CN117508262A - A rail irregularity detection system and its detection method - Google Patents

Abstract

The invention provides a steel rail irregularity detection system and a detection method thereof, wherein a transmitting end of the steel rail irregularity detection system comprises an external cavity type tunable laser, a beam splitter I, a delay optical fiber, a coupler I, a circulator, a coated optical fiber, a photodetector, a red light laser, a coupler II and a collimator, a receiving end comprises an imaging plate, an optical filter and an infrared camera, a signal processing end comprises a data acquisition card and a computer, a frequency modulation continuous wave is utilized to carry out distance measurement, and meanwhile, the infrared camera is utilized to track the laser spot position on the imaging plate, so that the distance measurement precision is high, non-cooperative target measurement can be realized, and the anti-interference capability is strong; the infrared camera is used for positioning the position of the measuring light spot, so that the measuring repeatability is high, and the influence of environmental factors such as sunlight and the like is avoided; and solving a measured value of the irregularity of the height of the steel rail and a measured value of the irregularity of the track direction by combining the distance information of the three selected measuring positions on the steel rail and the spot pixel coordinates, thereby realizing the detection of the irregularity of the steel rail.

Description

A rail irregularity detection system and its detection method

Technical field

The invention relates to the technical field of rail detection, in particular to a rail irregularity detection system and a detection method thereof.

Background technique

High-speed railways have become an important guarantee for my country's economic and social development because of their advantages such as comfort, safety, convenience and speed, strong carrying capacity, and low energy consumption. As the running speed of high-speed rail trains gradually increases, the rail smoothness and precision requirements are also getting higher and higher. At present, a common measurement method in China is to use a track inspection vehicle combined with a track control network to accurately measure the track geometry. However, this method has high measurement costs and long measurement period and is difficult to be used for lines under construction. The chord measurement method is also a traditional rail irregularity detection method. It uses a small track geometry detector that is smaller and easier to control. It detects rail irregularities by "pushing the small to the big". However, this method only relies on the track gauge. Wheel contact measurement can easily cause line deviation, and a small positive vector deviation will cause a large curve radius error. Therefore, there is an urgent need for a rail irregularity detection system and detection method that is convenient, efficient, simple in structure, and has accurate measurement results.

Contents of the invention

The technical problem to be solved by the present invention is to provide a rail irregularity detection system.

Another technical problem to be solved by the present invention is to provide a rail irregularity detection method using the above detection system.

In order to solve the above technical problems, the technical solution of the present invention is:

A rail irregularity detection system includes a transmitting end (A), a receiving end (B) and a signal processing end (C), where,

The transmitting end (A) includes an external cavity tunable laser (1), a beam splitter I (2), a delay fiber (3), a coupler I (4), a circulator (5), and a coated optical fiber (6) , photodetector (7), red laser (8), coupler II (9) and collimator (10), the coated optical fiber (6) is coated with a semi-transparent and semi-reflective film, the external cavity type The tunable laser (1), beam splitter I (2), coupler I (4), circulator (5) and photodetector (7) are connected in series through optical fibers in sequence. At the same time, the beam splitter I (2), delay The optical fiber (3) and the coupler I (4) are connected in series through the optical fiber, the circulator (5) and the coupler II (9) are connected through the coated optical fiber (6), the red laser (8) and the collimator The coupler (10) is respectively connected to the coupler II (9) through optical fibers;

Specifically, the external cavity tunable laser (1) emits a frequency modulated continuous wave after being modulated by a triangular wave, and the laser passes through a Mach path composed of a beam splitter I (2), a delay fiber (3) and a coupler I (4). After the Zehnder interference optical path enters the circulator (5), the frequency-modulated laser passes through the coated optical fiber (6) coated with a semi-transparent and semi-reflective film and then connects with the indicator emitted by the red laser (8) at the coupler II (9). The visible red light is coupled and emitted to the imaging plate (11) by the collimator (10). After the return light signal carrying the target distance information is coupled with the local oscillator light through the coated optical fiber (6), the coupled light passes through the ring The device (5) interferes at the light detector (7) to generate a distance measurement beat signal;

The receiving end (B) includes an imaging plate (11), a filter (12) and an infrared camera (13). The collimator (10) is arranged in front of the imaging plate (11). The imaging plate (11) ) is placed in front of the lens of the infrared camera (13), and a filter (12) is fixed on the lens of the infrared camera (13);

