Train brake pad wear detection system and detection method

Abstract

The application discloses a train brake pad abrasion detection system and a detection method, and relates to the field of train detection. The system comprises a trackside device, a field control mechanism and a remote control center, the field control mechanism is in communication connection with the trackside device and the remote control center respectively; the trackside device comprises an incoming line sensor, an offline sensor, a wheel sensor and an image acquisition device, the wheel sensor is located on both sides of the image acquisition device; the image acquisition device is provided with a plurality of image acquisition devices, which correspondingly acquire the images of each brake pad on the train axle and the wheel; the image acquisition device comprises a plane array camera and two angle lasers, the center lines of the light sections of the two angle lasers are parallel to the optical axis of the plane array camera, and are perpendicular to the ground. The application non-contactly carries out train brake pad image acquisition, train speed measurement, axle number statistics and non-contact train brake pad abrasion detection.

Description

Technical Field

[0001] This invention belongs to the field of train brake pad testing technology, and particularly relates to a train brake pad wear testing system and method. Background Technology

[0002] With the rapid development of railways, train operation safety has become a key research focus in the rail transit field. Brake pads, as braking devices for trains, are tightly fitted to the brake disc under the action of the brake calipers when the train brakes, generating enormous friction to decelerate the train. While the train is moving, the brake pads are in a state of disengagement from the brake disc. According to its principle, brake pads will become thinner due to prolonged wear. When the thickness falls below a certain value, its braking capacity will weaken, posing a safety hazard to the train. Therefore, how to effectively detect the thickness of brake pads is one of the key issues in ensuring train operation safety.

[0003] In existing train safety monitoring equipment, brake pad thickness detection relies primarily on manual measurement, which is labor-intensive, prone to errors, and requires the train to be stationary, impacting train operation efficiency. Other brake pad detection methods simply use edge detection techniques on images to locate the brake pad area and estimate its thickness. However, due to the influence of stray edges and edge breaks, the brake pad position cannot be reliably determined; furthermore, natural wheel wear and refining cause significant variations in the distance between the brake pad and the camera, resulting in large fluctuations in the size of the brake pad in the image and making accurate thickness calculation impossible.

[0004] Therefore, how to improve the existing train brake pad wear detection technology, or propose a new train brake pad wear detection technology, has become one of the urgent issues to be addressed by those skilled in the art. Summary of the Invention

[0005] In view of the shortcomings of current train brake pad wear detection technology, the purpose of this invention is to provide a train brake pad wear detection system.

[0006] Another objective of this invention is to provide a method for detecting wear on train brake pads.

[0007] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0008] A train brake pad wear detection system includes trackside equipment, a field control mechanism, and a remote control center. The field control mechanism is communicatively connected to both the trackside equipment and the remote control center. The trackside equipment includes an incoming line sensor, an offline sensor, wheel sensors, and an image acquisition device. The wheel sensors are located on both sides of the image acquisition device. Multiple image acquisition devices are provided to acquire images of each brake pad on the train axle and wheel. Each image acquisition device includes an area array camera and two angled lasers. The center line between the two angled lasers is parallel to the optical axis of the area array camera and perpendicular to the ground.

[0009] Furthermore, the wavelengths of the lasers in the two adjacent image acquisition devices and the filters of the corresponding area array cameras are different.

[0010] Furthermore, the ten image acquisition devices are arranged in a row and mounted on a mounting base between two sleeper rails.

[0011] Furthermore, the trackside equipment also includes a speed measuring radar and a vehicle number recognition device. The speed measuring radar detects the real-time speed of the trains entering the track, and the vehicle number recognition device identifies the vehicle number of the trains entering the track.

[0012] Furthermore, the lenses of the incoming line sensor, speed radar, vehicle number recognition device, wheel sensor, image acquisition device, and offline sensor are all coated with a waterproof nanomaterial membrane.

