CN111452830A - An imaging method and device for realizing automatic detection of track slab cracks - Google Patents

Abstract

The invention discloses an imaging method and device for realizing automatic detection of track slab cracks. The imaging method refers to firstly using a camera to detect surface defects of track slab cracks, and then using an ultrasonic transmitter and receiver to excite and transmit through an ultrasonic linear array probe. The ultrasonic signal with a frequency in the range of 1-5MHz is sent to the inside of the track plate. The ultrasonic linear array probe adopts the mode of self-sending and self-receiving to obtain the diffuse field signal in the track plate and transmits the diffuse field signal to the computer. Finally, the computer uses MATLAB software to analyze the signal. The received signal passively extracts the Green's function, performs cross-correlation on the diffuse field signal, reconstructs the Green's function between the array elements, obtains the undelayed response between the array elements, recovers the early defect information submerged by noise, and then according to the frequency Orbital plate crack imaging using domain synthetic aperture focusing imaging algorithm. The invention can efficiently, non-destructively and real-time detect the crack surface of the ballastless track slab and the defects inside the crack.

Description

An imaging method and device for realizing automatic detection of track slab cracks

technical field

The invention relates to an imaging method and device for realizing automatic detection of track slab cracks, belonging to the technical field of track defect detection.

Background technique

High-speed railways are of great significance for shortening urban distances and promoting coordinated regional development, and are an indispensable part of modern transportation methods. With the rapid development of high-speed railway, ballastless track, as a main track structure, is composed of ballastless track slab, CA mortar layer, support layer and foundation bed, and its application range is getting wider and wider. However, in recent years, more and more diseases have appeared in the structure of high-speed rail lines, including the under-line structure relief, through cracks, and voids in the CA mortar layer. The main reason is that, on the one hand, during the high-speed and heavy-load operation of the train, the ballastless track will be squeezed, impacted, etc., which may cause incompactness, cracks or voids inside, and the formation of damage layers or honeycomb hemp layers on the outside. On the other hand, the ballastless track may have problems in the pre-production due to construction technology and construction experience, resulting in its own defects; in addition, natural disasters such as rain and snow erosion and changes in ambient temperature will also lead to defect. The existence of defects will seriously affect the bearing capacity and durability of the ballastless track, which will lead to the failure of the ballastless track structure. It is precisely the important prerequisite for ensuring the rapid and safe operation of high-speed rail, which will be directly related to the normal operation of the train and the personal safety of passengers.

However, at present, the detection of ballastless track defects in my country mainly relies on manual static detection technology. Since the effective skylight time for rail transit to be used for line maintenance and maintenance is only 2-3 hours, and the thread of high-speed railway is very long, if the existing Some detection methods not only consume a lot of manpower and material resources, but also have very low efficiency and high detection and maintenance costs, so that they cannot meet the needs of rail safety early warning.

SUMMARY OF THE INVENTION

In view of the above-mentioned problems existing in the prior art, the purpose of the present invention is to provide an imaging method and device for realizing automatic detection of track slab cracks, so as to realize efficient, non-destructive and real-time detection of ballastless track slab crack defects, and for the safe operation of high-speed railways Provide timely warning and strong guarantee.

To achieve the above object, the present invention adopts the following technical solutions:

An imaging method for realizing automatic detection of track slab cracks, comprising the following steps:

a) Using a camera to detect surface defects of track slab cracks, specifically: using a camera to shoot the surface of the track slab, and transfer the captured image to the computer, and the computer compares it with the pre-stored normal track slab image to judge Whether there are cracks on the surface of the track plate;

b) Use the ultrasonic transmitter and receiver to excite and transmit ultrasonic signals with a frequency in the range of 1 to 5MHz to the inside of the track plate through the ultrasonic linear array probe. Transmit the diffused field signal to the computer, specifically: when the computer determines that there is a crack, adjust the position of the ultrasonic linear array probe to make it sink and place it at the crack, and then start the ultrasonic transmitter and receiver to transmit and receive the ultrasonic wave. The transmitting end of the ultrasonic sensor transmits ultrasonic signals with a frequency in the range of 1 to 5 MHz, and is transmitted to the inside of the track plate by the transmitting end of the ultrasonic linear array probe, and then the diffused field signal is collected by the receiving end of the ultrasonic linear array probe and transmitted to the ultrasonic wave. The receiving end of the transmitter and receiver, and then the received end of the ultrasonic transmitter receiver transmits the received diffuse field signal to the computer;

c) The computer uses MATLAB software to passively extract the Green's function from the received signal, cross-correlate the diffuse field signal, reconstruct the Green's function between the array elements, obtain the undelayed response between the array elements, and recover from being submerged by noise. The early defect information is obtained, and then the track plate crack imaging is carried out according to the frequency domain synthetic aperture focusing imaging algorithm.

In one embodiment, the step c) specifically comprises the following operations:

1) Data processing:

First of all, due to the interaction between the ultrasonic signal and the defect target during the propagation of the ultrasonic signal inside the ballastless track slab, an approximately uniform sound field will be formed after scattering and multiple reflections, which is the diffuse field. As shown in step b), the diffused field signal in the ballastless track can be obtained through the ultrasonic linear array probe, and any two receiving array elements in the ultrasonic linear array probe are set to be γ ₁ and γ ₂ respectively, and the two receiving elements are If the array elements γ ₁ and γ ₂ are in a closed surface space, then the integral of all sound sources of the sound field cross-spectrum at the receiving array elements γ ₁ and γ ₂ is equal to the frequency domain causal Green function and non-causal Green function between γ ₁ and γ ₂ Difference:

G(r ₁ , r ₂ ω)-G ^* (r ₁ ,r ₂ ,ω)=-2jω∫G(r ₁ ,r,ω)G ^* (r ₂ ,r,ω)dV(1):

In formula (1), the left side of the equation is the difference between the causal and non-causal functions of the frequency domain Green function between the two receiving elements γ ₁ and γ ₂ , and the right side of the equation represents the receiving elements γ ₁ and γ Integral of all sound sources in the cross-correlation of sound fields at ₂ places, specifically, G(γ ₁ ,γ ₂ ,ω) represents the causal Green function between γ ₁ and γ ₂ in the frequency domain, the superscript * represents the conjugate complex number, G ^* (γ ₁ ,γ ₂ ,ω) represents the non-causal Green's function between γ ₁ and γ ₂ in the frequency domain, j represents the imaginary unit, j ² =-1, ω represents the angular frequency of the transmitted signal, γ is in the diffusion The position of any point in the closed space V of field integration represents any defect target and can be regarded as a noise source, G(γ ₁ ,γ,ω) represents the frequency domain Green’s function propagation formula between γ and γ ₁ , G(γ ₂ , γ, ω) represents the frequency domain Green's function propagation formula between γ and γ ₂ , dV represents the differential of the closed space V;