Specifically, on the one hand, the imaging plate (11) serves as a target to reflect the measurement light emitted by the emission end (A), and on the other hand, images the spot of the measurement light; in front of the infrared camera The filter (12) is added to prevent stray light in the environment from affecting the measurement process; the combination of the imaging plate (11) and the infrared camera (13) enables tracking of the measurement light spot in a larger field of view;

The signal processing terminal (C) includes a data acquisition card (14) and a computer (15). The light detector (7) is connected to the data acquisition card (14) via lines. The data acquisition card (14) and an infrared camera (13) are respectively connected to the computer (15) lines;

Specifically, the data acquisition card (14) records the distance measurement beat signal at the photodetector (7) and transmits it to the computer ( 15); The computer (15) combines the distance information of the receiving end at different measurement points and the measurement spot position information to complete the measurement of the track direction information and height information of the rail to confirm whether there is track direction irregularity or height unevenness in the rail.

Preferably, in the above rail irregularity detection system, the modulation range of the external cavity tunable laser (1) is set to 1545-1555nm, and the basic modulation rate is set to 100.08nm/s.

A rail irregularity detection method using the above detection system, the specific steps are as follows:

(1) Take the first end of the rail to be measured as the measurement point S, the end of the rail to be measured as the measurement point E, and the midpoint of the rail to be measured as the measurement point M, outside the rail to be measured and close to point S Install the collimator (10), use frequency modulated continuous wave as the measurement light source, and build the frequency modulated continuous wave measurement optical path;

(2) Install the receiving end at point S, and the recording data processing end records the measured beat frequency signal at the photodetector and transmits it to the signal processing end. Use computer software to design a band-pass filter for software filtering to filter out the measurement signal and Auxiliary signal, determine the resampling point according to the peak and valley points of the auxiliary signal, resample the measurement signal at equal optical frequency intervals, determine the peak point of the spectrum of the resampled signal, and calculate the distance to be measured;

(3) Set the threshold to 80-130 to preprocess the images taken by the infrared camera, clear the gray values of pixels whose gray values are lower than the threshold, and eliminate the influence of ambient stray light. Perform Gaussian filtering to further eliminate the influence of Gaussian noise in the environment, and determine the pixel coordinates of the measurement spot in the camera coordinate system based on the gray value of each pixel;

(4) Move the receiving end to point E and point M respectively, repeat steps (2) and (3), and obtain the distance measurement value and light spot position measurement value of point E and point M;

(5) Based on the measurement results of points S, E and M, the lateral and longitudinal deviations of the measurement spot in the camera coordinate system caused by rail irregularities can be determined, and then the height and rail unevenness values of the rails can be calculated. Complete the detection of rail irregularities.

Preferably, according to the above-mentioned rail irregularity detection method, the expression of the measurement beat frequency signal I _b (t) measured by the light detector is:

I _b (t)＝2A ₀ ·{2cos[2π·(α(t)τ _m t+f ₀ τ _m )]+2cos[2π·(α(t)τ _r t+f ₀ τ _r )]

+cos[2π·(α(t)(τ _m +τ _r )t+f ₀ (τ _m +τ _r ))]+cos[2π·(α(t)(τ _m -τ _r )t+f ₀ (τ _m -τ _r ))]}

Among them, A ₀ represents the amplitude of the frequency modulated light source, f ₀ is the initial frequency of laser modulation, α(t) represents the real-time modulation rate of the frequency modulated light source containing nonlinearity, t represents time, and τ _m is the light emission caused by the optical path of the distance to be measured. The time delay between the optical fiber and the return light, τ _r is the time delay generated by the calibrated length of the delay fiber.

Preferably, in the above rail irregularity detection method, the pixel coordinates (x, y) of the measurement spot in the camera coordinate system are expressed as follows:

The resolution of the infrared camera is N×M, and a _ij represents the gray value of the pixel in the i-th row and j-th column of the camera after preprocessing and Gaussian filtering.

