Turnout frame size detection method

Abstract

The method comprises the steps of rail piece contour obtaining, rail piece area positioning, rail head contour external rectangle obtaining and frame size calculation, according to contour data collected by a high-precision line laser sensor and corresponding mileage data, all target detection positions are automatically positioned, and corresponding contour data are extracted; according to the relative position distribution of the rail piece at the target detection position, the stock rail head area and the stock rail head and switch rail head area of the switch rail section are automatically identified and positioned from the extracted contour data, the size of the turnout frame is calculated through the minimum enclosing rectangle, and the method is convenient to operate, high in detection precision and high in accuracy. The method can be applied to various tracks with different sizes, only one frame of contour at the preset detection position is extracted for subsequent calculation and analysis, three-dimensional modeling, three-dimensional model comparison and other operations are not needed according to contour data, the operand is very small, and a detection result can be quickly obtained on a track detection site.

Description

Technical Field

[0001] The invention belongs to the technical field of track detection, and in particular relates to a method for detecting the size of a turnout frame. Background Art

[0002] Track geometry is an important indicator for measuring the quality of railway lines and operational safety, and is directly related to the smoothness, comfort and driving safety of trains. Therefore, there are high requirements for the detection of track geometry. The detection of track geometry also includes the detection of various geometric dimensions of turnouts.

[0003] Turnouts are line connection devices that allow trains to transfer from one track to another. They are also one of the weak links of the track and are usually laid in large quantities at stations and marshaling yards. There are various track components in the turnout area, among which the turnout frame is the main load-bearing component of the turnout. It is fixed on the track and is responsible for supporting all the components of the turnout. The turnout frame needs to have sufficient rigidity, strength, and good geometric shape to prevent deformation or damage due to force when the train passes through the turnout, and to ensure that the turnout has good operating performance. Therefore, the size detection of the turnout frame is very important. The turnout frame size (frame width) specifically refers to the distance from the left side of the basic rail to the right side of the basic rail on each inspection surface of the point rail, that is, the inner spacing between the two basic rails. For the size detection of the turnout frame, the five section positions of the point rail with a section width of 0, 5, 20, 50, and 70 mm are measured on the conventional track; on the high-speed track, the measurement is made at the sleeper position from the tip of the point rail to the end of the point rail. At present, manual inspection is mainly used. Inspectors go to the above-mentioned inspection locations respectively, use the track gauge to measure according to the requirements, and manually record various information including the inspection value, track number, sequence, etc. The manual inspection method is time-consuming, the inspection efficiency is low, and the inspection accuracy is greatly affected by human operation. On the other hand, when the straight point rail fits the base rail, the manual measurement method has a large error. Summary of the invention

[0004] The present invention is made to solve the above problems, and aims to provide a detection method that can automatically obtain the turnout frame size and has high detection accuracy. The present invention adopts the following technical solutions:

[0005] The present invention provides a method for detecting the size of a turnout frame. The method has the following technical features: based on the profile data of a turnout point rail section collected by multiple line laser sensors and the mileage data collected by a mileage detector, the turnout frame sizes at multiple predetermined target detection positions are calculated. The method comprises the following steps: a rail profile acquisition step, in which the profile data at multiple target detection positions are respectively extracted according to the profile data; a rail area positioning step, in which two predetermined rail areas are respectively identified and positioned from the profile data based on the extracted profile data at the target detection position and according to the relative position relationship of multiple rails at the target detection position, which are the basic rail head area on one side and the point rail and basic rail head area on the other side; a rail head profile circumscribed rectangle acquisition step, in which, based on the located two predetermined rail areas, the profile between the highest point of the basic rail in the two predetermined rail areas and the predetermined descending distance thereof is respectively positioned, and the minimum circumscribed rectangle of the profile is acquired; a frame size calculation step, in which the size of the turnout frame at each of the predetermined detection positions is respectively calculated based on the two minimum circumscribed rectangles and the full section width value of the basic rail.

[0006] The turnout frame size detection method provided by the present invention may also have such a technical feature, wherein the rail area positioning step includes the following sub-steps: for the contour data of each target detection position, the tuple point data collected by the multiple line laser sensors are converted into corresponding contours respectively; the contour of the rail head area of ​​the rail member is screened out from each of the contours according to a predetermined height range; according to the calibration parameters of the multiple line laser sensors, the contours of the rail head area of ​​the rail member corresponding to the multiple line laser sensors are merged to obtain a merged contour; the merged contour is drawn in the image, and a morphological closing operation is performed on the target in the image, a connected area analysis is performed on the result of the closing operation, and two predetermined rail areas are positioned in combination with the predetermined width characteristics of the predetermined rail area and their relative position relationship.

