CN-115641319-B - Magnetic levitation train current collector detection method, device, computer equipment and storage medium

CN115641319BCN 115641319 BCN115641319 BCN 115641319BCN-115641319-B

Abstract

The application relates to a method and a device for detecting a current collector of a maglev train, computer equipment and a storage medium. The method comprises the steps of obtaining a plurality of original point cloud sets and a plurality of target structure images corresponding to a plurality of structures to be detected in a magnetic levitation train current collector, dividing the corresponding original point cloud sets by utilizing a point cloud model after each spatial transformation to obtain corresponding optimized point cloud sets, searching two edge point cloud sets along the running direction of the vertical magnetic levitation train in each optimized point cloud set to obtain a first straight line corresponding to each optimized point cloud set, searching edge pixel positions along the running direction of the vertical magnetic levitation train in each target structure image to obtain a second straight line, obtaining a slope threshold range according to the slope of the second straight line and a corresponding threshold value, and detecting whether the corresponding structure to be detected of the current collector is abnormal or not according to the size relation between the slope of the first straight line and the slope threshold value range. The method can improve the detection efficiency of the current collector, and has high detection precision and stable detection effect.

Inventors

ZHOU WENWU
ZHAO QINGLIN
JIANG TING
ZHANG CHENG
LI WEIJIN
ZHAO LONG
YU YANG
HOU SHIHAO
ZHANG XINGHUA

Assignees

湖南凌翔磁浮科技有限责任公司

Dates

Publication Date: 20260512
Application Date: 20221031

Claims (10)

1. The method for detecting the current collector of the maglev train is characterized by comprising the following steps of: Acquiring a plurality of original point cloud sets and a plurality of target structure images corresponding to a plurality of structures to be detected in a magnetic levitation train current collector; Carrying out point cloud registration on each original point cloud set and a point cloud model of a corresponding structure to be detected to obtain a corresponding point cloud model after space transformation, and dividing the corresponding original point cloud set by utilizing each point cloud model after space transformation to obtain a corresponding optimized point cloud set; Searching two edge point cloud sets along the running direction of the vertical maglev train in each optimized point cloud set, obtaining a first straight line corresponding to each optimized point cloud set according to edge point cloud coordinates in the two edge point cloud sets, searching an edge pixel position along the running direction of the vertical maglev train in each target structure image, and obtaining a second straight line corresponding to the target structure image according to the edge pixel position; And according to the second linear slope corresponding to the second linear and the corresponding threshold value, obtaining a slope threshold value range, and according to the magnitude relation between the first linear slope corresponding to each optimized point cloud set and the slope threshold value range, detecting whether the structures to be detected of the current collector are abnormal or not.
2. The method of claim 1, wherein the step of obtaining a plurality of original point cloud sets corresponding to a plurality of structures to be detected in the maglev train current collector comprises: The method comprises the steps of obtaining a current collector point cloud corresponding to a magnetic levitation train current collector, wherein the magnetic levitation train current collector comprises a plurality of target structures to be detected; And dividing the current collector point cloud by using a preset dividing line to obtain a plurality of original point cloud sets respectively containing the plurality of structures to be detected.
3. The method according to claim 2, wherein the step of dividing the current collector point cloud by a preset dividing line to obtain a plurality of original point cloud sets respectively including the plurality of structures to be detected includes: Constructing a space rectangular coordinate system, wherein the X-axis direction of the space rectangular coordinate system is perpendicular to the running direction of the magnetic levitation train, and the Y-axis direction of the space rectangular coordinate system is parallel to the running direction of the magnetic levitation train; Acquiring edge point clouds of the current collector point clouds in the X-axis direction, and acquiring two dividing lines according to edge point cloud coordinates in the two edge point cloud sets, physical parameters of structures of the current collector and a space topological relation; and dividing the point cloud of the current collector by adopting the two dividing lines to obtain a first original point cloud set, a second original point cloud set and a third original point cloud set which respectively contain the structures to be detected.
4. The method of claim 1, further comprising, prior to segmenting the corresponding original point cloud set using each of the spatially transformed point cloud models to obtain a corresponding optimized point cloud set: calculating the coincidence degree between each original point cloud set and the corresponding registered point cloud set, and outputting a detection result corresponding to the structure to be detected if the coincidence degree is not in a preset coincidence degree threshold range; If the coincidence degree is within a preset coincidence degree threshold range, calculating a coincidence region mean value between each original point cloud set and the corresponding registered point cloud set, and if the coincidence region mean value is not within the preset mean value threshold range, outputting a detection result corresponding to a structure to be detected; If the average value of the overlapping areas is within a preset average value threshold range, calculating the overlapping area variance between each original point cloud set and the corresponding registered point cloud set, and if the overlapping area variance is not within the preset variance threshold range, outputting a detection result corresponding to the structure to be detected, wherein the detection result is abnormal.
5. The method of claim 1, wherein the step of segmenting the corresponding original point cloud set using each spatially transformed point cloud model to obtain a corresponding optimized point cloud set comprises: And dividing the corresponding original point cloud set according to the boundary information of each point cloud model after the space transformation, extracting the point clouds of the overlapping area, clustering the point clouds of the overlapping area, and removing the point cloud clusters with small quantity in the adjacent point clouds to obtain the corresponding optimized point cloud set.
6. The method of claim 1, wherein the step of acquiring a plurality of target structure images corresponding to the plurality of structures to be detected comprises: acquiring an image to be detected of a magnetic levitation train current collector; and processing the image to be detected by adopting an outline matching method to obtain a plurality of target structure images corresponding to the plurality of structures to be detected.
7. The method of claim 6, wherein the method further comprises: Acquiring the train speed and the fluctuation quantity of the vertical train running direction when the magnetic levitation train runs at each moment; judging whether the image to be detected is smeared or not according to the magnitude relation between the fluctuation quantity and the image precision corresponding to the fluctuation time; When the image has smear, the exposure time of the current camera is adjusted according to the train running speed at the current moment, and the image to be detected is updated according to the adjusted exposure time.
8. A magnetic levitation train current collector detection device, the device comprising: The data acquisition module is used for acquiring a plurality of original point cloud sets and a plurality of target structure images corresponding to a plurality of structures to be detected in the maglev train current collector; The point cloud optimization module is used for carrying out point cloud registration on each original point cloud set and the point cloud model of the corresponding structure to be detected to obtain a corresponding point cloud model after spatial transformation, and dividing the corresponding original point cloud set by utilizing each point cloud model after spatial transformation to obtain a corresponding optimized point cloud set; The linear fitting module is used for searching two edge point cloud sets in each optimized point cloud set along the running direction of the vertical magnetic levitation train, obtaining a first line corresponding to each optimized point cloud set according to the edge point cloud coordinates in the two edge point cloud sets, searching the edge pixel position in each target structure image along the running direction of the vertical magnetic levitation train, and obtaining a second line corresponding to the target structure image according to the edge pixel position; the structure detection module is used for obtaining a slope threshold range according to the slope of the second straight line corresponding to the second straight line and a corresponding threshold value, and detecting whether the structures to be detected of the current collector are abnormal according to the magnitude relation between the slope of the first straight line corresponding to each optimized point cloud set and the slope threshold value range.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.