Preferably, the above rail irregularity detection method is based on the measurement results of the three selected measurement points, namely the distance measurement results R _S , RE _E , _RM and the coordinate measurement results (x _S , y _S ), (x _E , y _E ), (x _M , y _M ), the lateral and longitudinal deviations of the measurement spot in the camera coordinate system caused by rail irregularities can be determined:

Preferably, in the above-mentioned rail irregularity detection method, the measured rail irregularity value f _H and rail unevenness value f _L are expressed as follows:

f _L =β _L Δx+γ _L Δy

f _H =β _H Δx+γ _H Δy

Among them, β _L , β _H , γ _L , and γ _H are the conversion coefficients for converting pixel coordinates to world coordinates, which can be obtained by calibrating the laser tracker. Δx and Δy are the lateral deviations of the measurement spot in the camera coordinate system due to uneven rails. and longitudinal deviation.

Beneficial effects:

The above-mentioned rail irregularity detection system uses frequency modulated continuous waves for distance measurement, and at the same time uses an infrared camera to track the position of the laser spot on the imaging plate. It has high ranging accuracy, can achieve non-cooperative target measurement, and has strong anti-interference ability; using an infrared camera Positioning the measurement spot position, the measurement repeatability is high, and it is not affected by environmental factors such as sunlight; combining the distance information of the three selected measurement positions on the rail and the spot pixel coordinates, the rail unevenness measurement value and rail can be calculated The measured value of directional irregularity can be used to detect rail irregularities, especially the convenient, efficient and accurate detection of directional irregularities and unevenness of a single rail. This method does not require complex data processing and has high measurement accuracy. High, the measurement system has a simple structure, easy adjustment, efficient measurement process, and accurate measurement results.

Description of drawings

Figure 1 is a schematic diagram of the overall device of the rail irregularity detection system in the present invention;

Figure 2 is an installation diagram of the system in the process of detecting rail irregularities using infrared cameras and frequency modulated continuous waves in the present invention;

Figure 3 is a schematic diagram of the lateral deviation and longitudinal deviation of the measurement spot in the camera coordinate system caused by the rail irregularity during the detection of rail irregularities using an infrared camera and frequency modulated continuous wave in the present invention.

In the picture: 1-External cavity tunable laser 2-Beam splitter I 3-Delay fiber 4-Coupler I

5-Circulator 6-Fiber 7-Photodetector 8-Red laser 9-Coupler II

10-Collimator 11-Imaging plate 12-Optical filter 13-Infrared camera

14-Data acquisition card 15-Computer 16-Rail A-transmitter B-receiver

C-Signal processing terminal

Detailed ways

The rail irregularity detection system and its detection method according to the present invention will be described below with reference to the embodiments and drawings.

Example 1

As shown in Figure 1, a rail irregularity detection system includes a transmitting end A, a receiving end B and a signal processing end C. The transmitting end A includes an external cavity tunable laser 1, a beam splitter I2, a delay Optical fiber 3, coupler I4, circulator 5, coated optical fiber 6, photodetector 7, red laser 8, coupler II 9 and collimator 10. The coated optical fiber 6 is coated with a semi-transparent and semi-reflective film. The external cavity tunable laser 1, beam splitter I2, coupler I4, circulator 5 and photodetector 7 are connected in series through optical fibers. At the same time, beam splitter I2, delay fiber 3 and coupler I4 are connected in series through optical fibers, so The circulator 5 and the coupler II 9 are connected through a coated optical fiber 6, and the red laser 8 and the collimator 10 are respectively connected to the coupler II 9 through optical fibers; the receiving end B includes an imaging plate 11, a filter 12 and Infrared camera 13, the collimator 10 is arranged in front of the imaging plate 11, the imaging plate 11 is placed in front of the lens of the infrared camera 13, and the filter 12 is fixed on the lens of the infrared camera 13; the signal processing Terminal C includes a data acquisition card 14 and a computer 15. The light detector 7 is connected to the data acquisition card 14 via a circuit. The data acquisition card 14 and the infrared camera 13 are respectively connected to the computer 15 via a circuit.

The external cavity tunable laser 1 in the transmitting end A emits a frequency modulated continuous wave after being modulated by a triangular wave. The laser enters the circulator 5 after passing through the Mach-Zehnder interference optical path composed of the beam splitter I2, the delay fiber 3 and the coupler I4. The frequency-modulated laser passes through the coated optical fiber 6 coated with a semi-transparent and semi-reflective film and is coupled with the visible red light for indication emitted by the red laser 8 at the coupler II 9 and is emitted to the imaging plate 11 by the collimator 10 . After the return light signal carrying the target distance information is coupled with the local oscillator light through the coated optical fiber, the coupled light passes through the circulator and interferes at the photodetector 7 to generate a distance measurement beat frequency signal.

