Polygon online detection device and detection method for train wheel

Abstract

The invention relates to a train wheel polygon online detection device and method. The device comprises a bottom plate, a laser sensor and a high-definition night vision camera, wherein the laser sensor and the high-definition night vision camera are arranged along a track; when the laser sensor detects the wheel set of the train, the high-definition night vision camera starts to detect the wheel set of the train, and image acquisition is carried out on the rim of the wheel set and the surface of the tread; and three-dimensional reconstruction is carried out by combining data measured by a plurality of groups of high-definition night vision cameras, so that a three-dimensional model of the measured wheel set can be obtained, and a wheel polygon numerical value of the wheel set is analyzed and calculated by utilizing an algorithm. According to the device, online detection can be realized, a train does not need to stay, accurate and quantitative measurement of polygon parameters of wheels of all wheel pairs can be completed through one-time passing, the algorithm is convenient and fast, the system response speed is high, conflict with train overhaul or turning repair operation is avoided, and the train overhaul application efficiency is also improved while the measurement efficiency is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of rail transit train wheel detection, and in particular to a polygonal online detection device and a detection method for train wheels. Background Art

[0002] Wheel polygon refers to the polygonal shape of the wheel rim and wheel rim of the wheel cross section, which is not an ideal circle and will affect passenger comfort and may even have a negative impact on vehicle structural safety and track durability. The "polygonal" problem of the wheel is usually caused by long-term contact and dynamic interaction between the wheel and the track, and the wheel profile shows periodic non-circular wear. This problem will change the geometry of the wheel-rail contact, resulting in changes in the contact area and pressure, which in turn affects the efficiency and method of force transmission. In addition, the wheel polygon will bring additional periodic impact force to the wheelset and rails, causing cracks in the rails and wheelset. Severe wheel polygons can cause major accidents such as train derailment, resulting in heavy casualties and property losses.

[0003] At present, locomotives mostly use regular maintenance to detect wheel polygons, that is, after each locomotive runs for a period of time or a certain number of kilometers, the locomotive is sent to the maintenance depot for manual maintenance. Traditional train wheel set fault inspection methods mostly use manual inspection, which not only has low detection efficiency and a large detection blind spot, but also workers are easily affected by subjective judgment, resulting in missed or misdetected scratches. Therefore, relying on traditional methods cannot timely and accurately understand the actual use status of the wheelset, causing some wheelsets to "work with injuries" and some wheelsets to "retire prematurely", which not only poses a safety hazard, but also causes huge waste. Summary of the invention

[0004] Based on the above reasons, the present disclosure provides a method and device for online detection of train wheel polygons, which can realize convenient and efficient, highly automated, high-precision, and non-stop measurement online detection of train wheel tread polygons.

[0005] When the laser sensor detects the wheelset of the train, the corresponding high-definition night vision camera starts to detect the wheelset of the train, and collects images of the wheel flange and tread surface of the wheelset; combining the data measured by multiple high-definition night vision cameras, 3D reconstruction is performed to obtain the 3D model of the wheelset being measured, and then the wheel polygon value of the wheelset is calculated using algorithm analysis. During use, the detection cost can be reduced, the detection process is simple, and the detection efficiency is high.

[0006] The train wheel polygon online detection device provided by the present disclosure mainly comprises: a bottom plate, several groups of laser sensors, and at least 4 groups of high-definition night vision cameras; wherein:

[0007] The bottom plate is symmetrically arranged on the outside of the two rails and above the sleepers;

[0008] Each group of laser sensors and each group of high-definition night vision cameras are symmetrically arranged on the bottom plate along the center line of the track, wherein:

[0009] Laser sensors and high-definition night vision cameras are arranged in sequence along the direction of the train's travel;

[0010] Laser sensors are used to detect the passing of train wheelsets;

[0011] The high-definition night vision camera is used to collect images of the wheel flange and tread surface of the wheelset from the oblique front of the wheelset when the laser sensor detects the passing of the train wheelset;

[0012] The data measured by each set of high-definition night vision cameras are used to obtain a three-dimensional model of the measured wheelset through three-dimensional reconstruction, and then the wheel polygon value of the wheelset is obtained;

[0013] Each group of high-definition night vision cameras is arranged at a certain distance.

[0014] Furthermore, a guide rail is provided on the bottom plate;

[0015] Each laser sensor is slidably mounted on the guide rail of the bottom plate through a fixing seat;

[0016] Each high-definition night vision camera is slidably arranged on the guide rail of the bottom plate through a supporting seat.

[0017] Furthermore, a length scale is provided on the guide rail of the bottom plate.