[0013] A detection method for a train brake pad wear detection system, characterized in that the method includes:

[0014] (1) When a train approaches the detection system, the train approach signal is obtained through the approach sensor to start the detection system; the wheel sensor obtains the position of each wheel and starts each image acquisition device to acquire the image of each corresponding brake pad;

[0015] (2) The image acquisition device is calibrated using a checkerboard calibration plate. The physical size represented by the pixels captured by the camera at object distances of 400mm and 1000mm and the spacing between the parallel lines on the calibration plate are calibrated respectively. According to the imaging principle and projection relationship, it can be known that the physical size represented by the pixels at 400mm and 1000mm and the pixel spacing between the parallel laser lines on the calibration plate at the corresponding positions have a linear relationship.

[0016] (3) The image of the brake pad is analyzed by image processing technology. The position of the parallel laser line, the pixel spacing of the parallel laser line, and the average pixel length of the parallel laser line in the brake pad area are detected. Based on the pixel spacing of the parallel laser line and the relationship between the calibrated pixel spacing and the physical size represented by the pixel, the physical size represented by the pixel in the current image is calculated. The average physical length of the parallel laser line is further calculated, which is the remaining thickness of the brake pad. The wear condition of the brake pad is obtained by subtracting it from the standard thickness of the known brake pad standard parts.

[0017] Furthermore, steps (2) and (3):

[0018] The image acquired at 400mm is processed to obtain the pixel side length w1 of each block, the pixel size of the distance between the parallel lines of the two lasers is h1, and it is known that each block in the calibration block is a square with a physical size of a mm. The physical size represented by each pixel in the image taken by the camera at a distance of 400mm is calculated to be a / w1.

[0019] The image acquired at 1000mm is processed to obtain the pixel side length w2 of each block, and the pixel size of the distance between the parallel lines of the two lasers is h2. The physical size represented by each pixel in the image captured by the camera at a distance of 1000mm is calculated to be a / w2.

[0020] According to the projection relationship, at a distance x from the camera and 400mm≤x≤1000mm, the spacing h of the parallel laser lines in the image is linearly related to the distance x, and x=(400h / h1+1000h / h2) / 2.

[0021] Furthermore, in step (3): according to the principle of pinhole imaging, the pixel width of the checkerboard pattern captured at a distance x is w = (400w1 / x + 1000w2 / x) / 2 = (400w1 + 1000w2) / 2x;

[0022] The physical dimensions represented by each pixel at a distance x can be calculated as y = a / w = 2ax / (400w1 + 1000w2) = (400h / h1 + 1000h / h2)a / (400w1 + 1000w2).

[0023] Furthermore, in step (1): the speed measuring radar obtains the real-time passing speed of the train, calculates the acquisition frame rate of the matched area array camera based on the train speed, and sends pulses of the corresponding frequency to the image acquisition device; the train number recognition device obtains the train number of the passing train.

[0024] When a wheel passes the wheel sensor closest to the direction of the incoming line, the image acquisition device starts to continuously acquire images of the brake pads. When the wheel passes the wheel sensor furthest from the direction of the incoming line, the image acquisition device stops acquiring images. Images of the brake pads are acquired every time a wheel passes between two wheel sensor intervals.

[0025] The train brake pad wear detection system of the present invention has the following functions: (1) non-contact train brake pad image acquisition, train speed measurement, and axle count statistics; (2) non-contact train brake pad wear detection;

[0026] This invention employs an image acquisition device (image acquisition device) consisting of a laser beam with two angled surfaces and an area array camera. The angled laser beam is projected as auxiliary structured light onto the surface of the gate plate under test, while the area array camera simultaneously acquires images of the gate plate. The angled laser beam appears as two parallel bright lines on the plane within the gate plate area, and as zigzag lines between the gate plate edge and other adjacent components, effectively distinguishing the gate plate area from the non-gate plate area. Furthermore, the distance between the two parallel laser lines in the gate plate area of ​​the image can be used to determine the distance between the gate plate and the camera, allowing for the selection of different calibration coefficients to calculate the gate plate thickness.