[0007] The turnout frame size detection method provided by the present invention may also have such a technical feature, wherein the frame size calculation step includes the following sub-steps: for the two predetermined rail areas, respectively extracting the contour between the highest point of the basic rail and the predetermined spacing below it, and obtaining the minimum circumscribed rectangle of the contour parallel to the coordinate axis; for the two minimum circumscribed rectangles, calculating the distance from the inner side of the minimum circumscribed rectangle with a smaller width to the outer side of the minimum circumscribed rectangle with a larger width; subtracting the predetermined full-section width of the basic rail from the distance, thereby obtaining the turnout frame size.

[0008] The turnout frame size detection method provided by the present invention may also have such a technical feature, wherein, in the frame size calculation step, for the two predetermined rail areas, the contour between the highest point of the basic rail and 16 mm below it is extracted respectively, and the minimum circumscribed rectangle of the contour is obtained, and the full section width of the basic rail is 71 mm.

[0009] The turnout frame size detection method provided by the present invention may also have such a technical feature, wherein, for the two circumscribed rectangles, it is determined which side is the inner side of the circumscribed rectangle with a smaller width and the outer side of the circumscribed rectangle with a larger width based on their widths and relative positions, and the corner point coordinates of the corresponding side are taken based on the judgment result, and then the distance is calculated based on the two corner point coordinates.

[0010] The turnout frame size detection method provided by the present invention may also have such a technical feature, wherein the positioning and extraction step includes the following sub-steps: based on the contour data and the track structure characteristics at the starting point of the point rail, identifying the starting point of the point rail from multiple frames of the contour data and obtaining the corresponding mileage value; according to the mileage value of the starting point of the point rail and the distance from the predetermined starting point of the point rail to each target detection position, obtaining the mileage value of each target detection position; according to the mileage value of each target detection position, extracting the contour data of each target detection position from the multiple frames of the contour data.

[0011] The turnout frame size detection method provided by the present invention may also have such a technical feature, before the positioning and extraction step, it also includes: a data acquisition step, in which data of the track of the turnout point rail section is collected by a track detection device, wherein the plurality of line laser sensors and the mileage detector are arranged in the track detection device, and the plurality of line laser sensors respectively correspond to the two sides of the track of the point rail section, and are used to collect contour data of the same section of the track.

[0012] The turnout frame size detection method provided by the present invention may also have such a technical feature, wherein, there are at least six line laser sensors, and at least three line laser sensors are respectively arranged on both sides of the track detection device. When the track detection device is arranged on the track of the point rail section, the three line laser sensors on each side are respectively located above the outer side, above, and above the inner side of the basic rail on the corresponding side, and all face the basic rail. In the rail area positioning step, the tuple point data collected by the three line laser sensors on the corresponding side are converted into corresponding contours, and the contours of the three line laser sensors on the corresponding side are merged according to the calibration parameters of the multiple line laser sensors.

[0013] The turnout frame size detection method provided by the present invention may also have such a technical feature, and before the data collection step, it also includes: a preparation step, jointly calibrating the multiple line laser sensors of the track detection device, and time synchronizing the multiple line laser sensors and the mileage detector.

[0014] Functions and Effects of the Invention

[0015] According to the turnout frame size detection method provided by the present invention, there are a rail contour acquisition step, a rail area positioning step, a rail head contour circumscribed rectangle acquisition step, and a frame size calculation step, wherein each target detection position is automatically positioned by contour data collected by a high-precision line laser sensor and corresponding mileage data, and the corresponding contour data is extracted, and then the basic rail head area (i.e., the side where the point rail repels) and the basic rail head and point rail head area (i.e., the side where the point rail fits the basic rail) of the point rail segment are automatically identified and located from the extracted contour data according to the relative position distribution of the rail at the target detection position, and then the turnout frame size is calculated by the minimum circumscribed rectangle. The method is easy to operate and has high detection accuracy, and can be applied to tracks of various sizes. Moreover, since only one frame of contour at the predetermined detection position is extracted for subsequent calculation and analysis, and no operations such as three-dimensional modeling and three-dimensional model comparison are required based on the contour data, the amount of calculation is very small, which is conducive to quickly obtaining detection results at the track detection site. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of a method for detecting the size of a turnout frame in an embodiment of the present invention;

[0017] Figure 2 is a three-dimensional diagram of a track detection device in an embodiment of the present invention;

[0018] Figure 3 is a schematic diagram of the distribution of internal sensors of a track detection device in an embodiment of the present invention;

[0019] Figure 4 yes Figure 2 An enlarged view of the portion within the middle frame A;

[0020] Figure 5 is a flow chart of step S3 in this embodiment;

[0021] Figure 6 is a flow chart of step S4 in this embodiment;

[0022] Figure 7 is an example diagram of the merged contour in this embodiment;

[0023] Figure 8 is an example diagram of positioning two predetermined rail regions in this embodiment;

[0024] Fig. 9 is a flow chart of step S6 in this embodiment;

[0025] Fig.10 is an example diagram of extracting a circumscribed rectangle in this embodiment;

[0026] Fig.11 This is an example diagram for calculating the turnout frame size in this embodiment.

