CN-116026249-B - Real-time detection method and device for deformation of empty rail track beam

CN116026249BCN 116026249 BCN116026249 BCN 116026249BCN-116026249-B

Abstract

The embodiment of the application provides a method, a device, a storage medium and electronic equipment for detecting deformation of an empty track rail beam in real time, wherein the method comprises the steps of acquiring laser point cloud data generated by scanning the rail beam at a target position by a laser radar as original laser point cloud data; and detecting deformation of the track beam at the target position through a pre-constructed sample parameter based on the internal structure parameter, and recording a detection log. The technical scheme of the embodiment of the application can realize accurate real-time detection of the deformation of the empty rail track beam and realize remote operation and maintenance of the empty rail track beam.

Inventors

WANG WEI
SU LIJIE
CHEN ZHIGUO
WANG ZISHUN
LIU WEI

Assignees

中车长江运输设备集团有限公司

Dates

Publication Date: 20260512
Application Date: 20230113

Claims (8)

1. The method for detecting the deformation of the empty rail track beam in real time is characterized by comprising the following steps of: acquiring laser point cloud data generated by scanning a track beam at a target position by a laser radar, and taking the laser point cloud data as original laser point cloud data; The method comprises the steps of obtaining original laser point cloud data, calculating and extracting internal structural parameters of a track beam at a target position based on the original laser point cloud data, wherein the original laser point cloud data consists of laser data of a plurality of laser points, calculating and extracting the internal structural parameters of the track beam at the target position based on the original laser point cloud data, wherein the method comprises the steps of replacing noise laser data in the original laser point cloud data to obtain reference laser point cloud data, carrying out smoothing processing on the reference laser point cloud data to obtain target laser point cloud data, calculating and extracting the internal structural parameters of the track beam at the target position based on the target laser point cloud data, wherein the method comprises the steps of determining laser break points in the plurality of laser points based on the target laser point cloud data, dividing the plurality of laser points into a plurality of laser point sets through the laser break points, respectively generating characteristic lines based on the laser points in each laser point set, and obtaining characteristic line sets by fitting, and determining the characteristic line sets by the characteristic lines, wherein the characteristic line sets are used for characterizing the laser line sets in the radial direction of the track beam profile of the laser points; and detecting deformation of the track beam at the target position through a pre-constructed sample parameter based on the internal structure parameter, and recording a detection log.
2. The method of claim 1, wherein the laser data comprises distances between the number of laser points and the lidar, and wherein the replacing noise laser data in the original laser point cloud data comprises: for each target laser point, acquiring a distance corresponding to the target laser point, wherein the target laser point is any one laser point of the plurality of laser points, and taking the distance as a target distance; And if the target distance is greater than a first preset distance or less than a second preset distance, defining laser data corresponding to the target laser point as the noise laser data, and replacing the noise laser data by the laser data corresponding to the laser point adjacent to the target laser point, wherein the first preset distance is greater than the second preset distance.
3. The method of claim 1, wherein smoothing the reference laser point cloud data comprises: performing sliding segmentation on the reference laser point cloud data according to a preset window to obtain a plurality of groups of sub-reference laser point cloud data; Calculating smooth filtering laser data based on each laser data in each group of target sub-reference laser point cloud data, wherein the smooth filtering laser data is used for inhibiting fluctuation of each laser data in the target sub-reference laser point cloud data, and the target sub-reference laser point cloud data is any one group in the multiple groups of sub-reference laser point cloud data; for each set of target sub-reference laser point cloud data, replacing individual ones of the target sub-reference laser point cloud data by the smoothed filtered laser data.
4. The method of claim 1, wherein the determining a laser breakpoint among the number of laser points based on the target laser point cloud data comprises: selecting any two adjacent laser points from the plurality of laser points; Acquiring the distance corresponding to any two adjacent laser points based on the target laser point cloud data; And if the difference value of the distances corresponding to the adjacent arbitrary two laser points is larger than a preset parameterized self-adaptive threshold value, determining the adjacent arbitrary two laser points as the laser breakpoint.
5. The method of claim 1, wherein generating laser feature lines based on the laser points in each set of laser points by fitting respectively comprises: selecting a plurality of laser points from each target laser point set as seed laser points, wherein the target laser point set is any one of the plurality of laser point sets; and performing straight line fitting on seed laser points in the target laser point set to generate laser characteristic lines corresponding to the target laser point set.
6. An empty rail track beam deformation detection device, the device comprising: The acquisition unit is used for acquiring laser point cloud data generated by scanning the track beam at the target position by the laser radar, and the laser point cloud data is used as original laser point cloud data; A data processing unit for computing and extracting internal structural parameters of the track beam at the target position based on the original laser point cloud data; the original laser point cloud data comprises laser data of a plurality of laser points, the internal structural parameters of the track beam at the target position are calculated and extracted based on the original laser point cloud data, the method comprises the steps of replacing noise laser data in the original laser point cloud data to obtain reference laser point cloud data, conducting smoothing processing on the reference laser point cloud data to obtain target laser point cloud data, calculating and extracting the internal structural parameters of the track beam at the target position according to the target laser point cloud data, wherein the internal structural parameters of the track beam at the target position are calculated and extracted according to the target laser point cloud data, determining laser break points in the plurality of laser points based on the target laser point cloud data, dividing the plurality of laser points into a plurality of laser point sets based on the laser break points, respectively fitting and generating laser characteristic lines based on the laser points in each laser point set to obtain laser characteristic line sets, and the laser characteristic line sets are used for representing the profile of the track beam at the target position in the radial direction; And the detection recording unit is used for detecting the deformation of the track beam at the target position through a pre-constructed sample parameter based on the internal structure parameter and recording a detection log.
7. A computer readable storage medium having stored therein at least one program code loaded and executed by a processor to implement operations performed by the method of any of claims 1 to 5.
8. An electronic device comprising a memory, one or more processors, and one or more programs, wherein one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing the method of any of claims 1-5.