[0027] There are multiple brake pads on a single axle. Each brake pad is photographed by a shooting device consisting of two angled laser beams and an area array camera. The laser beams of two adjacent shooting devices use lasers of different wavelengths, and matching filters are installed on the camera lenses to avoid light source interference between adjacent shooting devices. By coordinating the work of multiple shooting devices, the wear of train brake pads can be detected online in real time without stopping. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the layout for image acquisition in the train brake pad wear detection system of the present invention;

[0029] Figure 2 This is a flowchart of the train brake pad wear detection system of the present invention;

[0030] Figure 3 This is a calibration schematic diagram of the image acquisition device for the train brake pad wear detection system of the present invention;

[0031] Figure 4 This is a schematic diagram of the parallel lines projected by the laser onto the brake pad in the image acquisition device of the train brake pad wear detection system of the present invention;

[0032] Among them, 1. First track, 2. Second track, 3. Incoming line sensor, 4. Speed ​​measuring radar, 5. Vehicle number recognition device, 6. Brake pad, 7. Image acquisition device, 8. Wheel sensor, 9. Offline sensor, 10. Checkerboard calibration plate. Detailed Implementation

[0033] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0034] The train brake pad wear detection system of this embodiment includes trackside equipment, a field control center, and a remote control center. The field control center is communicatively connected to both the trackside equipment and the remote control center. The trackside equipment includes a basic detection unit and a mounting base. The basic detection unit includes an incoming line sensor 3, a speed measuring radar 4, a vehicle number recognition device 5, a positioning device (wheel sensor 8), an imaging device (image acquisition device 7), and an offline sensor 9. The mounting base includes a concrete foundation between two sleepers.

[0035] refer to Figure 1 As shown, the approach sensor 3 is mounted on the inner side of the rail via a mounting bracket, serving as the starting position of the entire system. It detects the wheels of the approaching train, obtains the train approach signal, and activates the control system.

[0036] The speed measuring radar 4 is installed on a column outside the trackside safety clearance to detect the real-time speed of incoming trains.

[0037] The train number recognition device 5 is installed on a column outside the trackside safety clearance to identify the train number of the train entering the line.

[0038] Two wheel sensors 8 are combined with ten image acquisition devices 7. The two sets of wheel sensors 8 are mounted on the inside of a single track by mounting brackets and are located on both sides of the ten image acquisition devices 7 to locate the wheel position; that is, the two wheel sensors 8 control the ten image acquisition devices 7 to acquire images by locating the wheels.

[0039] Ten image acquisition devices 7 are arranged in a row and installed on the mounting base between two sleepers, respectively aligned with the positions of the brake pads 6 distributed on the train axle and wheels. Each image acquisition device 7 acquires image information of a brake pad 6 at a certain position.

[0040] like Figure 1 and Figure 4 As shown, the image acquisition device 7 includes an area scan camera and two angled lasers with an angle of 3-5° between them. The area scan camera is located behind the two angled lasers, and the emitting surfaces of the two angled lasers are parallel to the camera lens surface. The center line between the two angled lasers is parallel to the optical axis of the area scan camera and perpendicular to the ground.

[0041] As the train wheels pass over wheel sensors 8, two angled lasers in the image acquisition device 7 project two parallel laser lines onto the brake pad 6. Simultaneously, the area array camera in the image acquisition device 7 captures images of the brake pad. Wheel sensors 8 begin capturing images when one wheel passes and end capturing images when the second wheel passes.

[0042] Preferably, the laser wavelengths and corresponding camera filters in two adjacent image acquisition devices 7 are different. The two adjacent image acquisition devices 7 use lasers with wavelengths of 808nm and 915nm respectively to be distributed alternately, thereby avoiding light source interference between adjacent image acquisition devices 7.

[0043] Preferably, the image acquisition device 7 is equipped with a blower dust removal device for self-cleaning. This device is also small in size, low in cost, and very convenient for on-site installation and maintenance. The specific structure of the blower dust removal device can be found in CN217037269U, a blower dust removal hatch device, and will not be described further here.

[0044] The offline sensor 9 is mounted on the inside of the rail via a mounting bracket, serving as the termination point of the entire system. It detects the train wheels, receives a train departure signal, and shuts down the control system.