[0027] Reference numerals:

[0028] Track detection device 100; device body 10; main unit 11; unit housing 112; shading assembly 113; shading plate 1131; line laser sensors 1411~1146; moving mechanism 20; wheel frame 21; running wheel assembly 22; running wheel 222; auxiliary cone wheel 223; fixed side wheel assembly 23; detector assembly 25; pushing mechanism 30. DETAILED DESCRIPTION

[0029] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the turnout frame size detection method of the present invention is specifically described below in conjunction with embodiments and drawings.

[0030] <Example>

[0031] This embodiment provides a method for detecting the size of a turnout frame. In this embodiment, a track detection device is used to collect data of a turnout point rail segment, and then the method is used to automatically analyze and calculate the size of the turnout frame. The method requires that the track detection device has at least three line laser sensors on each side, that is, at least six in total, and has a mileage detector (mileage encoder). A track detection device is exemplarily shown below.

[0032] Figure 2 is a three-dimensional diagram of the track detection device in this embodiment, Figure 3 It is a schematic diagram of the distribution of internal detection components of the track detection device in this embodiment.

[0033] like Figure 2 and Figure 3 As shown, the track detection device 100 is an H-type comprehensive track detection device (detection vehicle), which includes a device body 10, a moving mechanism 20, and a pushing mechanism 30.

[0034] The device body 10 is roughly in the shape of a long rectangular parallelepiped as a whole, and includes two main units 11. Each main unit 11 is also roughly in the shape of a rectangular parallelepiped, including a unit frame, a unit shell 112, a shading component 113, and a plurality of detection components. The unit frame is a metal part and has a triangular grid structure. The unit shell 112 covers the three side surfaces and the top surface of the unit frame, and the two main units 11 are connected to each other on the side that is not covered. The shading component 113 includes a plurality of shading plates 1131 that are roughly in the shape of rounded rectangular plates, which are respectively installed under the three side surfaces of the unit shell 112 and are slightly inclined outward from the unit shell 112. In an alternative embodiment, the shading component can also be in the shape of a skirt with an opening on one side, and the shape is roughly the shape formed by connecting the gaps between the three shading plates 1131 in the figure.

[0035] Multiple detection components are fixed in the unit frame through corresponding mounting parts, that is, placed in the unit housing 112. The detection components of one main unit 11 include three line laser sensors 1411-1143, and the detection components of another main unit 11 include three line laser sensors 1144-1146. For the convenience of description, they are respectively recorded as line lasers No. 1 to No. 6 below.

[0036] like Figure 3 As shown, in a main body unit 11, three line laser sensors are arranged near the top of the main body unit 11, wherein the line laser sensor 1411 is fixed at the outer end of the unit body 11 in the length direction and is arranged obliquely, the line laser sensor 1412 is fixed at the middle of the unit body 11 in the length direction and is arranged obliquely, and the inclination angle is smaller than that of the line laser sensor 1141, and the line laser sensor 1143 is fixed at the inner end of the unit body 11 in the length direction and is arranged obliquely, and the inclination direction is different from that of the other two line laser sensors. When the device is placed on the track in the turnout area, the line laser sensor 1411 is fixed at the inner end of the unit body 11 in the length direction and is arranged obliquely, and the inclination direction is different from that of the other two line laser sensors. The laser sensor 1411 is tilted toward the outside of the basic rail, and can cover part of the outer side surface and top surface of the basic rail, as well as part of the outer side surface and top surface of the rail member outside the basic rail (if any, the same below); the line laser sensor 1412 is slightly tilted toward the top surface of the rail, and can completely cover the top surface and part of the inner side surface of the basic rail, as well as part of the top surface of the rail member outside the basic rail, and the top surface of the rail member inside the basic rail; the line laser sensor 1143 is tilted toward the inside of the rail, and can basically completely cover the top surface and part of the inner side surface of the basic rail, as well as the top surface and part of the inner side surface of the rail member inside the basic rail.

[0037] The three line laser sensors in the other main body unit 11 are arranged similarly, with the only difference being that the line laser sensors in the two main body units 11 are arranged in mirror symmetry along the center line of the device body 10 .

[0038] Optionally, an inclinometer, a gyroscope, a two-dimensional camera, etc. may be further provided in the main unit 11 so that the device can cover more track detection items.

[0039] The shading component 113 is used to block external light to reduce the impact of changes in external ambient light on the line laser sensor, and the brightness of the collected contour data is more uniform and there are fewer missing points.

[0040] Figure 4 yes Figure 2 An enlarged view of the portion within frame A.

[0041] like Figure 2 and Figure 4 As shown, the mobile mechanism 20 includes two wheel frames 21, four walking wheel assemblies 22, four fixed side wheel assemblies 23, a floating connection assembly (not shown in the figure), and at least one detector assembly 25.

