fiber amplifier (17) is connected to the port b of the second circulator (18); and the port c of the first optical cable (25) is connected to the port a of the galvanometer scanning probe (29). The port c of the second circulator (18) is connected to the input end of the first photodetector (19); the output end of the first photodetector (19) is connected to the input port of the distance measuring unit (20); the output port of the distance measuring unit (20) is connected to the input end of the angle encoder (21); the output end of the angle encoder (21) is connected to the input port c of the high-speed analog-to-digital converter (33); the output port f of the high-speed analog-to-digital converter (33) is connected to the input end of the main control unit (22); the output end of the main control unit (22) is connected to the input end of the point cloud generation module (23); the output port of the point cloud generation module (23) is connected to the input port a of the host computer (35); the output port c of the host computer (35) is connected to the input port of the driving module (36); the output port of the driving module (36) is connected to the input port b of the galvanometer scanning probe (29).
2. The high-speed railway turnout fault multi-parameter real-time dynamic monitoring system according to claim 1, characterized in that: The distributed optical fiber acoustic wave sensing system comprises a second optical fiber coupler (3), an acousto-optic modulator (4), an arbitrary waveform generator (5), a first erbium-doped optical fiber amplifier (6), a wavelength division multiplexer (7), a third optical fiber coupler (8), a first circulator (9), a fourth optical fiber coupler (10), a balanced photodetector (11), a first bandpass filter (12), a frequency mixing module (13), a sine wave generator (14), a low-pass filter (15), an orthogonal demodulator (16), a third circulator (24), a first optical cable (25), an optical cable connection box (26), a second optical cable (27), a pigtail terminal box (28), a second photodetector (30 ), a low noise amplifier (31), a second bandpass filter (32) and an adaptive filter (34), the laser light emitted by the narrow linewidth laser (1) is input to the first fiber coupler (2) and divided into two paths, one of which is output to the input end of the second erbium-doped fiber amplifier (17), and the other is output to the input port a of the second fiber coupler (3) and then divided into two paths, one of which is input from the output port b to the input port a of the acousto-optic modulator (4) as a detection light, and the other is input from the output port c to the input port a of the third fiber coupler (10) as a local light; the output end of the arbitrary waveform generator (5) is connected to the input port b of the acousto-optic modulator (4); The output port c of the acousto-optic modulator (4) is connected to the input end of the first erbium-doped fiber amplifier (6); the output end of the first erbium-doped fiber amplifier (6) is connected to the input end of the wavelength division multiplexer (7); the output end of the wavelength division multiplexer (7) is connected to the input port a of the third fiber coupler (8); The third optical fiber coupler (8) splits the light into two paths, and the light is respectively emitted from its output ports b and c into port a of the third circulator (24) and port a of the first circulator (9); port b of the third circulator (24) is connected to port a of the first optical cable (25); port b of the first circulator (9) is connected to port a of the optical cable connection box (26); port b of the optical cable connection box (26) is connected to port a of the second optical cable (27); port b of the second optical cable (27) is connected to the pigtail terminal box (28); The output port c of the first circulator (9) is connected to the input port a of the fourth optical fiber coupler (10); the output ports c and d of the fourth optical fiber coupler (10) are connected to the input ports a and b of the balanced photodetector (11); the output port c of the balanced photodetector (11) is connected to the input port a of the high-speed analog-to-digital converter (33); the output port d of the high-speed analog-to-digital converter (33) is connected to the input port a of the adaptive filter (34); the output port c of the adaptive filter (34) is connected to the input port of the first bandpass filter (12); the output port of the first bandpass filter (12) is connected to the input port a of the mixing module (13); the input ports b and c of the mixing module (13) are connected to the output ports a and b of the sine wave generator (14); the output port d of the mixing module (13) is connected to the input port of the low-pass filter (15); the output port of the low-pass filter (15) is connected to the input end of the orthogonal demodulator (16); the output end of the orthogonal demodulator (16) is connected to the input port b of the host computer (35); The output port c of the third circulator (24) is connected to the input end of the second photodetector (30); the output end of the second photodetector (30) is connected to the input end of the low-noise amplifier (31); the output end of the low-noise amplifier (31) is connected to the input end of the second band-pass filter (32); the output end of the second band-pass filter (32) is connected to the input port b of the high-speed analog-to-digital converter (33); and the output port e of the high-speed analog-to-digital converter (33) is connected to the input port b of the adaptive filter (34).