Then, the power spectral density |q(ω)| ² is also independent of the position of the noise source, assuming that the intensity of the diffuse-field noise source is spatially uniformly distributed and uncorrelated with each other:

(q(r ₁ , ω)q ^* (r ₂ , ω)>=δ(r ₁ -r ₂ )|q(ω)| ² (2);

Equation (2) represents the expected value of the cross-power spectrum of the signals at γ ₁ and γ ₂ in the frequency domain equal to the power spectral density of the noise source between γ ₁ and γ _2. Specifically, in formula (2), <q(γ ₁ , ω )q ^* (γ ₂ , ω)> represents the expected value of the cross-power spectrum of the signals at γ ₁ and γ ₂ in the frequency domain, q(γ ₁ , ω) represents the field strength at γ ₁ , q(γ ₂ , ω) represents the The field strength at γ ₂ , <> represents the expected value of the statistical average, the superscript * represents the conjugate complex number, δ(γ1-γ2) represents the cross-power spectrum of the signal between γ ₁ and γ ₂ is independent of the position of the noise source; The cross-correlation between the sound fields at the positions of the two receiving array elements γ ₁ and γ ₂ in the frequency domain is:

<p(r ₁ , ω)p ^* (r ₂ , ω)>=|q(ω)| ² ∫G(r ₁ , r, ω)G ^* (r ₂ , r, ω)dV (3);

In formula (3), p(γ ₁ , ω) represents the sound field at the receiving element γ ₁ , p(γ ₂ , ω) represents the sound field at the receiving element γ ₂ , <p(γ ₁ , ω)p ^* (γ ₂ , ω)> represents the cross-correlation function between the sound fields at the positions of the two receiving array elements γ ₁ and γ ₂ , and the superscript * represents a complex conjugate number;

Then, from formulas (1) and (3), we get:

(G(r ₁ , r ₂ , ω)-G ^* (r ₁ , r ₂ , ω))|q(ω)| ² =-2jω<p(r ₁ , ω)p ^* (r ₂ , ω) >(4);

In formula (4), the left side of the equation is the Green's function G(γ ₁ , γ ₂ , ω) and its corresponding time reversal, that is, the conjugate operation in the corresponding frequency domain. Multiply these two terms by the random noise. After the power spectral density, the result is equal to the cross-correlation result of the two receiving array elements γ ₁ and γ ₂ in the diffuse field on the right side of the equation, and the superscript * represents the conjugate complex number;

Then, the time domain expression corresponding to formula (4) is obtained. The frequency domain 2jω corresponds to the time domain 2d/dt. According to the convolution theorem, the product in the frequency domain corresponds to the convolution in the time domain, thus obtaining:

代表互相关运算，C_q(t)表示扩散场中的噪声q(t)的自相关结果，d/dt代表对t求导，p(γ₁,t)代表时域中接收阵元γ₁处的声场，p(γ₂,t)代表时域中接收阵元γ₂处的声场；In formula (5), G(γ ₁ , γ ₂ , t) represents the Green's function between γ ₁ and γ ₂ in the time domain, and G(γ ₁ , γ ₂ , -t) represents G(γ ₁ , γ ₂ , t) time reversal, which corresponds to the conjugate operation in the frequency domain, * represents the convolution operation,

represents the cross-correlation operation, C _q (t) represents the autocorrelation result of the noise q(t) in the diffuse field, d/dt represents the derivation of t, p(γ ₁ ,t) represents the receiving array element γ ₁ in the time domain The sound field at , p(γ ₂ ,t) represents the sound field at the receiving array element γ ₂ in the time domain;

The results of formulas (4) and (5) show that the sound fields p(γ ₁ ,t) and p(γ ₂ ,t) of the receiving array elements in the diffuse field are correlated and derived, and the result is equal to the difference between the two receiving array elements. Green's function response, according to the principle of reciprocity of sound waves, on the time axis, the Green's function response between the two receiving array elements has symmetry;

2) Frequency Domain Synthetic Aperture Focusing (SAFT) imaging:

First of all, since the propagation of ultrasonic waves in an infinite homogeneous medium satisfies the Helmholtz wave equation, it is assumed that the coordinates of the ultrasonic transmitting array element are (u, 0) and the coordinates of the receiving array element are (v, 0 under the xz coordinate axis of the two-dimensional space). ), the distances of the defect target from the transmitting array element and the receiving array element are ρ _in and ρ _out respectively, and the time domain receiving signal corresponding to any pair of transmitting and receiving array elements is e(v, u, t), then the corresponding transmitting and receiving The frequency response E(v, u, ω) corresponding to the array element is:

E(v, u, ω)=P(ω)∫∫f(x, z)g(ρ _tn , ω)g(ρ _out , ω)dxdz (6):

P(ω)是发射信号的频谱，ω为发射信号的角频率；f(x，z)代表缺陷目标的点扩散函数，g(ρ_in，ω)代表发射阵元到缺陷目标的空间格林函数频域响应，g(ρ_out，ω)代表接收阵元到缺陷目标的空间格林函数频域响应；In formula (6),

P(ω) is the spectrum of the transmitted signal, ω is the angular frequency of the transmitted signal; f(x, z) represents the point spread function of the defective target, and g(ρ _in , ω) represents the spatial Green's function from the transmitting array element to the defective target Frequency domain response, g(ρ _out , ω) represents the spatial Green's function frequency domain response from the receiving array element to the defect target;

Also, the Green's function is:

In formula (7), g(x, z, ω) represents the Green’s function of any point under the xz coordinate axis of the two-dimensional space, the wave number k=ω/c, c is the longitudinal wave velocity of the sound wave in the medium, and the spatial variables x and z Corresponds to the wave numbers k _x and k _z ;

Use the variables x and z to replace the variables ρ _in and ρ _out , that is, g(x, z, ω) instead of g(ρ _in , ρ _out , ω), bring the formula (7) into the formula (6), and regain the reception Response E(v, u, ω), and then use the wave numbers _ku and k _v to represent the spatial information of the transmitting array element (u, 0) and the receiving array element (v, 0), respectively. In order to realize the transformation of the integral into a two-dimensional Fourier Leaf transform, the Fourier transform of the variables u and v in the received response E(v, u, ω) is obtained:

In formula (8), E(ω, ku , K _v ) represents the spatial Fourier transform of the frequency domain response E(v, _u , ω), and F represents the spatial Fourier transform of the point spread function f(x, z) leaf transform;

Correspondingly, the spatial Fourier transform of the point spread function f(x, z) is F(k _x , k _z ). In order to obtain the image domain, the inverse transform model is:

代表缺陷目标等间隔采样的数据域模型，S^-1{}代表Stolt映射，完成了从数据域到图像域的转换，p^*(ω)代表发射信号的共轭频谱；In formula (9),

represents the data domain model of the defect target sampled at equal intervals, S ^-1 {} represents the Stolt map, which completes the conversion from the data domain to the image domain, p ^* (ω) represents the conjugate spectrum of the transmitted signal;

The nonlinear coordinate change of formula (9) is performed to map the wavenumbers ku, _kv and _k to the wavenumbers _kx and _kz associated with the defect target in the image domain, then:

k _x = k _u + k _v (10):

Then the image of the defect target distribution is obtained by the two-dimensional inverse Fourier transform

代表缺陷目标分布的图像域模型。In formula (12),

An image domain model representing the distribution of defect targets.