[0018] Furthermore, the lower end of the laser sensor fixing seat is slidably connected to the bottom plate guide rail, and its position on the bottom plate can be fixed by a first fastening bolt;

[0019] The upper end of the fixing seat is connected to the laser sensor through a clamp. The tightness of the clamp can be adjusted by a second fastening bolt and a nut, and is used for fixing and subsequent replacement of the laser sensor.

[0020] Furthermore, the lower end of the high-definition night vision camera support seat is slidably connected to the bottom plate guide rail, and its position on the bottom plate can be fixed by a third fastening bolt;

[0021] The upper end of the support base is rotatably connected to the high-definition night vision camera through an axial hole, and the relative position between the high-definition night vision camera support base and the high-definition night vision camera can be fixed by a fourth fastening bolt.

[0022] Furthermore, the high-definition night vision camera and the laser sensor remain relatively still with respect to the track during measurement.

[0023] Furthermore, the high-definition night vision camera collects images of the surface of the wheel rim and tread of the wheelset at an oblique upward angle, wherein the ground is assumed to be the XOY plane, the direction in which the train moves along the rails is the X-axis, and the Z-axis is perpendicular to the XOY plane. The high-definition night vision camera forms an angle of 30 degrees with the XOZ plane and an angle of 60 degrees with the XOY plane.

[0024] Furthermore, the device includes: 4 groups of laser sensors and 4 groups of high-definition night vision cameras; wherein:

[0025] When the i-th (i=1, 2, 3, 4) group of laser sensors detects the passing of a train wheel set, the i-th group of high-definition night vision cameras begins to collect images of the wheel flange and tread surfaces of the wheelset;

[0026] The distance between the i-th group of laser sensors and the i-th group of high-definition night vision cameras on the same side of the track is l1 = r·cot30°, where r is the wheel radius;

[0027] The distance between two adjacent laser sensors on the same side of the track r is the wheel radius.

[0028] A train wheel polygon online detection method using the above device comprises the following steps:

[0029] Move 4 sets of laser sensors and their fixing seats and 4 sets of high-definition night vision cameras and their supporting seats along the guide rails on the bottom plate to the calculated determined positions and fix them;

[0030] When the train wheelset rolls forward and the first set of laser sensors detects the passing of the train wheelset, the first set of high-definition night vision cameras starts to detect the train wheelset and collect images of the wheel flange and tread surfaces of the wheelset;

[0031] When the second set of laser sensors detects the passing of the train wheelset, the second set of high-definition night vision cameras begins to detect the train wheelset and collect images of the wheel flange and tread surfaces of the wheelset;

[0032] In sequence, when the fourth set of laser sensors detects the passing of a train wheel set, the fourth set of high-definition night vision cameras begins to detect the train wheel set and collect images of the wheel rim and tread surfaces of the wheel set;

[0033] Combining the data measured by four sets of high-definition night vision cameras, a three-dimensional model of the wheelset under test can be obtained through three-dimensional reconstruction.

[0034] The wheel polygon value of the wheelset is calculated by algorithm analysis.

[0035] Compared with the prior art, the beneficial effects of the present disclosure are: (1) high measurement accuracy and high degree of automation, the measurement process can be completed automatically, no manual operation is required on site, and errors caused by human factors are avoided;

[0036] (2) Efficient measurement and high integration: The measuring device can be installed at the throat track where the train passes frequently. The entire measurement process is online, and the train does not need to stop and can be measured immediately. The train can complete the accurate quantitative measurement of the wheel polygon parameters of all wheelsets in one pass;

[0037] (3) The algorithm is convenient and the system responds quickly. It will not conflict with train maintenance or turning operations. It greatly improves the measurement efficiency while also improving the efficiency of train maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The above and other objects, features and advantages of the present disclosure will become more apparent through a more detailed description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present disclosure.

[0039] Figure 1 It is a schematic structural diagram of an exemplary train wheel polygon online detection device according to the present disclosure;

[0040] Figure 2 It is a structural schematic diagram of the right side of an exemplary train wheel polygon online detection device;

[0041] Figure 3 is a schematic diagram of an exemplary laser sensor fixing device structure;

[0042] Figure 4 It is a schematic diagram of the structure of an exemplary high-definition night vision camera support device;

[0043] In the figure, 1 is a wheelset, 2 is a rail, 3 is a base plate, 4 is a laser sensor and its fixing seat, 5 is a high-definition night vision camera and its supporting seat; 41 is a first group of laser sensors, 42 is a second group of laser sensors, 43 is a third group of laser sensors, 44 is a fourth group of laser sensors, 51 is a first group of high-definition night vision cameras, 52 is a second group of high-definition night vision cameras, 53 is a third group of high-definition night vision cameras, and 54 is a fourth group of high-definition night vision cameras; 411 is a laser sensor, 412 is a laser sensor fixing seat, 413 is a 10mm fixing bolt, 414 is a 4mm fixing bolt, 415 is a 4mm fixing nut, 511 is a high-definition night vision camera, and 512 is a high-definition night vision camera supporting seat. DETAILED DESCRIPTION

[0044] The preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0045] The present invention provides a train wheel polygon online detection method and device, which can reduce the detection cost during use, and has a simple detection process and high detection efficiency.