[0042] The wheel frame 21 is a metal frame, and its overall shape is similar to the top and two sides of a trapezoid, and it also has a triangular grid structure. Its two ends are semi-enclosed shell-shaped and have mounting holes for mounting the travel wheel assembly 22. A pair of square mounting holes are provided on the two opposing sides of the above-mentioned unit housing 112, and the middle section of the wheel frame 21 passes through the pair of mounting holes to be connected to the device body 10. The size of the square mounting hole on one of the unit housings 112 is slightly larger than the cross-sectional size of the middle part of the wheel frame 21. The wheel frame 21 passes through the mounting hole on the unit housing 112 and is connected to the main unit 11 through a floating connection assembly. It can rotate slightly along the length direction of the device body 10, and its rotation is limited by the mounting holes. With such a design, the four travel wheels can be kept in contact with the rails as much as possible in places where the track is uneven, so that the overall movement of the device is more stable.

[0043] Two running wheel assemblies 22 are respectively arranged on the two ends of the wheel frame 21. The running wheel assembly 22 includes a rotating shaft, a bearing shaft, a running wheel 222, and an auxiliary cone wheel 223. The running wheel 222 is rotatably arranged on the wheel frame 21 through the rotating shaft and the bearing shaft. The running wheel 222 is cylindrical and is used to contact the top surface of the rail; the auxiliary cone wheel 223 is in the shape of a truncated cone and is coaxially arranged on one side of the running wheel 222, and its large end faces the running wheel 222. When the device is placed on the track, the auxiliary cone wheel 223 is located on the outside of the rail, and is used to avoid and assist the stability of the device at the positions of the point rail, the heart rail, etc. when the device enters the turnout.

[0044] The relatively large shading plate 1131 in the shading assembly 113 extends obliquely from the lower end of the main shell to the end of the wheel frame 21 (above the running wheel), that is, the distance between the lower end of the shading plate 1311 and the rail is very small, so it can effectively block most of the external light.

[0045] The four fixed side wheel assemblies 23 are respectively arranged on the ends of the wheel frame 21 next to the four running wheel assemblies 22. The fixed side wheel assemblies 23 include side wheel brackets extending downward from the ends of the wheel frame 21 and side wheels installed at the bottom of the side wheel brackets. Through the action of the fixed side wheel assemblies, the device always runs near the center of the track without deviating too far. When passing through the center rail area, the side wheels will abut against the guard rails, and by providing lateral restraint force, the device is laterally limited to ensure that the device can pass through the center rail area normally.

[0046] The detector assembly 25 includes a detector housing and a mileage detector (mileage encoder). The detector housing is fixed to the outside of one end of the wheel frame 21 and is arranged corresponding to the running wheel 222. The mileage detector is installed in the housing and is coaxially arranged with the running wheel 222. When the running wheel 222 contacts the track and rolls along the track, the mileage detector can obtain corresponding mileage data.

[0047] The pushing mechanism 30 includes a T-shaped push rod and an angle adjustment assembly, and is used to allow the inspector to push the device.

[0048] In addition, the device also has an electronic control component, including a battery, a data collector, a data calculation analyzer (such as a laptop computer), etc.

[0049] In addition, it is worth mentioning that, considering the movement and detection needs of the device at the turnout, the device does not have spring side wheels, but only fixed side wheels. After modeling analysis and repeated test adjustments, a suitable installation position is designed for the four fixed side wheels so that the device can pass through the center rail area normally. The installation of spring side wheels will affect the walking and detection accuracy of the device at the turnout. The following are a few examples to illustrate this:

[0050] For example, due to the gap between the wing rail and the center rail, when passing through this position, the gap may cause the spring side wheel to lose support, causing the device to deviate or jam.

[0051] In the front area of ​​the heart rail, there is a certain height difference between the top surface of the base rail and the top surface of the heart rail. The spring side wheels may not be able to adapt to the change in height difference due to elastic force limitations, causing the overall center of gravity of the device to shift or jump, thus affecting the detection accuracy; it may also cause the overall instability of the device and make it unable to continue moving along the track.

[0052] The width of the tip of the heart rail gradually narrows. When the spring side wheels are used to press the track components on both sides, the spring side wheels or fixed side wheels may interfere with the track components near the gauge point, resulting in unstable walking and affecting the detection accuracy.

[0053] Since the tip of the heart rail is relatively sharp, the side wheels will not be in sufficient contact with the track. Therefore, the spring side wheels cannot apply uniform pressure when passing through the pointed rail part, causing the entire device to drift or misalign relative to the track, thereby affecting the detection accuracy.

[0054] In the discontinuous area where the point rail and the heart rail transition, the side wheels become suspended or lose support, causing the device to be unable to maintain a stable posture.

[0055] Therefore, the device of this embodiment does not have spring side wheels. The fixed side wheels at the installation position designed through modeling enable the device to move stably and smoothly in the turnout area, and accurate detection results are obtained through algorithm design.

[0056] The method of this embodiment will be described in detail below in conjunction with the above device structure.