An imaging device for realizing automatic detection of track slab cracks includes a track inspection trolley, a camera and a computer arranged on the track inspection trolley, the camera is connected with a computer signal, and also includes an ultrasonic transmitter receiver and an ultrasonic linear array probe, The transmitting end of the ultrasonic linear array probe is signally connected to the transmitting interface of the ultrasonic transmitting receiver, the receiving end of the ultrasonic linear array probe is signally connecting to the receiving interface of the ultrasonic transmitting and receiving device, and the receiving interface of the ultrasonic transmitting and receiver is connected to the receiving interface of the ultrasonic transmitting receiver. computer signal connection, the ultrasonic linear array probe is connected with an adaptive telescopic mechanism, the adaptive telescopic mechanism is connected with a biaxial displacement adjustment mechanism, and the biaxial displacement adjustment mechanism is fixedly connected to the front end of the rail inspection trolley .

In one embodiment, the self-adaptive telescopic mechanism includes a mounting plate A and a mounting plate B that are arranged vertically and parallel to each other, and a horizontally telescopic telescopic bracket is connected between the mounting plate A and the mounting plate B. The ultrasonic linear array probe is connected with the mounting plate A, and the biaxial displacement adjusting mechanism is connected with the mounting plate B.

In a preferred solution, a probe connector is provided on the side of the mounting plate A away from the mounting plate B, and the ultrasonic linear array probe is fixedly installed on the probe connector.

In a preferred solution, a laser rangefinder is provided on at least one side of the ultrasonic linear array probe.

In one embodiment, the biaxial displacement adjustment mechanism includes an installation plate C, a longitudinal displacement adjustment mechanism and a lateral displacement adjustment mechanism, the adaptive telescopic mechanism is fixedly connected with the installation plate C, and the installation plate C is connected with the longitudinal displacement adjustment mechanism. Sliding connection, the longitudinal displacement adjusting mechanism is slidably connected with the lateral displacement adjusting mechanism, and the lateral displacement adjusting mechanism is fixedly connected to the front end of the rail inspection trolley.

In a preferred solution, the longitudinal displacement adjustment mechanism includes a longitudinal bracket and a longitudinal electric screw adjustment mechanism, the lateral displacement adjustment mechanism includes a transverse bracket and a transverse electric screw adjustment mechanism, and the mounting plate C is slidably connected to the longitudinal bracket up and down. , the longitudinal support and the transverse support are connected in a lateral sliding manner, and the transverse support is fixedly connected to the front end of the rail inspection trolley.

In a preferred solution, a lighting lamp is also provided at the front end of the rail inspection trolley.

In a preferred solution, a guide mechanism is also provided at the front end of the rail inspection trolley, and the guide wheels constituting the guide mechanism are connected with the rail in a rolling manner.

In a preferred solution, an alarm module is also provided on the rail inspection trolley, and the alarm module is signal-connected to the computer.

In a preferred solution, a mobile power supply is also provided on the rail inspection trolley.

Compared with the prior art, the present invention has the following beneficial technical effects:

1) In the present invention, the crack defects on the surface of the track plate can be photographed and detected by a camera first, and then the internal defects of the crack can be further detected by the ultrasonic transmitter receiver and the ultrasonic linear array probe, so that the automatic detection of the track plate crack can be realized;

2) The ultrasonic linear array probe used in the present invention has the advantages of high energy and low attenuation, the ultrasonic energy hardly leaks, and is better incident on the inside of the material to be tested and interacts with the defects, forming a scattering signal favorable for detection, And the array elements of each probe can be excited individually or in combination, which can realize focusing and deflection functions, and can better collect diffuse field signals;

3) When the present invention processes and image the signal, the computer first uses the principle of reconstructing the Green's function through the cross-correlation operation between the diffused field signals of the receiving array elements, obtains the un-delayed response between the array elements, and extracts the signal submerged by the noise. Early defect scattering information, and then using frequency-domain synthetic aperture focusing imaging technology can not only ensure the ability to have high resolution, but also effectively reduce the computing time of imaging, with high speed and high precision;

To sum up, the present invention can realize efficient, non-destructive, accurate and real-time detection of ballastless track crack surface and crack internal defects, can provide timely early warning and strong guarantee for the safe operation of high-speed rail, and can provide strong support for subsequent track maintenance work; Therefore, compared with the prior art, the present invention has significant progress and application value.

Description of drawings

1 is a schematic structural diagram of an imaging method and device for automatic detection of track plate cracks provided in an embodiment of the present invention;

FIG. 2 is a state diagram of the imaging device provided in the embodiment of the present invention when it is used for detection;

3 is a schematic structural diagram of an adaptive telescopic mechanism provided in an embodiment of the present invention;

The symbols in the figure are as follows: 1. Track inspection trolley; 2. Camera; 3. Ultrasonic transmitter and receiver; 4. Ultrasonic linear array probe; 5. Self-adaptive telescopic mechanism; 51. Mounting plate A; 53. Telescopic bracket; 54. Probe connector; 55. Laser distance meter; 6. Biaxial displacement adjustment mechanism; 61. Mounting plate C; 62. Longitudinal displacement adjustment mechanism; 621. Longitudinal support; Adjustment mechanism; 63, lateral displacement adjustment mechanism; 631, lateral support; 632, lateral electric screw adjustment mechanism; 7, track plate; 8, lighting lamp; 9, guide mechanism; 91, guide wheel; 10, steel rail; 11, Alarm module; 12, mobile power supply; 13, seat; 14, computer.

Detailed ways

The technical solutions of the present invention will be described in further detail below with reference to the accompanying drawings and embodiments.