3. The high-speed railway turnout fault multi-parameter real-time dynamic monitoring system according to claim 2, characterized in that: The first optical fiber coupler (2) adopts an optical fiber coupler with a coupling ratio of 50:
50.
4. The high-speed railway turnout fault multi-parameter real-time dynamic monitoring system according to claim 2, characterized in that: The second optical fiber coupler (3) uses an optical fiber coupler with a coupling ratio of 99:1, and 99% of the laser light is output as detection light, and 1% of the laser light is output as local light.
5. The high-speed railway turnout fault multi-parameter real-time dynamic monitoring system according to claim 2 is characterized by: The acousto-optic modulator (4) modulates light into pulsed light and introduces a frequency shift, and then outputs the pulsed light to the input end of the first erbium-doped fiber amplifier (6); the arbitrary waveform generator (5) generates an electrical pulse signal to drive the acousto-optic modulator (4).
6. A high-speed railway turnout fault multi-parameter real-time dynamic monitoring system according to any one of claims 1 to 5, characterized in that: At least one galvanometer scanning probe (29) is provided and arranged in the area where the guide rail (38), the wing rail (39), and the frog core (40) are located.
7. A multi-parameter real-time dynamic monitoring method for high-speed railway turnout faults, characterized by: The high-speed railway turnout fault multi-parameter real-time dynamic monitoring system according to any one of claims 1 to 6 comprises the following steps: S1: The continuous narrow linewidth laser light emitted by the narrow linewidth laser (1) is input to the first fiber coupler (2); the first fiber coupler (2) divides the laser light into two parts with a ratio of 50:50, one of which is used for the distributed fiber acoustic wave sensing system and is injected into the second fiber coupler (3) by the first fiber coupler (2); the second fiber coupler (3) divides the laser light into two parts with a ratio of 99:1, 99% of the laser light is output as detection light and input to the acousto-optic modulator (4), and 1% of the laser light is output as local light and input to the fourth fiber coupler (10); the acousto-optic modulator (4) modulates the light into pulse light and introduces a frequency shift, and then outputs it to the first erbium-doped fiber amplifier (6); the arbitrary waveform generator (5) generates an electrical pulse signal to drive the acousto-optic modulator (4); S2: The signal outputted by the acousto-optic modulator (4) is inputted into the third optical fiber coupler (8) after power amplification and filtering, and is divided into two output lights by the third optical fiber coupler (8). The detection light signal of the third optical fiber coupler (8) is inputted into the port a of the third circulator (24), and is transmitted to the first optical cable (25) through the port b of the third circulator (24). The first optical cable (25) collects the reference background noise signal; the collected background noise signal returns to the third circulator (24) and is transmitted to the second photodetector (3) through the port c of the third circulator (24). 0) converting the optical signal into an electrical signal; the reference signal that has completed the photoelectric conversion is pre-processed by the second photodetector (30) through a low-noise amplifier (31) and a second band-pass filter (32) to suppress high-frequency electromagnetic interference; the pre-processed reference signal is transmitted to a high-speed analog-to-digital converter (33) by the second band-pass filter (32), and the analog signal is converted into a digital signal by the high-speed analog-to-digital converter (33) and then transmitted to an adaptive filter (34). Based on the adaptive filtering algorithm, the filter parameters are adjusted in real time to minimize the interference of noise on the signal; S3: The detection light signal outputted by the third optical fiber coupler (8) is injected into the port a of the first circulator (9) and transmitted to the optical cable connection box (26) through the port b of the first circulator (9). The detection light signal is injected into the second optical cable (27) by the optical cable connection box (26) to directly detect and transmit the vibration signal on the pointed rail; the generated backward Rayleigh scattered light signal is returned to the port b of the first circulator (9) by the second optical cable (27) through the optical cable connection box (26); the Rayleigh scattered signal is transmitted to the fourth optical fiber coupler (10) through the port c of the first circulator (9), and realizes coherent beat frequency with the reference light outputted by the second optical fiber coupler (3) at the fourth optical fiber coupler (10), and after the beat frequency signal is generated, it is transmitted to the balanced photodetector (11) by the fourth optical fiber coupler (10); the beat