[0057] Figure 1 Flowchart of the turnout frame size detection method in this embodiment.

[0058] like Figure 1 As shown, the method comprises the following steps:

[0059] Step S1 (preparation step), before starting data collection, multiple line laser sensors of the track detection device are jointly calibrated, and multiple line laser sensors and mileage encoders are time synchronized.

[0060] Step S2 (data collection step), data collection is performed on the turnout point rail section through a track detection device, wherein contour data on both sides of the track are acquired respectively by a plurality of line laser sensors, and mileage data of the device moving along the track is acquired by a mileage detector.

[0061] Step S3 (track contour acquisition step), respectively acquiring contours of multiple target detection positions according to the contour data.

[0062] Step S4 (rail area positioning step), based on the acquired contour at the target detection position and according to the relative position relationship of the rail at the target detection position, two predetermined rail areas are identified and positioned respectively, namely the basic rail head area on one side and the point rail and basic rail head area on the other side.

[0063] Step S5 (the step of obtaining the circumscribed rectangle of the rail head contour), based on the two predetermined rail areas located, first locate the contour between the highest point of the basic rail on one side and the point 16 mm below it, and obtain the minimum circumscribed rectangle of the contour parallel to the coordinate axis, then locate the contour between the highest point of the point rail and the basic rail on the other side and the point 16 mm below it, and obtain the minimum circumscribed rectangle of the contour parallel to the coordinate axis.

[0064] Step S6 (frame size calculation step), based on the acquired column coordinate values ​​of the corresponding two circumscribed rectangles and the width value of the full cross section of the basic rail, the frame size of the turnout at each predetermined detection position is calculated respectively.

[0065] The above steps will be described in detail below.

[0066] Step S1 (preparation step), before starting data collection, multiple line laser sensors of the track detection device are jointly calibrated, and multiple line laser sensors and mileage encoders are time synchronized.

[0067] Among them, by designing and carefully adjusting the installation positions of multiple line laser sensors so that their projected line lasers are basically collinear, it is possible to collect contour data of the same cross-section of the track, and use a calibration plate with multiple square columnar teeth and triangular pointed columnar teeth to jointly calibrate multiple line laser sensors, so that the measured contour data can be subsequently integrated into a unified coordinate space, and the correspondence between the contour data measured by the line laser sensor and the actual physical scale can be obtained.

[0068] Time synchronization is the premise of multi-source sensor data fusion, which can avoid the accumulation of errors caused by asynchrony and provide accurate basic data for subsequent horizontal value calculation. For example, multiple line laser sensors, inclinometers and mileage detectors can be connected to the same system-on-chip (such as the system-on-chip of a data collector). Before starting to collect data, when starting the system-on-chip, or when sending corresponding instructions to it, the system-on-chip performs time synchronization operations on all line laser sensors and mileage detectors through local time and timer values ​​in the timer core.

[0069] Step S2 (data collection step), data collection is performed on the turnout point rail section through a track detection device, wherein contour data on both sides of the track are acquired respectively by a plurality of line laser sensors, and mileage data of the device moving along the track is acquired by a mileage detector.

[0070] Among them, the inspection personnel place the device on the track so that its four running wheels are respectively placed on the top surfaces of the two rails, and push the device to move along the track. During the movement, the data collector controls multiple sensors to synchronously collect data, thereby collecting the contour data of a certain cross section of the track. The data collection of the line laser sensor can be triggered at a predetermined distance according to a predetermined mileage interval; the data of the line laser sensor can also be triggered at a predetermined time interval according to a predetermined time interval. In this case, the inspection personnel need to push the track inspection device relatively evenly. In this embodiment, the collection is triggered according to the mileage, and a frame of contour is collected every 10mm traveled.

[0071] The contour data obtained by scanning the track with a line laser sensor is tuple point data, including multiple data points, each of which includes a point space coordinate value, a ray reflection intensity (brightness value) and a corresponding mileage value. In this embodiment, the contour data collected by the above-mentioned line lasers 1 to 6 are used.

[0072] Step S3 (track contour acquisition step), respectively acquiring contours of multiple target detection positions according to the contour data. Figure 5 is a flow chart of step S3 in this embodiment.

[0073] like Figure 5 As shown, step S3 specifically includes the following sub-steps:

[0074] Step S3-1, based on the contour data and the track structure characteristics at the starting point of the point rail, identify the starting point of the point rail from multiple frames of contour data and obtain the corresponding mileage value.

[0075] Step S3-2, obtaining the mileage value of each target detection position according to the mileage value of the starting point of the pointed rail and the distance from the predetermined starting point of the pointed rail to each target detection position.

[0076] For example, each frame of contour data can be extracted and analyzed in sequence according to the mileage value. Before the starting point of the sharp rail, the contour data only contains the contour of the basic rail; at the starting point of the sharp rail, the contour of the sharp rail further appears in the contour data. Therefore, a frame of contour data at the starting point of the sharp rail can be easily found to obtain the corresponding mileage value as the mileage value of the starting point of the sharp rail.