Example

1 to 3 : an imaging device for realizing automatic detection of track slab cracks provided by the present invention includes a track inspection trolley 1 and a camera 2 and a computer 14 arranged on the track inspection trolley 1. The camera 2 is signal-connected with the computer 14, and also includes an ultrasonic transmitter receiver 3 and an ultrasonic linear array probe 4, the transmitting end of the ultrasonic linear array probe 4 is signal-connected with the transmitter interface of the ultrasonic transmitter receiver 3, and the ultrasonic linear array probe 4 The receiving end of the ultrasonic transmitter and receiver 3 is signal-connected with the receiving interface of the ultrasonic transmitter receiver 3, the receiver interface of the ultrasonic transmitter receiver 3 is signal-connected with the computer 14, and the ultrasonic linear array probe 4 is connected with an adaptive telescopic mechanism 5. The adaptive telescopic mechanism 5 is connected with a biaxial displacement adjusting mechanism 6 , and the biaxial displacement adjusting mechanism 6 is fixedly connected to the front end of the rail inspection trolley 1 .

In this embodiment, the ultrasonic linear array probe 4 may be a commercially available product. For example, a contact-type ultrasonic linear array probe can be used, which has the advantages of high energy and low attenuation, and the ultrasonic energy hardly leaks. Incident to the interior of the material to be tested and interact with defects to form scattering signals favorable for detection, the ultrasonic linear array probe 4 is composed of several ultrasonic probes, for example, it can be composed of 24 ultrasonic probes in 8 rows and 3 columns. Array probe, correspondingly, the ultrasonic linear array probe 4 has 24 array elements, the center frequency of the ultrasonic probe can be 1.0MHz, and the array elements of each probe can be excited individually or in combination, which can realize focusing and deflection function. In addition, it can be seen from the above that the ultrasonic linear array probe 4 in this embodiment has a transmitting end and a receiving end, and has the function of self-transmitting and self-receiving. Due to the different propagation speed of ultrasonic waves in media with different densities, when it passes through the interface of two different media, reflection and scattering will occur. Because the ballastless track is a concrete component, it is composed of a mixture of sand, cement, and stones. Porous heterogeneous composite materials, the ultrasonic pulse wave emitted by the ultrasonic transmitter and receiver 3 will be diffracted when it encounters defects in the concrete, and will be scattered and reflected at the defect interface, resulting in a significant amplitude of the sound wave energy when it reaches the receiving probe. In this application, using the ultrasonic linear array probe 4 can not only collect the diffuse field signals more efficiently, but also superimpose the diffuse field signals received by the ultrasonic linear array probe 4, so that the radiated energy of the reflected diffuse field signals in a certain direction is the largest, The total radiation energy in other directions is smaller, so as to realize the focusing effect on the diffused field signal, so that the useful diffused field signal is enhanced, and then the interference signal is suppressed, thereby ensuring the accuracy of signal acquisition and detection results.

Referring to FIG. 1 to FIG. 3 , in this embodiment, the adaptive telescopic mechanism 5 includes a mounting plate A51 and a mounting plate B52 that are vertically arranged and parallel to each other, and the mounting plate A51 and the mounting plate B52 are connected to each other. There is a horizontally retractable telescopic bracket 53, the ultrasonic linear array probe 4 is connected with the mounting plate A51, the biaxial displacement adjusting mechanism 6 is connected with the mounting plate B52, and the adaptive telescopic mechanism 5 is connected with the biaxial displacement adjusting mechanism 6 In combination, the three-axis displacement adjustment of the ultrasonic linear array probe 4 can be realized, so that the position of the ultrasonic linear array probe 4 can be flexibly adjusted according to requirements, and then the detection range of the ultrasonic linear array probe 4 in each scan can be flexibly adjusted. The telescopic support 53 can be composed of a plurality of telescopic rods (not marked in the figure) that are cross-connected. The telescopic support 53 is connected with a telescopic drive mechanism (not shown), and the telescopic drive mechanism controls the telescopic support 53. The entire mechanism and the self-adaptive telescopic mechanism 5 belong to the prior art, and will not be repeated here.

In addition, the side of the mounting plate A51 away from the mounting plate B52 is provided with a probe connector 54, and the ultrasonic linear array probe 4 is fixedly installed on the probe connector 54. When in use, the ultrasonic linear array probe 4 is fixedly installed on the On the probe connector 54, the transmitting and receiving end faces of the ultrasonic linear array probe 4 are always kept perpendicular to the surface of the ballastless track.

In addition, a laser range finder 55 is provided on at least one side of the ultrasonic linear array probe 4, and the displacement detected by the laser range finder 55 can be used to determine the distance between the transmitting and receiving end face of the ultrasonic linear array probe 4 and the ballastless track plate 7. The vertical distance between the surfaces to keep the vertical distance within the preset range, thereby preventing the probe from being damaged due to the excessive settlement distance. In this embodiment, the left and right sides of the ultrasonic linear array probe are respectively provided with laser rangefinders 55 symmetrically, and the two laser rangefinders 55 can be set as one for use and one for backup.

1 and 2, the biaxial displacement adjusting mechanism 6 includes a mounting plate C61, a longitudinal displacement adjusting mechanism 62 and a lateral displacement adjusting mechanism 63, and the adaptive telescopic mechanism 5 is fixedly connected to the mounting plate C61 (specifically , the mounting plate B52 in the self-adaptive telescopic mechanism 5 is fixedly connected with the mounting plate C61, so that the self-adaptive telescopic mechanism 5 is connected with the biaxial displacement adjusting mechanism 6), and the mounting plate C61 is slidably connected with the longitudinal displacement adjusting mechanism 62, The longitudinal displacement adjusting mechanism 62 is slidably connected with the lateral displacement adjusting mechanism 63 , and the lateral displacement adjusting mechanism 63 is fixedly connected to the front end of the rail inspection trolley 1 .

Specifically, the longitudinal displacement adjustment mechanism 62 includes a longitudinal bracket 621 and a longitudinal electric screw adjustment mechanism 622, the lateral displacement adjustment mechanism 63 includes a lateral bracket 631 and a transverse electric screw adjustment mechanism 632, and the mounting plate C61 is connected to the longitudinal The bracket 621 is slidably connected up and down, and the longitudinal bracket 621 is slidably connected to the transverse bracket 631 , and the transverse bracket 631 is fixedly connected to the front end of the rail inspection trolley 1 . In this embodiment, the longitudinal electric screw adjustment mechanism 622 and the transverse electric screw adjustment mechanism 632 are both composed of a drive motor, a screw with one end fixed at the output end of the drive motor, and a slider connector threadedly connected to the screw. Since this composition is a known technology, it is not shown in detail in the figure.

In addition, as shown in FIG. 1 and FIG. 2 , the front end of the on-track inspection trolley 1 is also provided with a light 8 to avoid the problem of blurred and unrecognized cracks when the camera 2 is photographed at night or in a dark environment, avoiding missed inspections and increasing detection efficiency. The number of the lighting lamps 8 can be set flexibly according to the needs. For example, in this embodiment, two lighting lamps 8 are symmetrically arranged at the front end of the rail inspection trolley 1.

