A method for automatically measuring the spacing between cross-cords of contact network

Abstract

The invention relates to the technical field of contact network measurement, and relates to a method for automatically measuring the spacing of contact network crossing lines. The method comprises the following steps: step 1. parameter acquisition, step 2. line identification; step 3. line spacing calculation; step 4. supplementary manual measurement, any two points in a three-dimensional point cloud can be selected, and the straight-line distance between the two points and the projection distance of each two-dimensional surface can be directly calculated; step 5. the above steps 1-4 are repeated until the measurement is completed. The invention can calculate the shortest distance between all lines of the entire line in batches at one time without using additional measuring equipment to perform individual distance measurement on the line, and the measurement result is both accurate and fast. At the same time, the manual measurement is supplemented, and any two points in a three-dimensional point cloud scene can be selected to view their straight-line distance and the projection distance of each two-dimensional surface, so as to meet various requirements for the spacing of contact network lines.

Description

Technical Field

[0001] The invention relates to the technical field of contact network measurement, in particular to a method for automatically measuring the spacing between cross-wires of a contact network. Background Art

[0002] In electrified railways, in order to ensure the normal operation of electric locomotives, the contact network needs to provide reliable and uninterrupted power supply to electric locomotives. As an important power supply line, the contact network is a key link to ensure the safe operation of trains. The proportion of contact network failures is relatively high, so it is necessary to conduct regular inspections to detect potential safety hazards in a timely manner. Among the inspection items, the geometric parameters of various contact lines in the contact network, especially the spacing between cross lines, become an important item. According to the distance between the lines, it can be judged whether the various lines are within the safe distance to ensure that they will not affect the stable power supply of the train's pantograph. In the current industry, there are generally two ways to measure the line spacing. One is to use a laser measuring instrument for manual measurement during on-site operations, that is, manually control the measuring instrument, control the laser to aim at the line and then make points, and the measuring instrument calculates the results based on the laser points. This method is greatly affected by human factors, and the laser point needs to be adjusted back and forth repeatedly. The laser point standards aimed by different people are not uniform, and only one line spacing can be measured at a time. The efficiency is extremely low and time-consuming. The other method is combined with image processing. The laser no longer uses a single point, but a fan-shaped scan to form a light spot on the contact line. Then the camera is used to take an image. In the image, the distance parameter is calculated based on the light spot. Although this method is more efficient than the first method, the imaging quality of the light spot will affect the subsequent calculation, and it is unknown whether the accuracy can be guaranteed. Summary of the invention

[0003] The purpose of the present invention is to provide a method for automatically measuring the spacing between crossing lines of a contact network, so as to solve the problem that the prior art cannot perform batch, rapid and automatic line spacing measurement.

[0004] The purpose of the present invention is achieved by the following technical scheme, which includes the following steps: step 1. parameter acquisition, reading and identifying measurement parameters according to an external measuring device, wherein the measurement parameters include railway contact network body data and contact network peripheral data, and the specific acquisition method is as follows: step 1-1. Relying on an external measuring device, the laser radar is used to scan the measurement line along the way to obtain point data of the contact network and the surrounding environment; step 1-2. The data measured in 1-1 is imported into the contact network measurement and analysis system, and the system will read the point data and form a three-dimensional point cloud to restore the measurement line scene; step 2. clue identification, according to The three-dimensional point cloud in step 1-2 uses the clue recognition algorithm to complete the grouping, calculation and judgment of the point cloud. First, the points are identified as line segments, and then the line segments are reorganized to form a complete long line, and then the clue data is stored by root; step 3. Line spacing calculation, mileage segmentation is performed according to the clue data obtained in step 2, and the line spacing between any two clues at each mileage segment is calculated; when all mileage segments are calculated, the spacing between the same two clues is taken as the minimum value for optimal storage; step 4. Supplement manual measurement, select any two points in the three-dimensional point cloud, and directly calculate the straight-line distance between the two points and the projection distance of each two-dimensional surface. Step 5. Repeat the above steps 1-4 until the measurement is completed.