frequency signal is converted into an electrical signal by the balanced photodetector (11) and then output to the high-speed analog-to-digital converter (33) to realize the conversion from analog signal to digital signal; S4: The high-speed analog-to-digital converter (33) outputs the digital signal to the adaptive filter (34), and the background noise is filtered out by the adaptive filtering algorithm; the digital signal is transmitted to the first band-pass filter (12) through the adaptive filter (34) to filter out the noise signal; the digital signal filtered by the first band-pass filter (12) is transmitted to the mixing module (13), the sine wave generator (14) outputs the standard sine wave and cosine wave signals and transmits them to the mixing module (13), and the signals are orthogonally mixed in the mixing module (13) to generate two signals, I and Q; The mixing module (13) inputs the I and Q signals after orthogonal mixing into the low-pass filter (15) for low-pass filtering; the filtered I and Q signals are transmitted from the low-pass filter (15) to the orthogonal demodulator (16), and the amplitude and phase of the vibration signal are obtained by using the coordinate rotation digital calculation method, and the phase is unwound; the demodulated amplitude signal and phase signal are transmitted to the host computer (35) for preliminary fault classification processing; S5: The host computer (35) is connected to the drive module (36), and once the host computer (35) identifies and locates the abnormal vibration, it will send a control command signal to the drive module (36); the drive module (36) is connected to the galvanometer scanning probe (29), which is used to control the galvanometer scanning probe (29) to perform three-dimensional laser scanning on the key track structure of the turnout to collect its three-dimensional geometric information; the detection laser output by the first optical fiber coupler (2) is transmitted to the second erbium-doped fiber amplifier (17) for optical power amplification; the detection light after optical power amplification is injected into the port a of the second circulator (18) and output through the port b of the second circulator (18), and is transmitted to the galvanometer scanning probe (29) through the first optical cable (25), focusing the laser beam and controlling the emission angle for three-dimensional scanning to obtain three-dimensional information of the turnout point rail, and then collecting the reflected laser and focusing to filter out the ambient stray light, and transmitting it back to the port b of the second circulator (18) through the first optical cable (25); S6: The reflected laser signal is transmitted to the first photodetector (19) through the port c of the second circulator (18) to convert it into an electrical signal; the electrical signal after the photoelectric conversion is transmitted to the distance measuring unit (20), and the distance is calculated by modulating the laser phase difference; the output end of the distance measuring unit (20) is connected to the input end of the angle encoder (21), and the angle encoder (21) records the deflection angle of the galvanometer in real time to provide spatial positioning information; the angle encoder (21) is connected to the high-speed analog-to-digital converter (33) to convert the analog signal into a digital signal and transmit it to the main control unit (22) through the high-speed analog-to-digital converter (33), and the main control unit (22) processes the distance and angle data in real time to generate three-dimensional coordinates; the generated three-dimensional coordinate data is transmitted by the main control unit (22) to the point cloud generation module (23), and the single-point measurement data is integrated to generate a three-dimensional point cloud, which is then transmitted to the host computer (35) to process the point cloud data for three-dimensional modeling and analysis of the geometric parameter changes of the switch point rail, and to perform alarms and fault identification based on the track standard parameters.
8. The multi-parameter real-time dynamic monitoring method for high-speed railway turnout faults according to claim 7, characterized in that: In step S4, the upper computer (35) performs disturbance location by using the difference, cumulative average, and threshold disturbance location methods to obtain the specific location of the abnormal vibration, and performs phase restoration on the disturbance signal according to the disturbance location to obtain the original waveform of the disturbance signal, extracts its signal features, and performs preliminary fault type identification based on the support vector machine.
9. The multi-parameter real-time dynamic monitoring method for high-speed railway turnout faults according to claim 7, characterized in that: The frequency range of the low-pass filter (15) is 0-50 MHz, which filters out noise outside this frequency and only retains the signal with zero intermediate frequency.