Description

Real-time detection method and device for deformation of empty rail track beam Technical Field The application relates to the technical field of track detection, in particular to a method and a device for detecting deformation of an empty track beam in real time, a storage medium and electronic equipment. Background With the running of an aerial track in China, such as the running of an demonstration line of a collection and delivery system in a Qingdao port, the daily operation and maintenance inspection work of the interior of a high overhead track is carried out, because the settlement, the deformation, the groove clearance deformation and the like of the aerial track rail beam can be caused under the long-term load running of a vehicle, if the deformation of the aerial track rail beam is not timely processed, the potential safety hazard of the vehicle running can be generated, but because of the high pressure in the interior of the aerial track, the wide groove and the dark interior environment bring great safety risks and work load intensity to daily maintenance personnel of the aerial track. Based on the above, how to realize intelligent and accurate detection of the deformation of the empty rail track beam and realize the remote operation and maintenance of the empty rail track beam are technical problems to be solved urgently. Disclosure of Invention The embodiment of the application provides a real-time detection method and device for deformation of an empty rail track beam, a storage medium and electronic equipment, so that intelligent accurate detection of the deformation of the empty rail track beam can be realized, and meanwhile, remote operation and maintenance of the empty rail track beam can be realized. Other features and advantages of the application will be apparent from the following detailed description, or may be learned by the practice of the application. According to a first aspect of an embodiment of the application, a real-time detection method for deformation of an empty track rail beam is provided, and the method comprises the steps of obtaining laser point cloud data generated by a laser radar for scanning the rail beam at a target position as original laser point cloud data, calculating and extracting internal structural parameters of the rail beam at the target position based on the original laser point cloud data, and detecting the deformation of the rail beam at the target position through pre-constructed sample parameters based on the internal structural parameters. In some embodiments of the present application, based on the foregoing scheme, the original laser point cloud data is composed of laser data of a plurality of laser points, and the calculating and extracting internal structural parameters of the track beam at the target position based on the original laser point cloud data includes replacing noise laser data in the original laser point cloud data to obtain reference laser point cloud data, performing smoothing processing on the reference laser point cloud data to obtain target laser point cloud data, and calculating and extracting internal structural parameters of the track beam at the target position according to the target laser point cloud data. In some embodiments of the present application, based on the foregoing solution, the replacing noise laser data in the original laser point cloud data includes, for each target laser point, obtaining a distance corresponding to the target laser point, as a target distance, where the target laser point is any one of the plurality of laser points, defining the laser data corresponding to the target laser point as the noise laser data if the target distance is greater than a first preset distance or less than a second preset distance, and replacing the noise laser data with the laser data corresponding to a laser point adjacent to the target laser point, where the first preset distance is greater than the second preset distance. In some embodiments of the present application, based on the foregoing scheme, the smoothing processing is performed on the reference laser point cloud data, which includes performing sliding segmentation on the reference laser point cloud data according to a preset window to obtain a plurality of groups of sub-reference laser point cloud data, calculating, based on each laser data in each group of target sub-reference laser point cloud data, smoothing filter laser data, where the smoothing filter laser data is used to suppress fluctuation of each laser data in the target sub-reference laser point cloud data, the target sub-reference laser point cloud data is any one of the groups of sub-reference laser point cloud data, and replacing each laser data in the target sub-reference laser point cloud data by the smoothing filter laser data for each group of target sub-reference laser point cloud data. In some embodiments of the present application, based on the foregoing solution, the computing and ex