[0005] It should be noted that, according to the above steps, batch calculations between high-speed rail contact network lines can be carried out efficiently, and at the same time, the calculation efficiency can be improved while the calculation accuracy can be improved.

[0006] The step 2 specifically includes the following contents: Step 2-1. Line segment fitting: group clue data by mileage, take a frame every 380mm-410mm, and judge the continuous points of the current frame and the frame data 550mm-620mm away from the current frame; use a random sampling algorithm to judge whether there is a line segment and save these line segment data to form a line segment set, wherein the algorithm uses the RANSAC algorithm, and the saved data includes the coordinate data of the starting point and the end point of the line segment, the coordinate data of the center point of the line segment, and the direction vector of the line segment; Step 2-2. Long-line fitting, long-line matching is performed according to the line segment set obtained in step 2-1. If the long-line matching process is passed, it means that the line segments to be determined belong to the same long line, and then they are integrated and saved. After the determination is completed, a long-line set of all mileages is obtained. The specific process of long-line matching is as follows: S1. Adjacent single line segments are matched according to matching rule 1 to obtain a longer line segment set, where matching rule 1 is as follows: the angle between the direction vectors of the line segments tends to 0; if it is not satisfied, it means that the next line segment may have an angle deviation, and the slope from the center point of the line segment to be determined to the end point of the line segment set to be determined is calculated. To obtain this deflection angle, it is necessary to be less than 4 degrees; S2. After S1, a set of longer line segments is obtained. For these sets, matching rule 2 is used to match the long line segments again to obtain longer long lines. Matching rule 2 is as follows: For line segments with gaps not exceeding 4.5 meters and the angles of the long line segment direction vectors approaching 0, combined features are used for comprehensive consideration, and the minimum eigenvalue is taken as the best match, where the combined features are: the slope of the long line segment connection and the projection distance of the long line segment on the XOY plane perpendicular to the track travel.

[0007] It should be noted that the use of matching rules and matching measurements can maximize the measurement accuracy and overcome the interference of many uncertain factors.

[0008] The step 3 specifically includes the following contents: step 3-1. Divide the mileage into segments, divide all the clues into segments according to the mileage, and divide them into segments every 100mm. At every mileage of 100mm, take out the intersection points of all the clues at that mileage and save them as the data set for line spacing calculation; step 3-2. Spacing calculation, according to the points obtained in step 3-1, according to their coordinates, use the Pythagorean theorem to calculate the spacing between each pair.

[0009] It should be noted that the application is based on multiple experimental improvements to obtain relevant measurement data. Based on the above data information, the data accuracy can be greatly improved.

[0010] The step 3 also specifically includes the following contents: step 3-3. retain the optimal value. After obtaining the distance between two points according to step 3-2, compare it with the distance between two corresponding long lines calculated by the mileage segmentation points in step 3-1, save the smaller value, and discard the larger one; step 3-4. after all mileage calculations are completed, obtain the shortest line distance between all clues, and then perform data selection for the user, and the three-dimensional point cloud will automatically focus on the shortest distance.

[0011] It should be noted that the above trade-off rules were obtained by the applicant after a long period of extensive experiments and can control the accuracy of measurement to the greatest extent.

[0012] The step 4 specifically includes the following contents: step 4-1. Select any two points in the three-dimensional point cloud: select "point selection mode" in the contact network measurement and analysis system in step 1-2, and select any two points; step 4-2. Automatically calculate the distance between the two points. After selecting the two points, the contact network measurement and analysis system will convert them into coordinate points in three-dimensional space, and calculate the distance between the two points in three-dimensional space and the projection distance in each two-dimensional plane based on the coordinate values ​​and the Pythagorean theorem.

[0013] It should be noted that the mode is selected according to actual measurements, and the prior art generally adopts a fixed mode corresponding selection, which cannot improve the adaptability.

[0014] The step 5 specifically includes the following contents: according to the operations of step 1 to step 4, the operations are repeated, and the operation data parameters are measured and recorded one by one.