In addition, a guide mechanism 9 is also provided at the front end of the rail inspection trolley 1 , and the guide wheels 91 constituting the guide mechanism 9 are rollingly connected with the rail 10 to guide the running direction of the rail inspection trolley 1 .

In addition, an alarm module 11 is also provided on the rail inspection trolley 1, and the alarm module 11 is connected with the computer 14 by signal, so that when a crack is found on the track, the computer 14 can issue an instruction to drive the alarm module 11 to give an alarm. The alarm module 11 can be a commercially available alarm device.

In addition, a mobile power supply 12 is also provided on the rail inspection trolley 1 to implement mobile power supply for the device.

In addition, a seat 13 is provided on the rail inspection trolley 1 so that the inspector can sit on the rail inspection trolley 1 to perform manual signal collection and detection operations.

The imaging method for realizing automatic detection of track slab cracks using the imaging device of the present invention is as follows:

a) Make the track inspection trolley 1 run along the direction of the track plate 7. When driving, the camera 2 takes a picture of the surface of the track plate 7, and transmits the photographed picture to the computer 14, and the computer 14 matches it with the pre-stored normal track plate picture Compare to judge whether there are cracks on the surface of the track plate 7;

b) When the judgment result shows that there is a crack, the controller controls the alarm module 11 to give an alarm, the rail inspection trolley 1 stops running, adjusts the position of the ultrasonic linear array probe 4 to make it sink and press at the crack, and then starts the ultrasonic transmission and reception 3, so that the transmitting end of the ultrasonic transmitter and receiver 3 transmits ultrasonic signals with a frequency in the range of 1 to 5 MHz, and is transmitted from the transmitting end of the ultrasonic linear array probe 4 to the inside of the track plate 7, and the ultrasonic signal propagates inside the track plate 7. In the medium propagation, scattering and reflection occur at the defect interface to form a diffuse field signal, and then the diffuse field signal is collected by the receiving end of the ultrasonic linear array probe 4 and transmitted to the receiving end of the ultrasonic transmitter receiver 3, and then the ultrasonic transmitter receiver 3. The receiving end of the device transmits the received diffuse field signal to the computer 14; namely:

c) The computer uses MATLAB software to passively extract the Green's function from the received signal, cross-correlate the diffuse field signal, reconstruct the Green's function between the array elements, obtain the undelayed response between the array elements, and recover from being submerged by noise. Then, the track plate crack imaging is performed according to the frequency domain synthetic aperture focusing imaging algorithm, which is as follows:

1) Data processing:

First of all, due to the interaction between the ultrasonic signal and the defect target during the propagation of the ultrasonic signal inside the ballastless track slab, an approximately uniform sound field will be formed after scattering and multiple reflections, which is the diffuse field. As shown in step b), the diffused field signal in the ballastless track can be obtained through the ultrasonic linear array probe, and any two receiving array elements in the ultrasonic linear array probe are set to be γ ₁ and γ ₂ respectively, and the two receiving elements are If the array elements γ ₁ and γ ₂ are in a closed surface space, then the integral of all sound sources of the sound field cross-spectrum at the receiving array elements γ ₁ and γ ₂ is equal to the frequency domain causal Green function and non-causal Green function between γ ₁ and γ ₂ Difference:

G(r ₁ , r ₂ , ω)-G ^* (r ₁ , r ₂ , ω)=-2jω∫G(r ₁ , r, ω)G ^* (r ₂ , r, ω)dV (1);

In formula (1), the left side of the equation is the difference between the causal and non-causal functions of the frequency domain Green function between the two receiving elements γ ₁ and γ ₂ , and the right side of the equation represents the receiving elements γ ₁ and γ Integral of all sound sources for cross-correlation of sound fields at ₂ places, specifically, G(γ ₁ , γ ₂ , ω) represents the causal Green function between γ ₁ and γ ₂ in the frequency domain, the superscript * represents the conjugate complex number, G ^* (γ ₁ , γ ₂ , ω) represents the non-causal Green's function between γ ₁ and γ ₂ in the frequency domain, j represents the imaginary unit, j ² =-1, ω represents the angular frequency of the transmitted signal, γ is in the diffusion The position of any point in the closed space V of field integration represents any defect target and can be regarded as a noise source, G(γ ₁ , γ, ω) represents the frequency domain Green’s function propagation formula between γ and γ ₁ , G(γ ₂ , γ, ω) represents the frequency domain Green's function propagation formula between γ and γ ₂ , dV represents the differential of the closed space V;

Then, the power spectral density |q(ω)| ² is also independent of the position of the noise source, assuming that the intensity of the diffuse-field noise source is spatially uniformly distributed and uncorrelated with each other:

<q(r ₁ , ω)q ^* (r ₂ , ω)>=δ(r ₁ -r ₂ )|q(ω)| ² (2);

Equation (2) represents the expected value of the cross-power spectrum of the signals at γ ₁ and γ ₂ in the frequency domain equal to the power spectral density of the noise source between γ ₁ and γ _2. Specifically, in formula (2), <q(γ ₁ , ω )q ^* (γ ₂ , ω)> represents the expected value of the cross-power spectrum of the signals at γ ₁ and γ ₂ in the frequency domain, q(γ ₁ , ω) represents the field strength at γ ₁ , and q(γ ₂ , ω) represents the The field strength at γ ₂ , <> represents the expected value of the statistical average, the superscript * represents the conjugate complex number, δ(γ1-γ2) represents the cross-power spectrum of the signal between γ ₁ and γ ₂ is independent of the position of the noise source; The cross-correlation between the sound fields at the positions of the two receiving array elements γ ₁ and γ ₂ in the frequency domain is:

<p(r ₁ , ω)p ^* (r ₂ , ω)>=|q(ω)| ² ∫G(r ₁ , r, ω)G ^* (r ₂ , r, ω)dV (3);

In formula (3), p(γ ₁ , ω) represents the sound field at the receiving element γ ₁ , p(γ ₂ , ω) represents the sound field at the receiving element γ ₂ , <p(γ ₁ , ω)p ^* (γ ₂ , ω)> represents the cross-correlation function between the sound fields at the positions of the two receiving array elements γ ₁ and γ ₂ , and the superscript * represents a complex conjugate number;

Then, from formulas (1) and (3), we get:

(G(r ₁ , r ₂ , ω)-G ^* (r ₁ , r ₂ , ω))|q(ω)| ² =-2jω<p(r ₁ , ω)p ^* (r ₂ , ω) > (4):