[0045] Preferably, the lenses of each detection module (incoming sensor 3, speed radar 4, vehicle number recognition device 5, wheel sensor 8, image acquisition device 7, and offline sensor 9) of the basic detection unit are coated with a waterproof nanomaterial membrane.

[0046] The on-site control center is responsible for power supply, control, data and image acquisition, analysis, processing, and storage of the basic detection units, while also communicating with the remote control center. The on-site control center includes an image processing computer, equipment control box, wheel sensor processing device, vehicle number industrial control computer, switch, data server, uninterruptible power supply, and surge protection box.

[0047] The workflow of the detection method of the train brake pad wear detection system in this embodiment is as follows: Figure 2 As shown:

[0048] When no train is approaching the track, the system is in a dormant state, waiting to receive a train.

[0049] When the train enters the track, the wheels pass the entry sensor 3 mounted on the inner side of the second track 2. The entry sensor 3 triggers a valid entry signal, and the system starts to enter the train receiving state.

[0050] The image acquisition device 7 opens the protective door and turns on the air blowing dust removal device on the window glass. At the same time, the window glass is heated to remove water mist.

[0051] The system activates the speed measuring radar 4 to detect the speed of passing vehicles, calculates the acquisition frame rate of the matched area array camera based on the vehicle speed, and sends pulses of the corresponding frequency to the image acquisition device 7.

[0052] When a wheel passes the wheel sensor 8, which is mounted on the inner side of the first track 1 and is close to the direction of the incoming line, the image acquisition device 7 starts to continuously acquire images of the brake pad 6. When the wheel passes the wheel sensor 8, which is away from the direction of the incoming line, the image acquisition device 7 stops acquiring images. Each wheel will acquire images of the brake pad 6 within the interval between the two wheel sensors 8.

[0053] When the train passes through and goes offline, the wheels trigger the offline sensor 9, and the number of times the offline sensor 9 triggers the wheels is consistent with the number of times the incoming line sensor 3 is triggered. The system completes image acquisition, the train goes offline, the image acquisition device 7 is turned off, and the system enters a sleep state waiting for the train to arrive.

[0054] Note: The units for pixel length, pixel spacing, and pixel distance in the following text are pixels, which are the pixels in the image, while the units for physical size and physical length are mm.

[0055] In this embodiment, the image acquisition device 7 acquires images of the train brake pad 6. Two angled laser beams are projected onto the lower surface of the brake pad 6, for reference. Figure 4 Two parallel lines appear in the area of ​​the brake disc 6, and a zigzag line appears in the transition area between the brake disc 6 and other equipment. The area array camera acquires images of the brake disc 6. By analyzing the characteristics of the laser lines in the image, image processing technology can be used to obtain the pixel length of the parallel laser lines in the area of ​​the brake disc 6 (the length of the parallel laser lines in the image of the brake disc area, i.e., the corresponding pixel length) and the pixel distance between the parallel lines (the distance h between the parallel laser lines in the image of the brake disc area, i.e., the corresponding pixel distance). The physical length of the parallel laser lines in the area of ​​the brake disc 6 is obtained according to the calibration parameters. The average value is taken as the thickness of the brake disc 6. The difference between this and the standard thickness of the brake disc 6 is used to obtain the wear of the brake disc 6, which can improve the accuracy of the wear detection of the brake disc 6.

[0056] Specifically, a checkerboard calibration plate is used to calibrate the image acquisition device 7. The physical size represented by the pixels captured by the camera at object distances of 400mm and 1000mm is calibrated, as is the spacing between the parallel laser lines on the calibration plate. According to the imaging principle and projection relationship, it can be known that the physical size represented by the pixels at 400mm and 1000mm and the pixel spacing between the parallel laser lines on the calibration plate at the corresponding positions have a linear relationship.