[0077] Step S3-3, extracting contour data of each target detection position from multiple frames of contour data according to the mileage value of each detection position.

[0078] In this embodiment, the contour data of the five cross-sectional positions of the point rail, namely 0, 5, 20, 50, and 70 mm cross-sectional width, are extracted to calculate the corresponding turnout frame size. In an alternative solution, according to actual needs, only some of the positions or more positions of contour data can be taken for calculation.

[0079] Step S4 (rail area positioning step), based on the acquired contour at the target detection position and according to the relative position relationship of multiple rails at the target detection position, two predetermined rail areas are identified and positioned, namely the basic rail head area on one side and the point rail and basic rail head area on the other side.

[0080] Figure 6 It is a flow chart of step S4 in this embodiment.

[0081] like Figure 6 As shown, step S4 specifically includes the following sub-steps:

[0082] Step S4-1, for the contour data of each target detection position, convert the tuple point data collected by multiple line laser sensors into corresponding contours respectively.

[0083] Step S4-2: Filter out the contour of the rail head area from various contours according to a predetermined height range.

[0084] Step S4-3, according to the calibration parameters, the contours of the rail head area of ​​the rail member corresponding to the multiple line laser sensors are merged, and certain miscellaneous edges are removed after merging to obtain a merged contour (target contour).

[0085] Figure 7 is an example diagram of merging contours in this embodiment.

[0086] like Figure 7 As shown, the merged contour line of one target detection position has multiple contour parts. The left side of the figure is the contour collected by the laser lines 4, 5, and 6, and the right side of the figure is the contour collected by the laser lines 1, 2, and 3. The two contour parts include the contours of the corresponding basic rail and pointed rail respectively.

[0087] Step S4-4, draw the merged contour in the image, perform morphological closing operation on the target in the image, perform connected region analysis on the result of the closing operation, and locate the two predetermined track areas in combination with the predetermined width features of the two predetermined track areas and their relative position relationship.

[0088] Figure 8 This is an example diagram of positioning two predetermined rail areas in this embodiment.

[0089] like Figure 8 As shown, the two predetermined rail parts areas located are the base rail head area on one side (left side in the figure) and the point rail and base rail head area on the other side (right side in the figure). The left side is the contour area where the point rail excludes the base rail, and the right side is the contour area where the point rail fits the base rail.

[0090] Step S5 (the step of obtaining the circumscribed rectangle of the rail head contour), based on the two predetermined rail areas located, first locate the contour between the highest point of the basic rail on one side and the point 16 mm below it, and obtain the minimum circumscribed rectangle of the contour parallel to the coordinate axis, then locate the contour between the highest point of the point rail and the basic rail on the other side and the point 16 mm below it, and obtain the minimum circumscribed rectangle of the contour parallel to the coordinate axis.

[0091] Fig.10 This is an example diagram of extracting a circumscribed rectangle in this embodiment.

[0092] like Fig.10As shown, the contour between the highest point of the left basic rail and the point 16mm below it is extracted, and its minimum circumscribed rectangle surrounds the upper part of the left basic rail head. The contour between the highest point of the right basic rail and the point 16mm below it is extracted, and its minimum circumscribed rectangle surrounds the upper part of the right basic rail head and the upper part of the point rail head.

[0093] Step S6 (frame size calculation step), based on the acquired column coordinate values ​​of the corresponding two circumscribed rectangles and the width value of the full cross section of the basic rail, the frame size of the turnout at each predetermined detection position is calculated respectively.

[0094] Fig. 9 This is a flow chart of step S6 in this embodiment.

[0095] like Fig. 9 As shown, step S6 includes the following sub-steps:

[0096] Step S6-1: For two circumscribed rectangles, calculate the distance from the inner side of the circumscribed rectangle with a smaller width to the outer side of the circumscribed rectangle with a larger width.

[0097] Among them, first, according to the widths and relative positions of the two, it is determined which side is the inner side of the circumscribed rectangle with the smaller width and which side is the outer side of the circumscribed rectangle with the larger width. Then, according to the determination result, the column coordinate value of the corresponding side is obtained, and then the above distance is calculated based on the two column coordinate values.

[0098] For example, for Fig.10 The two circumscribed rectangles shown are judged according to their widths and relative positions that the right side is the inner side of the circumscribed rectangle with a smaller width and the outer side of the circumscribed rectangle with a larger width, and the coordinates of their lower right corner points are obtained accordingly. Then, the difference between the x value of the lower right corner point of the circumscribed rectangle on the left and the x value of the lower right corner point of the circumscribed rectangle on the right is calculated to obtain the above distance.

[0099] Step S6-2, subtract the predetermined basic rail full section width (71 mm) from the above distance to obtain the turnout frame size (frame width).

[0100] Fig.11 This is an example diagram for calculating the turnout frame size in this embodiment.