[0046] According to an exemplary embodiment of the present disclosure, Figure 1 As shown, the train wheel polygon online detection device includes a bottom plate 3, 4 sets of laser sensors and their fixing devices 4, and 4 sets of high-definition night vision cameras and their supporting devices 5. The bottom plate 3 is installed on the outer side of the rail 2 and above the sleeper 6.

[0047] like Figure 3 As shown, the laser sensor and its fixing device 4 include: a laser sensor 411 and a laser sensor fixing seat 412. The lower end of the laser sensor fixing seat 412 can be slidably connected to the guide rail of the base plate 3. The 10mm bolt 413-1 can be tightened to fix the position of the laser sensor fixing seat on the base plate 3. The upper end of the laser sensor fixing seat 412 is connected to the laser sensor 411 through a clamp. The 4mm bolt 414 and the 4mm nut 415 can control the tightness of the clamp to facilitate the fixation and subsequent replacement of the laser sensor 411.

[0048] like Figure 4 As shown, the high-definition night vision camera and its supporting device 5 include: a high-definition night vision camera supporting seat 512 and a high-definition night vision camera 511. The lower end of the high-definition night vision camera supporting seat 512 is slidably connected to the guide rail of the bottom plate 3. The 10mm bolt 413-2 is tightened to fix the position of the high-definition night vision camera supporting seat 512 on the bottom plate 3. The upper end of the high-definition night vision camera supporting seat 512 is rotatably connected to the high-definition night vision camera 511 through the shaft hole. The 10mm bolt 413-3 can be tightened to determine the relative position between the high-definition night vision camera supporting seat 512 and the high-definition night vision camera 511.

[0049] Preferably, assuming that the ground is an XOY plane, the train's advancing direction along the rail 2 is an X axis, and the Z axis is perpendicular to the XOY plane, the high-definition night vision camera 511 forms an angle of 30 degrees with the XOZ plane and an angle of 60 degrees with the XOY plane. The high-definition night vision camera 511 collects images of the surfaces of the wheel rim and tread of the wheelset 1.

[0050] Preferably, a guide rail is provided on the upper portion of the base plate 3 .

[0051] Preferably, a length scale is provided on the upper portion of the bottom plate 3 .

[0052] Preferably, the distance l1 between the same group of laser sensors 411 and the high-definition night vision camera 511 on the same side of the train wheel polygon online detection device is equal to r·cot30°, where r is the wheel radius.

[0053] As a preferred embodiment, in the train wheel polygon online detection device, the distance between two adjacent laser sensors on the same side is r is the wheel radius.

[0054] The train wheel polygon online detection method based on the above device mainly includes the following steps:

[0055] S100: Move 4 sets of laser sensor fixing devices and 4 sets of high-definition night vision camera supporting devices along the guide rails on the bottom plate to the calculated determined positions and fix them. When the train wheelset rolls forward and the first set of laser sensors detects the passing of the train wheelset, the first set of high-definition night vision cameras starts to detect the train wheelset and collect images of the wheel flange and tread surfaces of the wheelset;

[0056] S200: When the second group of laser sensors detects that a train wheelset has passed, the second group of high-definition night vision cameras starts to detect the train wheelset and collects images of the surfaces of the wheel rim and tread of the wheelset;

[0057] S300: When the third group of laser sensors detects that a train wheel set has passed, the third group of high-definition night vision cameras starts to detect the train wheel set and collect images of the wheel rim and tread surfaces of the wheel set;

[0058] S400: When the fourth group of laser sensors detects that a train wheel set has passed, the fourth group of high-definition night vision cameras starts to detect the train wheel set and collect images of the wheel rim and tread surfaces of the wheel set;

[0059] S500: Combine the data measured by four high-definition night vision cameras to perform 3D reconstruction to obtain a 3D model of the wheelset being measured;

[0060] S600: Calculate the wheel polygon value of the wheelset by using algorithm analysis.

[0061] The train wheel polygon online detection device provided in this embodiment has high measurement efficiency and high integration. The measuring device can be installed at the throat track where the train passes frequently. The entire measurement process is online, and the train does not need to stop, and it can measure and go immediately; the train can complete the accurate quantitative measurement of the wheel polygon parameters of all wheel pairs in one pass; the algorithm is convenient and the system responds quickly, and there will be no conflict with the train maintenance or turning and repair operations, which greatly improves the measurement efficiency and also improves the train maintenance efficiency.