In formula (4), the left side of the equation is the Green's function G(γ ₁ , γ ₂ , ω) and its corresponding time reversal, that is, the conjugate operation in the corresponding frequency domain. Multiply these two terms by the random noise. After the power spectral density, the result is equal to the cross-correlation result of the two receiving array elements γ ₁ and γ ₂ in the diffuse field on the right side of the equation, and the superscript * represents the conjugate complex number;

Then, the time domain expression corresponding to formula (4) is obtained. The frequency domain 2jω corresponds to the time domain 2d/dt. According to the convolution theorem, the product in the frequency domain corresponds to the convolution in the time domain, thus obtaining:

代表互相关运算，C_q(t)表示扩散场中的噪声q(t)的自相关结果，d/dt代表对t求导，p(γ₁,t)代表时域中接收阵元γ₁处的声场，p(γ₂,t)代表时域中接收阵元γ₂处的声场；In formula (5), G(γ ₁ ,γ ₂ ,t) represents the Green’s function between γ ₁ and γ ₂ in the time domain, and G(γ ₁ ,γ ₂ ,-t) represents G(γ ₁ ,γ ₂ , t) time reversal, which corresponds to the conjugate operation in the frequency domain, * represents the convolution operation,

represents the cross-correlation operation, C _q (t) represents the autocorrelation result of the noise q(t) in the diffuse field, d/dt represents the derivation of t, p(γ ₁ ,t) represents the receiving array element γ ₁ in the time domain The sound field at , p(γ ₂ ,t) represents the sound field at the receiving array element γ ₂ in the time domain;

The results of formulas (4) and (5) show that the sound fields p(γ ₁ ,t) and p(γ ₂ ,t) of the receiving array elements in the diffuse field are correlated and derived, and the result is equal to the difference between the two receiving array elements. Green's function response, according to the principle of reciprocity of sound waves, on the time axis, the Green's function response between the two receiving array elements has symmetry;

2) Frequency Domain Synthetic Aperture Focusing (SAFT) imaging:

The frequency-domain synthetic aperture focusing uses the phase delay method. The Green's function is decomposed by Fourier transform on the basis of the wave equation inversion theory, so it is also called the wavenumber algorithm. After that, the coordinate transformation is realized by using Stolt Mapping and interpolation. After filtering and post-processing Then the two-dimensional inverse Fourier transform is performed to finally generate the detection image in space-time. Therefore, the idea of explosion mode can be used for the defect target, and a defect target can be discretized into independent scattering points, which are regarded as secondary sound sources. , the total scattered sound field size is the superposition of the sound fields of each independent scattering point, and then the specific position and size of the defect can be reversed according to the received signal of each array element; the propagation path (sound path) of spontaneous and self-receiving is exactly equal to the array 2 times the distance from the element to the defect target;

First of all, since the propagation of ultrasonic waves in an infinite homogeneous medium satisfies the Helmholtz wave equation, it is assumed that the coordinates of the ultrasonic transmitting array element are (u, 0) and the coordinates of the receiving array element are (v, 0 under the xz coordinate axis of the two-dimensional space). ), the distances of the defect target from the transmitting array element and the receiving array element are ρ _in and ρ _out respectively, and the time domain receiving signal corresponding to any pair of transmitting and receiving array elements is e(v, u, t), then the corresponding transmitting and receiving The frequency response E(v, u, ω) corresponding to the array element is:

E(v, u, ω)=P(ω)∫∫f(x, z)g(ρ _in ,ω)g(ρ _out ,ω)dxdz (6);

P(ω)是发射信号的频谱，ω为发射信号的角频率；f(x，z)代表缺陷目标的点扩散函数，g(ρ_in，ω)代表发射阵元到缺陷目标的空间格林函数频域响应，g(ρ_out，ω)代表接收阵元到缺陷目标的空间格林函数频域响应；In formula (6),

P(ω) is the spectrum of the transmitted signal, ω is the angular frequency of the transmitted signal; f(x, z) represents the point spread function of the defective target, and g(ρ _in , ω) represents the spatial Green's function from the transmitting array element to the defective target Frequency domain response, g(ρ _out , ω) represents the spatial Green's function frequency domain response from the receiving array element to the defect target;

Also, the Green's function is:

In formula (7), g(x, z, ω) represents the Green’s function of any point under the xz coordinate axis of the two-dimensional space, the wave number k=ω/c, c is the longitudinal wave velocity of the sound wave in the medium, and the spatial variables x and z Corresponds to the wave numbers k _x and k _z ;

Use the variables x and z to replace the variables ρ _in and ρ _out , that is, g(x, z, ω) instead of g(ρ _in , ρ _out , ω), bring the formula (7) into the formula (6), and regain the reception Response E(v, u, ω), and then use the wave numbers _ku and k _v to represent the spatial information of the transmitting array element (u, 0) and the receiving array element (v, 0), respectively. In order to realize the transformation of the integral into a two-dimensional Fourier Leaf transform, the Fourier transform of the variables u and v in the received response E(v, u, ω) is obtained:

In formula (8), E(ω,k _u ,k _v ) represents the spatial Fourier transform of the frequency domain response E(v, u, ω), and F represents the spatial Fourier transform of the point spread function f(x, z) leaf transform;

Correspondingly, the spatial Fourier transform of the point spread function f(x, z) is F(k _x , k _z ). In order to obtain the image domain, the inverse transform model is:

代表缺陷目标等间隔采样的数据域模型，S^-1{}代表Stolt映射，完成了从数据域到图像域的转换，p^*(ω)代表发射信号的共轭频谱；In formula (9),

represents the data domain model of the defect target sampled at equal intervals, S ^-1 {} represents the Stolt map, which completes the conversion from the data domain to the image domain, p ^* (ω) represents the conjugate spectrum of the transmitted signal;

The nonlinear coordinate change of formula (9) is performed to map the wave numbers _ku , k _v and k to the wave numbers k _x and k _z associated with the defect target in the image domain, then:

k _x =k _u +k _v (10);

Then the image of the defect target distribution is obtained by the two-dimensional inverse Fourier transform

代表缺陷目标分布的图像域模型。In formula (12),

An image domain model representing the distribution of defect targets.