[0057] Image processing techniques are used to analyze the acquired brake pad images. The position x of the parallel laser lines in the brake pad area, the pixel spacing of the parallel laser lines (i.e., the pixel size h of the spacing between parallel laser lines below), and the average pixel length of the parallel laser lines are detected (the brake pad image is converted to grayscale, two parallel laser lines are obtained through threshold segmentation, the pixel lengths of the two laser lines are counted, and the average pixel length is calculated). Based on the relationship between the pixel spacing of the parallel laser lines and the calibrated pixel spacing and the physical size represented by the pixels, the physical size represented by the pixels in the current image is calculated. Further conversion (physical length = physical size y * number of pixels on the laser line, the number of pixels is obtained through image processing) yields the average physical length of the parallel laser lines. The average physical length of the parallel laser lines represents the remaining thickness of the brake pad. The wear of the brake pad can be obtained by subtracting the standard thickness of the standard brake pad component. Detected brake pad wear is located and alarmed according to vehicle number, axle number, and over-limit conditions, notifying manual confirmation and repair.

[0058] like Figure 3 As shown, more specifically, including:

[0059] 1) Image acquisition: The image acquisition device 7 is a device for image acquisition, including an area array camera and two sets of line lasers at different angles. In this embodiment, there are a total of 10 image acquisition devices 7, which correspond to the brake pads 6 at various positions on the axle and wheel, respectively, to realize the image acquisition of the brake pads 6 at each position.

[0060] 2) Image calibration: The image acquisition devices 7 are calibrated using a checkerboard calibration plate. Each image acquisition device 7 is calibrated separately. The checkerboard calibration plate is calibrated by taking pictures with the camera in the acquisition device at object distances of 400mm and 1000mm respectively. The plane of the calibration plate is perpendicular to the optical axis of the camera.

[0061] The image acquired at 400mm is processed to obtain the pixel side length w1 of each block, the pixel size of the distance between the parallel lines of the two lasers is h1, and it is known that each block in the calibration block is a square with a physical size a (in mm). After calculation, the physical size represented by each pixel in the image taken by the camera at a distance of 400mm is a / w1.

[0062] Similarly, the pixel side length w2 of each square is obtained through image processing from the image acquired at 1000mm, and the pixel size of the distance between the parallel lines of the two lasers is h2. After calculation, the physical size represented by each pixel in the image captured by the camera at a distance of 1000mm is a / w2.

[0063] According to the projection relationship, when the distance between the gate plate 6 and the image acquisition device 7 is x mm and 400mm≤x≤1000mm, the pixel size h of the spacing between parallel laser lines in the image is linearly related to the distance x, and x=(400h / h1+1000h / h2) / 2.

[0064] According to the principle of pinhole imaging in cameras, the pixel width of the checkerboard pattern (i.e., the pixel side length of each square in the checkerboard pattern) taken at a distance x is w = (400w1 / x + 1000w2 / x) / 2 = (400w1 + 1000w2) / 2x.

[0065] Substituting w and x into the formula y = a / w, the physical dimensions represented by each pixel at a distance x can be calculated as y = a / w = 2ax / (400w1 + 1000w2) = (400h / h1 + 1000h / h2)a / (400w1 + 1000w2). The brake pad wear detection system processes the acquired image of brake pad 6. Using image processing algorithms, the system detects parallel laser lines in the region of brake pad 6, identifies the average pixel length of the parallel laser lines and the pixel spacing between them, and obtains the average physical length of the parallel laser lines, i.e., the remaining thickness of brake pad 6, based on the calibration relationship. The difference between this average and the standard thickness of brake pad 6 is the wear value of the brake pad. An alarm is then triggered based on whether the wear value exceeds the limit.

[0066] The present invention has been disclosed above with reference to preferred embodiments, but these are not intended to limit the present invention. Anyone skilled in the art can make various changes or modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be defined by the scope of the claims in this application.