The receiving end B is composed of an imaging plate 11 , a filter 12 and an infrared camera 13 . On the one hand, the imaging plate 11 serves as a target to reflect the measurement light emitted from the emission end A, and on the other hand, it will image the spot of the measurement light. The filter 12 is added in front of the infrared camera to prevent stray light in the environment from affecting the measurement process. The combination of the imaging plate 11 and the infrared camera 13 enables tracking of the measurement light spot in a larger field of view.

The PicoScope 6000E data acquisition card 14 in the signal processing terminal C records the distance measurement beat signal at the photodetector 7 and transmits it to the computer 15 together with the imaging plate spot image captured by the infrared camera 13 . The computer 15 combines the distance information of the receiving end at different positions and the measurement spot position information to complete the measurement of the track irregularity and height irregularity of the rail.

Specifically, the specific model of the external cavity tunable laser 1 in the rail irregularity detection system is the LUNAPHOENIX 1400 frequency modulated continuous wave laser. The frequency modulated continuous wave modulation range is set to 1545-1555nm, and the basic modulation rate is set to 100.08nm/s. The specific model of the infrared camera is CMLN-13S2M-CS from Gray Point Company, with a resolution of 1294×964 and responds to light in the 1550nm band.

Taking the frequency modulation stage of the frequency modulated light source as an example, the instantaneous electric field _Em (t) of the frequency modulated laser emitted by the external cavity tunable laser 1 can be expressed as:

where A ₀ represents the amplitude of the FM light source, f ₀ is the initial frequency of laser modulation, α(t) represents the real-time modulation rate of the FM light source containing nonlinearity, t represents time, Represents the initial phase of the FM continuous wave.

The distance measurement signal at the photodetector 7 is a beat frequency signal generated by the coherence of four optical signals containing different distance information. After ignoring the light intensity loss caused by the optical fiber and filtering out the DC component, the beat frequency signal I _b (t) It can be expressed as follows:

I _b (t)＝2A ₀ ·{2cos[2π·(α(t)τ _m t+f ₀ τ _m )]+2cos[2π·(α(t)τ _r t+f ₀ τ _r )]

+cos[2π·(α(t)(τ _m +τ _r )t+f ₀ (τ _m +τ _r ))]+cos[2π·(α(t)(τ _m -τ _r )t+f ₀ (τ _m -τ _r ))]}

Among them, A ₀ represents the amplitude of the frequency modulated light source, f ₀ is the initial frequency of laser modulation, α(t) represents the real-time modulation rate of the frequency modulated light source containing nonlinearity, t represents time, and τ _m is the light emission caused by the optical path of the distance to be measured. The time delay between the optical fiber and the return light, τ _r is the time delay generated by the calibrated length of the delay fiber.

Although the modulation rate of the FM light source has been set before the measurement begins, the modulation rate still has serious nonlinearity due to the influence of environmental changes and mechanical vibration, which will cause the spectrum of the measurement signal to be seriously broadened and make it difficult to calculate the distance information to be measured. , the measurement error is large. Use Matlab to design a band-pass filter for software filtering, and filter out the measurement signal I _m (t) and the auxiliary signal I _r (t) as follows:

I _m (t)＝4A ₀ cos[2π·(α(t)τ _m t+f ₀ τ _m )]

I _r (t)＝4A ₀ cos[2π·(α(t)τ _r t+f ₀ τ _r )]

Among them, A ₀ represents the amplitude of the frequency modulated light source, f ₀ is the initial frequency of laser modulation, α(t) represents the real-time modulation rate of the frequency modulated light source containing nonlinearity, t represents time, and τ _m is the light emission caused by the optical path of the distance to be measured. The time delay between the optical fiber and the return light, τ _r is the time delay generated by the calibrated length of the delay fiber.

The auxiliary signal is used as the resampling clock signal, and the extreme point of the auxiliary signal is used as the resampling point. At this time, the corresponding optical frequencies between each sampling point are equally spaced, and the sampling frequency F _s that triggers resampling is :

F _s =2πα(t)τ _r

α(t) is the instantaneous modulation rate of the FM light source containing nonlinearity, and τ _r is the time delay generated by the calibrated length of the delay fiber.

The measurement signal _Im (t) is resampled at equal optical frequency intervals using the resampling point, and the peak frequency f of the spectrum of the resampled measurement signal is:

Among them, τ _m is the time delay between light emission and return light caused by the optical path of the distance to be measured, and τ _r is the time delay caused by the calibrated length of the delay fiber.