To sum up, the present invention first uses a camera to detect the surface defects of the track plate cracks, and then uses the ultrasonic transmitter receiver to transmit ultrasonic signals with a frequency in the range of 1 to 5MHz to the inside of the track plate. The mode scans the cracks of the track slab at equal intervals, and uses the full-matrix capture function of the ultrasonic linear array probe to obtain the diffuse field signal in the track slab and transmits the diffuse field signal to the computer. Finally, the diffuse field signal is processed by the computer MATLAB software and based on the The frequency domain synthetic aperture focusing imaging algorithm is used to image the track slab cracks, which can intuitively obtain the surface defects of the track slab cracks and the internal defects of the cracks; High sensitivity, high resolution, high speed, high efficiency, and it is not easily disturbed by external conditions and has high precision, which makes subsequent non-destructive testing have the advantages of fast detection speed, high efficiency and high detection accuracy, and can achieve efficient, non-destructive, accurate and real-time detection. The detection of ballastless track cracks can provide timely early warning and strong guarantee for the safe operation of high-speed rail, and can provide strong support for subsequent track maintenance work; therefore, compared with the prior art, the present invention has significant progress and application value.

Finally, it is necessary to point out here: the above is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited to this, and any person skilled in the art is within the technical scope disclosed by the present invention. Any changes or substitutions that can be easily thought of within the scope of the present invention should be covered within the protection scope of the present invention.

Claims (10)

1. an imaging method that realizes automatic detection of track plate cracks, is characterized in that, comprises the steps: a) Use a camera to detect surface defects of track slab cracks; b) Use the ultrasonic transmitter and receiver to excite and transmit ultrasonic signals with a frequency in a wide range of 5MHz to the inside of the track plate through the ultrasonic linear array probe, and the ultrasonic linear array probe adopts the mode of spontaneous and self-receiving to obtain the diffused field signal in the track plate and send it to the track plate. The diffuse field signal is transmitted to the computer; c) The computer uses MATLAB software to passively extract the Green's function from the received signal, cross-correlate the diffuse field signal, reconstruct the Green's function between the array elements, obtain the undelayed response between the array elements, and recover from being submerged by noise. The early defect information is obtained, and then the track plate crack imaging is carried out according to the frequency domain synthetic aperture focusing imaging algorithm. 2. The imaging method according to claim 1, wherein the step c) specifically comprises the following operations: 1) Data processing: First of all, due to the interaction between the ultrasonic signal and the defect target during the propagation of the ultrasonic signal inside the ballastless track slab, an approximately uniform sound field will be formed after scattering and multiple reflections, which is the diffusion field. As shown in step b), the diffused field signal in the ballastless track can be obtained through the ultrasonic linear array probe, and any two receiving array elements in the ultrasonic linear array probe are set to be γ ₁ and γ ₂ respectively, and the two receiving elements are If the array elements γ ₁ and γ ₂ are in a closed surface space, then the integral of all sound sources of the sound field cross-spectrum at the receiving array elements γ ₁ and γ ₂ is equal to the frequency domain causal Green function and non-causal Green function between γ ₁ and γ ₂ Difference: G(r ₁ , r ₂ , ω)-G ^* (r ₁ , r ₂ , ω)=-2jω∫G(r ₁ , r, ω)G ^* (r ₂ , r, ω)dV (1); In formula (1), the left side of the equation is the difference between the causal and non-causal functions of the frequency domain Green function between the two receiving elements γ ₁ and γ ₂ , and the right side of the equation represents the receiving elements γ ₁ and γ Integral of all sound sources for cross-correlation of sound fields at ₂ places, specifically, G(γ ₁ , γ ₂ , ω) represents the causal Green function between γ ₁ and γ ₂ in the frequency domain, the superscript * represents the conjugate complex number, G ^* (γ ₁ , γ ₂ , ω) represents the non-causal Green's function between γ ₁ and γ ₂ in the frequency domain, j represents the imaginary unit, j ² =-1, ω represents the angular frequency of the transmitted signal, γ is in the diffusion The position of any point in the closed space V of field integration represents any defect target and can be regarded as a noise source, G(γ ₁ , γ, ω) represents the frequency domain Green’s function propagation formula between γ and γ ₁ , G(γ ₂ , γ, ω) represents the frequency domain Green's function propagation formula between γ and γ ₂ , dV represents the differential of the closed space V; Then, the power spectral density |q(ω)| ² is also independent of the position of the noise source, assuming that the intensity of the diffuse-field noise source is spatially uniformly distributed and uncorrelated with each other: <q(r ₁ , ω)q ^* (r ₂ , ω)>=δ(r ₁ -r ₂ )|q(ω)| ² (2); Equation (2) represents the expected value of the cross-power spectrum of the signals at γ ₁ and γ ₂ in the frequency domain equal to the power spectral density of the noise source between γ ₁ and γ _2. Specifically, in formula (2), <q(γ ₁ , ω )q ^* (γ ₂ , ω/)> represents the expected value of the cross-power spectrum of the signals at γ ₁ and γ ₂ in the frequency domain, q(γ ₁ , ω) represents the field strength at γ ₁ , q(γ ₂ , ω) represents the field strength at γ ₂ , <> represents the expected value of the statistical average, the superscript * represents the conjugate complex number, δ(γ1-γ2) represents the cross-power spectrum of the signal between γ ₁ and γ ₂ is independent of the location of the noise source ; The cross-correlation between the sound fields at the positions of the two receiving array elements γ ₁ and γ ₂ in the frequency domain is: <p(r ₁ , ω)p ^* (r ₂ , ω)>=|q(ω)| ² ∫G(r ₁ , r, ω)G ^* (r ₂ , r, ω)dV (3); In formula (3), p(γ ₁ , ω) represents the sound field at the receiving element γ ₁ , p(γ ₂ , ω) represents the sound field at the receiving element γ ₂ , <p(γ ₁ , ω)p ^* (γ ₂ , ω)> represents the cross-correlation function between the sound fields at the positions of the two receiving array elements γ ₁ and γ ₂ , and the superscript * represents a complex conjugate number; Then, from formulas (1) and (3), we get: (G(r ₁ , r ₂ , ω)-G ^* (r ₁ , r ₂ , ω))|q(ω)| ² =-2jω<p(r ₁ , ω)p ^* (r ₂ , ω) >(4); In formula (4), the left side of the equation is the Green's function G(γ ₁ , γ ₂ , ω) and its corresponding time reversal, that is, the conjugate operation in the corresponding frequency domain. Multiply these two terms by the random noise. After the power spectral density, the result is equal to the cross-correlation result of the two receiving array elements γ ₁ and γ ₂ in the diffuse field on the right side of the equation, and the superscript * represents the conjugate complex number; Then, the time domain expression corresponding to formula (4) is obtained. The frequency domain 2jω corresponds to the time domain 2d/dt. According to the convolution theorem, the product in the frequency domain corresponds to the convolution in the time domain, thus obtaining:

In formula (5), G(γ ₁ , γ ₂ , t) represents the Green's function between γ ₁ and γ ₂ in the time domain, and G(γ ₁ , γ ₂ , -t) represents G(γ ₁ , γ ₂ , t) time reversal, which corresponds to the conjugate operation in the frequency domain, * represents the convolution operation,

represents the cross-correlation operation, C _q (t) represents the autocorrelation result of the noise q(t) in the diffuse field, d/dt represents the derivation of t, p(γ ₁ , t) represents the receiving array element γ ₁ in the time domain The sound field at , p(γ ₂ , t) represents the sound field at the receiving array element γ ₂ in the time domain; The results of formulas (4) and (5) show that the sound fields p(γ ₁ , t) and p(γ ₂ , t) of the receiving array elements in the diffuse field are correlated and derived, and the result is equal to the difference between the two receiving array elements. Green's function response, according to the principle of reciprocity of sound waves, on the time axis, the Green's function response between the two receiving array elements has symmetry; 2) Frequency Domain Synthetic Aperture Focusing (SAFT) imaging: First of all, since the propagation of ultrasonic waves in an infinite homogeneous medium satisfies the Helmholtz wave equation, it is assumed that the coordinates of the ultrasonic transmitting array element are (u, 0) and the coordinates of the receiving array element are (v, 0 under the xz coordinate axis of the two-dimensional space). ), the distances of the defect target from the transmitting array element and the receiving array element are ρ _in and ρ _out respectively, and the time domain receiving signal corresponding to any pair of transmitting and receiving array elements is e(v, u, t), then the corresponding transmitting and receiving The frequency response E(v, u, ω) corresponding to the array element is: E(v, u, ω)=P(ω)∫∫f(x, z)g(ρ _in ,ω)g(ρ _out ,ω)dxdz (6); In formula (6),

P(ω) is the spectrum of the transmitted signal, ω is the angular frequency of the transmitted signal; f(x, z) represents the point spread function of the defective target, and g(ρ _in , ω) represents the spatial Green's function from the transmitting array element to the defective target Frequency domain response, g(ρ _out , ω) represents the spatial Green's function frequency domain response from the receiving array element to the defect target; Also, the Green's function is:

In formula (7), g(x, z, ω) represents the Green’s function of any point under the xz coordinate axis of the two-dimensional space, the wave number k=ω/c, c is the longitudinal wave velocity of the sound wave in the medium, and the spatial variables x and z Corresponds to the wave numbers k _x and k _z ; Use the variables x and z to replace the variables ρ _in and ρ _out , that is, g(x, z, ω) instead of g(ρ _in , ρ _out , ω), bring the formula (7) into the formula (6), and regain the reception Response E(v, u, ω), and then use the wave numbers _ku and k _v to represent the spatial information of the transmitting array element (u, 0) and the receiving array element (v, 0), respectively. In order to realize the transformation of the integral into a two-dimensional Fourier Leaf transform, the Fourier transform of the variables u and v in the received response E(v, u, ω) is obtained:

In formula (8), E(ω, k _u , k _v ) represents the spatial Fourier transform of the frequency domain response E(v, u, ω), and F represents the spatial Fourier transform of the point spread function f(x, z) leaf transform; Correspondingly, the spatial Fourier transform of the point spread function f(x, z) is F(k _x , k _z ). In order to obtain the image domain, the inverse transform model is:

In formula (9),

represents the data domain model of the defect target sampled at equal intervals, S ^-1 {} represents the Stolt map, which completes the conversion from the data domain to the image domain, p ^* (ω) represents the conjugate spectrum of the transmitted signal; The nonlinear coordinate change of formula (9) is performed to map the wavenumbers ku, _kv and _k to the wavenumbers _kx and _kz associated with the defect target in the image domain, then: k _x =k _u +k _v (10);

Then the image of the defect target distribution is obtained by the two-dimensional inverse Fourier transform

In formula (12),

An image domain model representing the distribution of defect targets. 3. an imaging device that realizes automatic detection of track plate cracks, comprising a track inspection trolley and a camera and a computer arranged on the track inspection trolley, the camera is connected with the computer signal, and it is characterized in that: also include an ultrasonic transmitter receiver and an ultrasonic linear array probe, the transmitting end of the ultrasonic linear array probe is signal-connected with the transmitting interface of the ultrasonic transmitting receiver, the receiving end of the ultrasonic linear array probe is signally connected with the receiving interface of the ultrasonic transmitting and receiving device, and the ultrasonic transmitting The receiving interface of the receiver is connected with the computer signal, the ultrasonic linear array probe is connected with an adaptive telescopic mechanism, the adaptive telescopic mechanism is connected with a biaxial displacement adjustment mechanism, and the biaxial displacement adjustment mechanism is fixedly connected At the front end of the rail inspection trolley. 4 . The imaging device according to claim 3 , wherein the self-adaptive telescopic mechanism comprises a mounting plate A and a mounting plate B that are vertically arranged and parallel to each other, and the mounting plate A and the mounting plate B are located between the mounting plate A and the mounting plate B. 5 . A horizontally retractable telescopic bracket is connected, the ultrasonic linear array probe is connected with the mounting plate A, and the biaxial displacement adjusting mechanism is connected with the mounting plate B. 5 . The imaging device according to claim 4 , wherein a probe connector is provided on the side of the mounting plate A away from the mounting plate B, and the ultrasonic linear array probe is fixedly installed on the probe connector. 6 . 6 . The imaging device according to claim 4 , wherein a laser range finder is provided on at least one side of the ultrasonic linear array probe. 7 . 7 . The imaging device according to claim 3 , wherein the biaxial displacement adjusting mechanism comprises a mounting plate C, a longitudinal displacement adjusting mechanism and a lateral displacement adjusting mechanism, and the adaptive telescopic mechanism is fixedly connected with the mounting plate C 7 . , the mounting plate C is slidably connected with the longitudinal displacement adjustment mechanism, the longitudinal displacement adjustment mechanism is slidably connected with the lateral displacement adjustment mechanism, and the lateral displacement adjustment mechanism is fixedly connected to the front end of the rail inspection trolley. 8 . The imaging device according to claim 7 , wherein the longitudinal displacement adjustment mechanism comprises a longitudinal bracket and a longitudinal electric screw adjustment mechanism, and the lateral displacement adjustment mechanism includes a transverse bracket and a transverse electric screw adjustment mechanism, 9 . The mounting plate C is slidably connected up and down with the longitudinal bracket, the longitudinal bracket is slidably connected with the transverse bracket, and the transverse bracket is fixedly connected to the front end of the rail inspection trolley. 9 . The imaging device according to claim 3 , wherein a guide mechanism is further provided at the front end of the on-rail inspection trolley, and the guide wheels constituting the guide mechanism are connected with the rail in a rolling manner. 10 . 10 . The imaging device according to claim 3 , wherein an alarm module is further provided on the rail inspection trolley, and the alarm module is signal-connected to the computer. 11 .

Images