[0101] like Fig.11 As shown, according to Fig.10 After calculating the distance value of the two circumscribed rectangles, subtract the full section width of the right basic rail from the distance value to get the distance between the effective edge of the left basic rail and the effective edge of the right basic rail, thereby obtaining the turnout frame size.

[0102] Through the above steps, the turnout frame dimensions at multiple target detection positions can be obtained, and in actual detection, the detection personnel only need to push the above track detection device to move along the turnout point rail section once, and then the results can be automatically obtained.

[0103] The above method can be run as a corresponding program on the data calculation and analysis device (laptop computer) of the device, or as a corresponding program on a remote data calculation and analysis device, and the device only collects data and transmits the data to the remote device.

[0104] For the device and method of the present embodiment, the inventors used them to perform multiple turnout frame size detections at multiple target detection positions on an actual turnout point rail segment, and also used a 0-level track gauge to perform turnout frame size detection at the same position of the point rail segment to verify the effectiveness of the method of the present invention. Through multiple measurements and comparisons, it was proved that the turnout frame size measured by the track detection device and method of the present embodiment was highly consistent with the value measured by the 0-level track gauge.

[0105] Furthermore, the track detection device shown in this embodiment is only an exemplary structure. The method of the present invention only requires the track detection device to have at least six line laser sensors and mileage detectors, and does not depend on a specific vehicle body structure. Therefore, it can also be used for track detection devices of other structures.

[0106] Functions and Effects of the Embodiments

[0107] According to the turnout frame size detection method provided by the present embodiment, there are a rail contour acquisition step, a rail area positioning step, a rail head contour circumscribed rectangle acquisition step, and a frame size calculation step, wherein each target detection position is automatically positioned by contour data collected by a high-precision line laser sensor and corresponding mileage data, and the corresponding contour data is extracted, and then the basic rail head area (i.e., the side where the point rail is repelled) and the basic rail head and point rail head area (i.e., the side where the point rail fits the basic rail) of the point rail segment are automatically identified and located from the extracted contour data according to the relative position distribution of the rail at the target detection position, and then the turnout frame size is calculated by the minimum circumscribed rectangle. The method is easy to operate and has high detection accuracy, and can be applied to tracks of various sizes. Moreover, since only one frame of contour at the predetermined detection position is extracted for subsequent calculation and analysis, and no operations such as three-dimensional modeling and three-dimensional model comparison are required based on the contour data, the amount of calculation is very small, which is conducive to quickly obtaining detection results at the track detection site.

[0108] In the embodiment, for two predetermined rail regions, the turnout frame size is calculated by taking the minimum circumscribed rectangle between the basic rail surface and 16 mm below it, which can well fit the contour shape and reduce the blank area, thereby obtaining a more accurate turnout frame size.

[0109] Furthermore, after extracting the profile of each rail head, it is only necessary to find the corresponding corner points of the two circumscribed rectangles and calculate the coordinate difference according to the detection requirements. The subsequent calculation and analysis is relatively simple and intuitive, with small amount of calculation and high calculation accuracy, which can avoid the problem of cumulative errors caused by algorithms such as multiple profile matching.

[0110] Furthermore, the method of the embodiment can automatically and accurately locate multiple target detection positions, extract corresponding contour data, automatically identify the rail head area, and calculate the target detection value. The degree of automation is high. During detection, the detection personnel only need to push the track detection device to walk in the switch area once, and basically no human intervention is required afterwards, which can greatly reduce the burden on the detection personnel, and will not cause differences in detection accuracy due to the detection personnel's experience, operation differences, etc.

[0111] Furthermore, before starting data collection, multiple line laser sensors were jointly calibrated, so that the data collected by multiple line laser sensors can be easily merged later and the profile lines can be made to correspond to the real physical scale. In addition, all sensors were time synchronized before data collection to avoid error accumulation caused by multi-sensor time asynchrony during data collection, providing accurate basic data for subsequent calculation and analysis.

[0112] Furthermore, the line laser sensors are all arranged inside a semi-enclosed device casing, and the device is also equipped with a light-shielding blocker extending above the walking wheel, which can effectively block most of the external ambient light and avoid the influence of excessive or changing external ambient light on the line laser sensor, making the sensor data more accurate and also helping to improve the accuracy of the key dimensions of the guardrail.

[0113] Furthermore, the track detection device is an H-type integrated track detection device, which can cover multiple items of detection in the mainline track and turnout area. After adopting the above method, the integrated track detection device does not need to be equipped with spring side wheels, which can avoid the problems of multiple walking and detection in the turnout area caused by the installation of spring side wheels, so that the integrated track detection device can detect accurate key dimensions of the guardrail while also being able to stably move in the turnout area and measure more accurate data. Furthermore, the device is equipped with auxiliary cone wheels, and its multiple fixed side wheels have been modeled, designed, and carefully tested and adjusted, so that it can move more stably and smoothly in the turnout area.