[0015] It should be noted that implementing one measurement and one record can improve the stability of the overall operating parameters, while helping to improve the convenience of recording parameters.

[0016] Compared with the prior art, the present invention has the following advantages and beneficial effects: without using additional measuring equipment to perform individual distance measurements on the line, the shortest distances between all clues on the entire line can be calculated in batches at one time. The clues are identified by a clue recognition algorithm, and the optimal points of the clues are selected to calculate the spacing. Compared with the traditional manual measurement of spacing, it not only greatly saves time and manpower, but also eliminates the influence of human factors. The measurement results are accurate and fast. At the same time, it supplements manual measurement. Any two points in the three-dimensional point cloud scene can be selected to view their straight-line distance and the projection distance of each two-dimensional surface, meeting various requirements for contact network clue spacing. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is the overall flow chart of the present invention;

[0018] Figure 2 is a clue identification flow chart of the present invention;

[0019] Figure 3 It is a flow chart of line spacing calculation of the present invention. DETAILED DESCRIPTION

[0020] Please refer to the attached instructions Figure 1-Figure 3 This embodiment provides a method for automatically measuring the spacing between cross-wires of a contact network. The method for automatically measuring the spacing between cross-wires of a contact network is mainly used to solve the problem that the prior art cannot perform batch, fast and automatic line spacing measurement. The method for automatically measuring the spacing between cross-wires of a contact network is already in the actual use stage.

[0021] The specific implementation methods of this application are as follows:

[0022] Example 1

[0023] Reference Notes Figure 1 , the present application includes the following steps, step 1. parameter acquisition, reading and identifying the measurement parameters according to the external measurement device, it should be noted that the preferred external device of this embodiment is an intelligent dynamic parameter detection device for the contact network line, wherein the measurement parameters include the railway contact network body data and the contact network surrounding data, and the specific acquisition method is as follows: step 1-1. Relying on the external measurement device, the laser radar is used to scan the measurement line along the way to obtain the point data of the contact network and the surrounding environment; step 1-2. The data measured in 1-1 is imported into the contact network measurement and analysis system, the system will read the point data and form a three-dimensional point cloud to restore the measurement line scene; step Step 2. Clue identification: Based on the three-dimensional point cloud in step 1-2, use the clue identification algorithm to complete the grouping, calculation and judgment of the point cloud. First, identify the points as line segments, then reorganize the line segments to form a complete long line, and then store the clue data by root; Step 3. Line spacing calculation: Mileage segmentation is performed according to the clue data obtained in step 2, and the line spacing between any two clues at each mileage segment is calculated; when all mileage segments are calculated, the spacing between the same two clues is taken as the minimum value for optimal storage; Step 4. Supplement manual measurement, select any two points in the three-dimensional point cloud, and directly calculate the straight-line distance between the two points and the projection distance of each two-dimensional surface. Step 5. Repeat the above steps 1-4 until the measurement is completed.

[0024] Example 2

[0025] Reference Notes Figure 2 , 3 Another preferred implementation of this embodiment is to use line segment fitting, group the clue data by mileage, take a frame every 380mm-410mm, where the preferred value in this embodiment is 400mm, and judge the continuous points of the current frame and the frame data 550mm-620mm away from the current frame, where the preferred value in this embodiment is 600mm.

[0026] A random sampling algorithm is used for these points to determine whether there are line segments and save the line segment data to form a line segment set. The preferred algorithm in this embodiment is the RANSAC algorithm. Of course, the Sherwood elimination algorithm, Las Vegas algorithm, and Monte Carlo algorithm can also be used. The RANSAC algorithm can maximize the accuracy and reduce errors.