At this time, the distance to be measured R can be expressed as:

where f is the peak frequency of the resampled measurement signal spectrum, τ _r is the time delay generated by the calibrated delay fiber, c represents the speed of light, and n represents the refractive index of air.

The determination of the measurement spot position is based on the gray value of each pixel recorded after the infrared camera shoots the imaging plate. A dynamic threshold is set according to the ambient light intensity to artificially eliminate the interference of stray light, and Gaussian filtering is used to eliminate the influence of Gaussian noise in the environment. , the pixel coordinates (x, y) of the measurement spot in the camera coordinate system are expressed as follows:

The resolution of the infrared camera is N×M, and a _ij represents the gray value of the pixel in the i-th row and j-th column of the camera.

Take the first end of the rail 16 to be measured as the measurement point S, the end of the rail 16 to be measured as the measurement point E, and the midpoint of the rail 16 to be measured as the measurement point M. Record the frequency modulated continuous wave distance measurement results and the infrared camera spot positioning position as follows :R _S , RE _E , R _M , (x _S ,y _S ), (x _E ,y _E ), (x _M ,y _M ). Then the lateral deviation Δx and longitudinal deviation Δy of the measurement spot in the camera coordinate system due to the unevenness of the rail are expressed as follows:

Furthermore, the height unevenness value f _H and rail unevenness value f _L of the measured rail in this section are expressed as follows:

f _L =β _L Δx+γ _L Δy

f _H =β _H Δx+γ _H Δy

Among them, β _L , β _H , γ _L , and γ _H are the conversion coefficients for converting pixel coordinates to world coordinates, which can be obtained by calibrating the laser tracker. Δx and Δy are the lateral deviations of the measurement spot in the camera coordinate system due to uneven rails. and longitudinal deviation.

This invention uses frequency-modulated continuous waves for distance measurement, and uses an infrared camera to track the position of the laser spot on the imaging plate to achieve convenient, efficient and accurate detection of rail irregularities, that is, track direction and height. The specific design process includes:

Step 1: Figure 2 shows the installation diagram of the system in the process of detecting rail irregularities using infrared cameras and frequency modulated continuous waves. The first end of the rail to be tested 16 is used as the measurement point S, and the end of the rail to be tested is used as the measurement point E. Take the midpoint of the rail to be measured as the measurement point M. Install the collimator 10 outside the rail to be measured and close to the side of point S, use frequency modulated continuous wave as the measurement light source, the laser wavelength modulation method is sawtooth wave modulation, the modulation range is set to 1545-1555nm, and the basic modulation rate is set to 100.08 nm/s, build a frequency modulated continuous wave measurement optical path.

Step 2: First, install the receiving end B at point S. The data processing end records the measured beat frequency signal at the light detector and transmits it to the data acquisition card. The measured beat frequency signal I _bS (t) is:

I _bS (t)＝2A ₀ ·{2cos[2π·(α(t)τ _m t+f ₀ τ _m )]+2cos[2π·(α(t)τ _r t+f ₀ τ _r )]

+cos[2π·(α(t)(τ _m +τ _r )t+f ₀ (τ _m +τ _r ))]+cos[2π·(α(t)(τ _m -τ _r )t+f ₀ (τ _m -τ _r ))]}

Among them, A ₀ represents the amplitude of the frequency modulated light source, f ₀ is the initial frequency of laser modulation, α(t) represents the real-time modulation rate of the frequency modulated light source containing nonlinearity, t represents time, and τ _m is the light emission caused by the optical path of the distance to be measured. The time delay between the optical fiber and the return light, τ _r is the time delay generated by the calibrated length of the delay fiber.

Use Matlab to design a band-pass filter for software filtering, and filter out the measurement signal I _mS (t) and auxiliary signal I _rS (t) as follows:

I _mS (t)＝4A ₀ cos[2π·(α(t)τ _m t+f ₀ τ _m )]

I _rS (t)＝4A ₀ cos[2π·(α(t)τ _r t+f ₀ τ _r )]

Among them, A ₀ represents the amplitude of the frequency modulated light source, f ₀ is the initial frequency of laser modulation, α(t) represents the real-time modulation rate of the frequency modulated light source containing nonlinearity, t represents time, and τ _m is the light emission caused by the optical path of the distance to be measured. The time delay between the optical fiber and the return light, τ _r is the time delay generated by the calibrated length of the delay fiber.