Claims

1. A detection method for a train brake pad wear detection system, characterized in that, The method includes: (1) When the train approaches the detection system, the train approach signal is obtained through the approach sensor to start the detection system; the wheel sensor obtains the position of each wheel and starts each image acquisition device to acquire the image of each corresponding brake pad; (2) The image acquisition device was calibrated using a checkerboard calibration board. The physical dimensions of the pixels captured by the camera at object distances of 400mm and 1000mm and the spacing between the parallel lines on the calibration board were calibrated respectively. According to the imaging principle and projection relationship, it can be known that the physical dimensions of the pixels at 400mm and 1000mm and the pixel spacing between the parallel laser lines on the calibration board at the corresponding positions have a linear relationship: the pixel side length w1 of each square in the image acquired at 400mm is obtained through image processing, and the pixel size of the spacing between the parallel lines of the two lasers is h1. It is known that each square in the calibration block is a square with a physical size of a. mm, the physical size represented by each pixel in the image taken by the camera at a distance of 400mm is calculated as a / w1; the pixel side length w2 of each block in the image acquired at 1000mm is obtained through image processing, and the pixel size of the distance between the parallel lines of the two lasers is h2. The physical size represented by each pixel in the image taken by the camera at a distance of 1000mm is calculated as a / w2; according to the projection relationship, at a distance x from the camera and 400mm≤x≤1000mm, the distance h of the parallel laser lines in the image is linearly related to the distance x, and x=(400h / h1+1000h / h2) / 2. (3) Analyze the acquired gate image through image processing technology, detect the position of the parallel laser line in the gate area, the pixel spacing of the parallel laser line, and the average pixel length of the parallel laser line. Based on the relationship between the pixel spacing of the parallel laser line and the calibrated pixel spacing and the physical size represented by the pixel, calculate the physical size represented by the pixel in the current image: According to the principle of pinhole imaging, the width of the checkerboard pixel at a distance x is w=(400w1 / x+1000w2 / x) / 2=(400w1+1000w2) / 2x; the physical size represented by each pixel at a distance x can be calculated as y=a / w=2ax / (400w1+1000w2)=(400h / h1+1000h / h2)a / (400w1+1000w2); Further calculations yield the average physical length of the parallel laser lines, which is the remaining thickness of the brake pad: Average physical length of parallel laser lines = Remaining thickness of the brake pad = w*y; The wear condition of the brake pad is then obtained by subtracting it from the standard thickness of a known standard brake pad component.

2. The detection method as described in claim 1, characterized in that, Step (1): The speed measuring radar obtains the real-time passing speed of the train, calculates the acquisition frame rate of the matching area array camera based on the train speed, and sends pulses of the corresponding frequency to the image acquisition device. The train number recognition device obtains the train number of the passing train; When a wheel passes the wheel sensor closest to the direction of the incoming line, the image acquisition device starts to continuously acquire images of the brake pads. When the wheel passes the wheel sensor furthest from the direction of the incoming line, the image acquisition device stops acquiring images. Images of the brake pads are acquired every time a wheel passes between two wheel sensor intervals.

3. The detection method as described in claim 1, characterized in that, The system includes trackside equipment, a field control mechanism, and a remote control center. The field control mechanism is communicatively connected to both the trackside equipment and the remote control center. The trackside equipment includes incoming line sensors, offline sensors, wheel sensors, and image acquisition devices. The wheel sensors are located on both sides of the image acquisition devices. Multiple image acquisition devices are provided to acquire images of each brake pad on the train axle and wheel. Each image acquisition device includes an area array camera and two angled lasers. The center line between the two angled lasers is parallel to the optical axis of the area array camera and perpendicular to the ground.

4. The detection method as described in claim 3, characterized in that, The wavelengths of the lasers in the two adjacent image acquisition devices and the filters of the corresponding area array cameras are different.

5. The detection method as described in claim 3, characterized in that, The ten image acquisition devices are arranged in a row and mounted on a mounting base between two sleeper rails.

6. The detection method as described in claim 3, characterized in that, The trackside equipment also includes a speed measuring radar and a vehicle number recognition device. The speed measuring radar detects the real-time speed of the trains entering the track, and the vehicle number recognition device identifies the vehicle number of the trains entering the track.

7. The detection method as described in claim 6, characterized in that, The lenses of the incoming line sensor, speed radar, vehicle number recognition device, wheel sensor, image acquisition device, and offline sensor are all coated with a waterproof nanomaterial membrane.

Images