[0027] The data then saved include the coordinate data of the starting point and the ending point of the line segment, the coordinate data of the center point of the line segment, and the direction vector of the line segment; long line fitting, long line matching is performed based on the obtained line segment set. If the long line matching process is passed, it means that the line segments to be determined belong to the same long line, and then they are integrated and saved. After the determination is completed, a long line set of all mileage is obtained, and the specific process of long line matching is as follows: S1. Adjacent single line segments are matched according to matching rule 1 to obtain a longer line segment set, and matching rule 1 is as follows: the angle between the direction vectors of the line segments tends to 0; if it is not satisfied, it means that the next line segment may have an angle deviation, and the slope from the center point of the line segment to be determined to the end point of the line segment set to be determined is calculated. To obtain this deflection angle, it is necessary to be less than 4 degrees; S2. After S1, a set of longer line segments is obtained. For these sets, matching rule 2 is used to match the long line segments again to obtain longer long lines. Matching rule 2 is as follows: For line segments with gaps not exceeding 4.5 meters and the angles of the long line segment direction vectors approaching 0, combined features are used for comprehensive consideration, and the minimum eigenvalue is taken as the best match, where the combined features are: the slope of the long line segment connection and the projection distance of the long line segment on the XOY plane perpendicular to the track travel.

[0028] In the prior art, a single matching rule is generally used for judgment. This judgment method cannot make judgments with high accuracy. The reason is that a single match can only make judgments on a single factor. At this time, the single factor belongs to a single angle judgment in the eyes of technical personnel in this field. It cannot guarantee the comprehensiveness of the data factor judgment and is bound to have low accuracy.

[0029] Example 3

[0030] Reference Notes Figure 2 , 3 Another preferred implementation of the present application is to segment the mileage, and segment all the clues according to the mileage, with each 100mm as a segment, and take out the intersection of all the clues at the mileage at each 100mm mileage. It should be noted that the selection of 100mm as the segmentation is the best segmentation interval obtained by the applicant after a lot of practical work experience. The segmentation interval of 100mm can better avoid the measurement error of the remaining measurement distance in the later stage of the measurement process. Save it as a data set for line spacing calculation;

[0031] Then the spacing calculation is performed. According to the points obtained in the steps, the spacing between the two points is calculated based on their coordinates using the Pythagorean theorem, and the optimal value is retained. After the spacing between the two points is obtained, it is compared with the spacing between the corresponding long lines calculated from the mileage segmentation points. The smaller value is saved and the larger value is discarded. After all the mileage calculations are completed, the shortest line spacing between all the clues is obtained, and then data selection is performed for the user. The three-dimensional point cloud will automatically focus on the shortest distance. Select any two points in the three-dimensional point cloud: select "point selection mode" in the contact network measurement and analysis system in step 1-2, and select any two points; automatically calculate the distance between the two points. After selecting the two points, the contact network measurement and analysis system will convert them into coordinate points in three-dimensional space. According to the coordinate values ​​and the Pythagorean theorem, the distance between the two points in three-dimensional space and the projection distance in each two-dimensional plane are calculated.

[0032] According to the above measurement steps, the applicant obtains the following analysis compared with the prior art:

[0033]

[0034] The above data can intuitively show that the measurement method and related matching principles of the present application can ensure the high efficiency and accuracy of the contact network measurement to the greatest extent, achieve high measurement efficiency, improve the accuracy of the overall operation, and thus realize batch, fast and automatic line spacing measurement.