The peak and valley points of the auxiliary signal are extracted as resampling points and the measurement signal I _mS (t) is resampled at equal optical frequency intervals. The resulting peak frequency f _S of the resampled measurement signal spectrum is:

Among them, τ _m is the time delay between light emission and return light caused by the optical path of the distance to be measured, and τ _r is the time delay caused by the calibrated length of the delay fiber.

At this time, the distance to be measured R _S can be expressed as:

where f is the peak frequency of the resampled measurement signal spectrum, τ _r is the time delay generated by the calibrated delay fiber, c represents the speed of light, and n represents the refractive index of air.

Repeat the measurement five times, and use the average of the five measurement results as the final distance measurement value of point S.

Step 3: Set an appropriate threshold to preprocess the images taken by the infrared camera, and clear the gray values of pixels whose gray values are lower than the threshold to eliminate the influence of ambient stray light. Gaussian filtering is performed on the preprocessed image to further eliminate the influence of Gaussian noise in the environment. The pixel coordinates (x _S , y _S ) of the light spot in the camera coordinate system are measured according to the gray value of each pixel and are expressed as follows:

Repeat the measurement five times, and use the average of the five measurement results as the final position of the light spot measured by the receiving end at point S.

Step 4: Move the receiving end to point E and point M respectively, repeat steps 2 and 3, and obtain the distance measurement value and light spot position measurement value of point B and point C:

Step 5: Figure 3 is a schematic diagram of the lateral deviation and longitudinal deviation of the measurement spot in the camera coordinate system due to the uneven rail. According to the measurement results of points S, E and M, it can be determined that the measurement spot in the camera coordinate system is caused by the uneven rail. The lateral deviation Δx and the longitudinal deviation Δy under the system:

In addition, the height unevenness value f _H and rail unevenness value f _L of the measured rail in this section are expressed as follows:

f _L =β _L Δx+γ _L Δy

f _H =β _H Δx+γ _H Δy

Among them, β _L , β _H , γ _L , and γ _H are constants, representing the conversion coefficients from pixel coordinates to world coordinates, which can be obtained by calibrating the laser interferometer. Δx and Δy are the measurement spots in the camera coordinate system due to uneven rails. lateral deviation and longitudinal deviation.

In summary, by applying the rail irregularity detection system of the present invention and combining the distance information of the three measurement positions and the light spot pixel coordinate information, the rail irregularity can be detected, and the rail irregularity measurement value and the rail irregularity can be completed. The calculation of unevenness measurement values does not require complex data processing, the measurement system is simple, the measurement process is convenient, the measurement results are less affected by environmental factors, and the measurement accuracy is high.

The above are only preferred embodiments of the present invention. It should be noted that those skilled in the art can make several improvements and modifications without departing from the principles of the present invention. These improvements and modifications can also be made. should be regarded as the protection scope of the present invention.

Claims (7)

1. A rail irregularity detecting system is characterized in that: comprises a transmitting end (A), a receiving end (B) and a signal processing end (C), wherein,

the emitting end (A) comprises an external cavity type tunable laser (1), a beam splitter I (2), a delay optical fiber (3), a coupler I (4), a circulator (5), a coated optical fiber (6), a light detector (7), a red light laser (8), a coupler II (9) and a collimator (10), wherein a semi-transparent semi-reflective film is plated on the coated optical fiber (6), the external cavity type tunable laser (1), the beam splitter I (2), the coupler I (4), the circulator (5) and the light detector (7) are sequentially connected in series through the optical fibers, meanwhile, the beam splitter I (2), the delay optical fiber (3) and the coupler I (4) are connected in series through the optical fibers, the circulator (5) and the coupler II (9) are communicated through the coated optical fiber (6), and the red light laser (8) and the collimator (10) are respectively connected with the coupler II (9) through the optical fibers;

the receiving end (B) comprises an imaging plate (11), an optical filter (12) and an infrared camera (13), the collimator (10) is arranged in front of the imaging plate (11), the imaging plate (11) is arranged in front of a lens of the infrared camera (13), and the optical filter (12) is fixed on the lens of the infrared camera (13);

the signal processing end (C) comprises a data acquisition card (14) and a computer (15), the optical detector (7) is connected with the data acquisition card (14) in a circuit manner, and the data acquisition card (14) and the infrared camera (13) are respectively connected with the computer (15) in a circuit manner.