[0114] The above embodiments are only used to illustrate the specific implementation of the present invention, and the present invention is not limited to the description scope of the above embodiments. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only to illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for detecting the size of a turnout frame, characterized in that: The method comprises the following steps: calculating the turnout frame dimensions at a plurality of predetermined target detection positions based on the profile data of the turnout point rail segment collected by a plurality of line laser sensors and the mileage data collected by the mileage detector; A rail profile acquisition step, extracting the profile data at a plurality of target detection positions respectively according to the profile data; A rail area positioning step, based on the extracted contour data of the target detection position, and according to the relative position relationship of multiple rails at the target detection position, two predetermined rail areas are respectively identified and positioned from the contour data, namely, a base rail head area on one side and a point rail and base rail head area on the other side; A step of obtaining a circumscribed rectangle of a rail head contour, based on the located two predetermined rail regions, respectively locating the contour between the highest point of the basic rail in the two predetermined rail regions and the point where it descends a predetermined distance, and obtaining a minimum circumscribed rectangle of the contour; The frame size calculation step is to calculate the frame size of the turnout at each of the predetermined detection positions based on the two minimum circumscribed rectangles and the full section width of the basic rail.

2. The turnout frame size detection method according to claim 1, Features: The rail area positioning step includes the following sub-steps: For the contour data of each target detection position, the tuple point data collected by the plurality of line laser sensors are converted into corresponding contours respectively; Selecting the profile of the rail head area of ​​the rail member from each of the profiles according to a predetermined height range; According to calibration parameters of the plurality of line laser sensors, contours of the rail head region of the rail member corresponding to the plurality of line laser sensors are merged to obtain a merged contour; The merged contour is drawn in the image, and a morphological closing operation is performed on the target in the image. A connected region analysis is performed on the result of the closing operation, and the two predetermined track areas are located in combination with the predetermined width features of the predetermined track areas and their relative positional relationship.

3. The turnout frame size detection method according to claim 1, characterized in that: in, The frame size calculation step includes the following sub-steps: For the two predetermined rail regions, respectively extract the contour between the highest point of the basic rail and the point where it descends by a predetermined distance, and obtain the minimum circumscribed rectangle of the contour parallel to the coordinate axis; For the two minimum circumscribed rectangles, calculate the distance from the inner side of the minimum circumscribed rectangle with a smaller width to the outer side of the minimum circumscribed rectangle with a larger width; The distance is subtracted from the predetermined full-section width of the stock rail to obtain the turnout frame size.

4. The method for detecting the size of a turnout frame according to claim 3, characterized in that: in, In the frame size calculation step, for the two predetermined rail regions, the contour between the highest point of the basic rail and the point 16 mm below it is extracted, and the minimum circumscribed rectangle of the contour is obtained. The full section width of the basic rail is 71 mm.

5. The method for detecting the size of a turnout frame according to claim 3, characterized in that: in, For the two circumscribed rectangles, determine which side is the inner side of the circumscribed rectangle with a smaller width and the outer side of the circumscribed rectangle with a larger width based on their widths and relative positions, and take the column coordinate value of the corresponding side based on the judgment result, and then calculate the distance based on the two column coordinate values.

6. The turnout frame size detection method according to claim 1, Features: The positioning extraction step includes the following sub-steps: Based on the profile data and the track structure characteristics at the starting point of the point rail, identifying the starting point of the point rail from multiple frames of the profile data, and obtaining corresponding mileage values; According to the mileage value of the starting point of the pointed rail and the distance from the predetermined starting point of the pointed rail to each of the target detection positions, the mileage value of each of the target detection positions is obtained; The contour data of each target detection position is extracted from the contour data of multiple frames according to the mileage value of each target detection position.

7. The method for detecting the size of a turnout frame according to claim 1, characterized in that: Before the positioning extraction step, the method further includes: Data collection step: collect data on the track of the turnout point rail section through the track detection device. Among them, multiple line laser sensors and mileage detectors are all arranged in the track detection device, and multiple line laser sensors correspond to both sides of the track of the pointed rail section respectively, and are used to collect contour data of the same section of the track.

8. The method for detecting the size of a turnout frame according to claim 7, characterized in that: in, There are at least six line laser sensors, and at least three line laser sensors are respectively arranged on both sides of the track detection device. When the track detection device is arranged on the track of the pointed track section, the three line laser sensors on each side are respectively located above the outer side, above the upper side, and above the inner side of the basic track on the corresponding side, and all face the basic track. In the rail area positioning step, the tuple point data collected by the three line laser sensors on the corresponding side are converted into corresponding contours, and the contours of the three line laser sensors on the corresponding side are merged according to the calibration parameters of the multiple line laser sensors.

9. The method for detecting the size of a turnout frame according to claim 6, characterized in that: Before the data collection step, the method further includes: The preparation step is to jointly calibrate the multiple line laser sensors of the track detection device and to time synchronize the multiple line laser sensors and the mileage detector.

Images