[0035] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for automatically measuring the spacing between cross-cable lines of a contact network, characterized in that: The following steps are included Step 1. Parameter acquisition: Read and identify the measurement parameters according to the external measurement device, where the measurement parameters include the railway contact network body data and the contact network surrounding data. The specific acquisition method is as follows: Step 1-1. Relying on an external measuring device, the laser radar is used to scan the measuring line along the way to obtain point data of the contact network and the surrounding environment; Step 1-2. Import the data measured in 1-1 into the contact network measurement and analysis system. The system will read the point data and form a three-dimensional point cloud to restore the measurement line scene; Step 2. Clue recognition: Based on the three-dimensional point cloud in step 1-2, the clue recognition algorithm is used to complete the grouping, calculation and judgment of the point cloud. The points are first recognized as line segments, and then the line segments are reorganized to form a complete long line. The clue data is then stored by root. Step 3. Calculate the line spacing. Divide the mileage into segments according to the clue data obtained in step 2, and calculate the line spacing between any two clues at each mileage segment. When all mileage segments are calculated, the minimum spacing between the same two clues is taken as the optimal value. Step 4. Supplement manual measurement, select any two points in the 3D point cloud, and directly calculate the straight-line distance between the two points and the projection distance of each 2D surface; Step 5. Repeat the above steps 1 to 4 until the measurement is completed; The step 2 specifically includes the following contents: Step 2-1. Line segment fitting: group clue data by mileage, take a frame every 380mm-410mm, and judge the continuous points of the current frame and the frame data 550mm-620mm away from the current frame; A random sampling algorithm is used for these points to determine whether there is a line segment and save the line segment data to form a line segment set. The algorithm uses the RANSAC algorithm. The saved data includes the coordinate data of the starting point and the end point of the line segment, the coordinate data of the center point of the line segment, and the direction vector of the line segment. Step 2-2. Long-line fitting: Long-line matching is performed based on the line segment set obtained in step 2-1. If the long-line matching process is passed, it means that the line segments to be determined belong to the same long line, and then they are integrated and saved. After the determination is completed, a long-line set of all mileages is obtained. The specific process of long-line matching is as follows: S1. Adjacent single line segments are matched according to the matching rule 1 to obtain a longer line segment set, where the matching rule 1 is as follows: the angle between the direction vectors of the line segments tends to 0; if it is not satisfied, it means that the next line segment may have an angle deviation, and the slope from the center point of the line segment to be determined to the end point of the line segment set to be determined is calculated, and the deflection angle needs to be less than 4 degrees; S2. After S1, a set of longer line segments is obtained. For these sets, matching rule 2 is used to match the long line segments again to obtain longer long lines. Matching rule 2 is as follows: for line segments with gaps of no more than 4.5 meters and the angle of the direction vector of the long line segment approaching 0, combined features are used for comprehensive consideration, and the minimum eigenvalue is taken as the best match, where the combined features are: the slope of the long line segment connection and the projection distance of the long line segment on the XOY plane perpendicular to the track travel; The step 3 specifically includes the following contents: Step 3-1. Divide all clues into segments according to the mileage, with each segment being 100 mm. At each mileage of 100 mm, take out the intersection points of all clues at that mileage and save them as the data set for line spacing calculation; Step 3-2. Spacing calculation: Based on the points obtained in step 3-1, use the Pythagorean theorem to calculate the spacing between the points according to their coordinates.

2. A method for automatically measuring the spacing between cross-cord wires of a contact network according to claim 1, characterized in that: The step 3 also specifically includes the following contents: Step 3-3. Keep the optimal value. After obtaining the distance between two points according to step 3-2, compare it with the distance between two corresponding long lines calculated by the mileage segmentation points in step 3-1, save the smaller value, and discard the larger one; Step 3-4. After all mileage calculations are completed, the shortest line distances between all clues are obtained, and then data selection is performed for the user, and the 3D point cloud will automatically focus on the shortest distance.

3. The method for automatically measuring the spacing between cross-cords of a contact network according to claim 1, characterized in that: The step 4 specifically includes the following contents: Step 4-1. Select any two points in the 3D point cloud: Select "point selection mode" in the contact network measurement and analysis system in step 1-2, and select any two points; Step 4-2. Automatically calculate the distance between two points. After selecting two points, the contact network measurement and analysis system will convert them into coordinate points in three-dimensional space. According to the coordinate values ​​and the Pythagorean theorem, the distance between the two points in three-dimensional space and the projection distance in each two-dimensional plane are calculated.

4. The method for automatically measuring the spacing between cross-cords of a contact network according to claim 1, characterized in that: The step 5 specifically includes the following contents: according to the operations of step 1 to step 4, the operations are repeated, and the operation data parameters are measured and recorded one by one.

Images