[0062] The above technical scheme is only an exemplary embodiment of the present invention. For those skilled in the art, it is easy to make various types of improvements or modifications based on the application methods and principles disclosed in the present invention, and it is not limited to the method described in the above specific embodiment of the present invention. Therefore, the method described above is only preferred and does not have a restrictive meaning.

Claims

1. A train wheel polygon online detection device, characterized in that: include: Base plate, several sets of laser sensors, and at least 4 sets of high-definition night vision cameras; among them: The bottom plate is symmetrically arranged on the outside of the two rails and above the sleepers; Each group of laser sensors and each group of high-definition night vision cameras are symmetrically arranged on the bottom plate along the center line of the track, wherein: Laser sensors and high-definition night vision cameras are arranged in sequence along the direction of the train's travel; Laser sensors are used to detect the passing of train wheelsets; The high-definition night vision camera is used to collect images of the wheel flange and tread surface of the wheelset from the oblique front of the wheelset when the laser sensor detects the passing of the train wheelset; The data measured by each set of high-definition night vision cameras are used to obtain a three-dimensional model of the measured wheelset through three-dimensional reconstruction, and then the wheel polygon value of the wheelset is obtained; Each group of high-definition night vision cameras is arranged at a certain distance.

2. The device according to claim 1, characterized in that The bottom plate is provided with a guide rail; Each laser sensor is slidably mounted on the guide rail of the bottom plate through a fixing seat; Each high-definition night vision camera is slidably arranged on the guide rail of the bottom plate through a supporting seat.

3. The device according to claim 2, characterized in that The guide rail of the bottom plate is provided with a length scale.

4. The device according to claim 2, characterized in that The lower end of the laser sensor fixing seat is slidably connected to the bottom plate guide rail, and its position on the bottom plate can be fixed by a first fastening bolt; The upper end of the fixing seat is connected to the laser sensor through a clamp. The tightness of the clamp can be adjusted by a second fastening bolt and a nut, and is used for fixing and subsequent replacement of the laser sensor.

5. The device according to claim 2, characterized in that The lower end of the high-definition night vision camera support seat is slidably connected to the bottom plate guide rail, and its position on the bottom plate can be fixed by a third fastening bolt; The upper end of the support base is rotatably connected to the high-definition night vision camera through an axial hole, and the relative position between the high-definition night vision camera support base and the high-definition night vision camera can be fixed by a fourth fastening bolt.

6. The device according to claim 2, characterized in that The high-definition night vision camera and the laser sensor remain relatively still with the track during measurement.

7. The device according to claim 2, characterized in that The high-definition night vision camera collects images of the wheel flange and tread surface of the wheelset at an oblique upward angle, wherein the ground is assumed to be the XOY plane, the direction in which the train moves along the rails is the X-axis, and the Z-axis is perpendicular to the XOY plane. The high-definition night vision camera forms an angle of 30 degrees with the XOZ plane and an angle of 60 degrees with the XOY plane.

8. The device according to any one of claims 2 to 7, characterized in that: include: 4 sets of laser sensors and 4 sets of high-definition night vision cameras; including: When the i-th (i=1, 2, 3, 4) group of laser sensors detects the passing of a train wheel set, the i-th group of high-definition night vision cameras begins to collect images of the wheel flange and tread surfaces of the wheelset; The distance between the i-th group of laser sensors and the i-th group of high-definition night vision cameras on the same side of the track is l1 = r·cot30°, where r is the wheel radius; The distance between two adjacent laser sensors on the same side of the track r is the wheel radius.

9. A train wheel polygon online detection method using the device of claim 8, comprising the following steps: Move 4 sets of laser sensors and their fixing seats and 4 sets of high-definition night vision cameras and their supporting seats along the guide rails on the bottom plate to the calculated determined positions and fix them; When the train wheelset rolls forward and the first set of laser sensors detects the passing of the train wheelset, the first set of high-definition night vision cameras starts to detect the train wheelset and collect images of the wheel flange and tread surfaces of the wheelset; When the second set of laser sensors detects the passing of the train wheelset, the second set of high-definition night vision cameras begins to detect the train wheelset and collect images of the wheel flange and tread surfaces of the wheelset; In sequence, when the fourth set of laser sensors detects the passing of a train wheel set, the fourth set of high-definition night vision cameras begins to detect the train wheel set and collect images of the wheel rim and tread surfaces of the wheel set; Combining the data measured by four sets of high-definition night vision cameras, a three-dimensional model of the wheelset under test can be obtained through three-dimensional reconstruction. The wheel polygon value of the wheelset is calculated by algorithm analysis.

Images