2. The rail irregularity detecting system of claim 1, wherein: the modulation range of the external cavity type tunable laser (1) is 1545-1555nm, and the basic modulation rate is 100.08nm/s.

3. A method for detecting rail irregularities using the detection system of claim 1, characterized by: the method comprises the following specific steps:

(1) Taking the head end of a steel rail to be measured as a measurement S point, the tail end of the steel rail to be measured as a measurement E point, the middle point of the steel rail to be measured as a measurement M point, installing a collimator (10) on one side, which is outside the steel rail to be measured and is close to the S point, of the measured steel rail, and constructing a frequency modulation continuous wave measurement light path by taking a frequency modulation continuous wave as a measurement light source;

(2) Installing a receiving end at the S point, recording a measurement beat frequency signal at the light detector by a data processing end, transmitting the measurement beat frequency signal to a signal processing end, designing a band-pass filter by computer software for software filtering, respectively filtering out a measurement signal and an auxiliary signal, determining a resampling point according to the peak-to-valley point of the auxiliary signal, resampling the measurement signal at equal optical frequency intervals, determining the spectrum peak point of the resampling signal, and solving the distance to be measured;

(3) Preprocessing a picture shot by an infrared camera by setting a threshold value to be 80-130, resetting the gray value of a pixel point with a gray value lower than the threshold value, eliminating the influence of ambient stray light, performing Gaussian filtering processing on the preprocessed image, further eliminating the influence of Gaussian noise in the environment, and determining the pixel coordinates of a measuring facula under a camera coordinate system according to the gray value of each pixel;

(4) Respectively moving the receiving end to the E point and the M point, and repeating the step (2) and the step (3) to obtain a distance measurement value and a light spot position measurement value of the E point and the M point;

(5) According to S, E and the measurement results of the M points, the transverse deviation and the longitudinal deviation of the measurement light spots under the camera coordinate system due to the height irregularity of the steel rail can be determined, so that the height irregularity value and the track direction irregularity value of the steel rail can be calculated, and the detection of the rail irregularity is completed.

4. A rail irregularity detecting method according to claim 3, characterized in that: the expression I of the measured beat frequency signal measured by the photodetector _b (t) is:

I _b (t)＝2A ₀ ·{2cos[2π·(α(t)τ _m t+f ₀ τ _m )]+2cos[2π·(α(t)τ _r t+f ₀ τ _r )]+cos[2π·(α(t)(τ _m +τ _r )t+f ₀ (τ _m +τ _r ))]+cos[2π·(α(t)(τ _m -τ _r )t+f ₀ (τ _m -τ _r ))]}

wherein A is ₀ Representing the amplitude of the FM light source, f ₀ Is the initial frequency of laser modulation, alpha (t) represents the real-time modulation rate of the frequency modulation light source containing nonlinear quantity, t represents time, tau _m For the time delay between light-out and light-back caused by the optical path of the distance to be measured, τ _r The time delay generated for the delay fiber after the length has been calibrated.

5. A rail irregularity detecting method according to claim 3, characterized in that: the pixel coordinates (x, y) of the measurement spot in the camera coordinate system are expressed as follows:

wherein the resolution of the infrared camera is N×M, a _ij Representing the gray value of the pixel in the ith row and jth column of the camera after preprocessing and gaussian filtering.

6. A rail irregularity detecting method according to claim 3, characterized in that: based on the measurement results of the three selected measurement points, i.e. the distance measurement result R _S 、R _E 、R _M And coordinate measurement result (x _S ,y _S )、(x _E ,y _E )、(x _M ,y _M ) The transverse deviation and the longitudinal deviation of the measuring light spot under the camera coordinate system caused by the rail irregularity can be determined:

7. a rail irregularity detecting method according to claim 3, characterized in that: measuring the irregularity value f of the rail _H Sum track irregularity value f _L The expression is as follows:

f _L ＝β _L Δx+γ _L Δy

f _H ＝β _H Δx+γ _H Δy

wherein beta is _L 、β _H 、γ _L 、γ _H The conversion coefficient for converting the pixel coordinate into the world coordinate can be obtained by calibrating a laser tracker, and the deltax and deltay are the transverse deviation and the longitudinal deviation of the measuring light spot under a camera coordinate system due to the irregularity of the steel rail.