Description

Magnetic levitation train current collector detection method, device, computer equipment and storage medium Technical Field The application relates to the technical field of detection of magnetic levitation train current collectors, in particular to a method, a device, computer equipment and a storage medium for detecting a magnetic levitation train current collector. Background The medium-low speed magnetic levitation train is used as a new generation of transportation means, has the advantages of high speed, strong adaptability to terrains, flexible line selection, safety, environmental protection and the like, and the medium-low speed magnetic levitation train current collector system consists of a current collector arranged below a levitation frame and a contact rail of a rail Liang Yaobu, and a current collector sliding plate carries out electric energy transmission and reception in a sliding contact manner with the contact rail. The long sand magnetic levitation fast line and the clear magnetic levitation are of a three-rail contact net current-receiving type, and the three-rail contact net is respectively an upper contact type, a lower contact type and a side contact type according to different relative positions of a current collector and a contact rail. The current collector adopting the side current collector is relatively simple, the installation structure is compact, the current collector cannot exceed the stipulated limit of a vehicle, the current collector adopts a hinged structure and is installed on the installation frame on the side face of the bracket, high pressure is sent to the magnetic levitation train through mechanical sliding contact with the power supply rail, and the power supply of the middle-low speed magnetic levitation train is realized, so that the integrity of the current collector can directly influence the power supply system of the magnetic levitation train, and the operation safety of the train is further influenced. The method has very important significance for improving the maintenance efficiency and reducing the on-line fault rate of the train through detecting integrity indexes such as defect, deformation, displacement and the like of the current collector and detecting the thickness of the carbon sliding plate. Three-dimensional detection techniques based on point cloud registration have been widely used in the field of industrial manufacturing. Taking the application of the method to the surface quality detection of the part as an example, the three-dimensional detection technology is realized by registering the scanning point cloud of the target part with a corresponding standard model, and then performing contrast evaluation on the difference between the two point clouds. However, under the influence of the operation of the magnetic levitation train, the installation position of the acquisition device and the complex hinged hollow structure of the current collector, the acquisition of the complete point cloud data of the current collector is difficult, and only the point cloud of the current collector in a certain range can be acquired, so that the point cloud registration accuracy and stability are affected. Disclosure of Invention In view of the foregoing, it is desirable to provide a method, an apparatus, a computer device, and a storage medium for detecting a magnetically levitated train current collector, which can improve the accuracy of detecting magnetically levitated train current collectors. A method for detecting a magnetically levitated train current collector, the method comprising: Acquiring a plurality of original point cloud sets and a plurality of target structure images corresponding to a plurality of structures to be detected in a magnetic levitation train current collector; Carrying out point cloud registration on each original point cloud set and a point cloud model of a corresponding structure to be detected to obtain a corresponding point cloud model after space transformation, and dividing the corresponding original point cloud set by utilizing each point cloud model after space transformation to obtain a corresponding optimized point cloud set; Searching two edge point cloud sets along the running direction of the vertical maglev train in each optimized point cloud set, obtaining a first straight line corresponding to each optimized point cloud set according to edge point cloud coordinates in the two edge point cloud sets, searching an edge pixel position along the running direction of the vertical maglev train in each target structure image, and obtaining a second straight line corresponding to the target structure image according to the edge pixel position; And according to the second linear slope corresponding to the second linear and the corresponding threshold value, obtaining a slope threshold value range, and according to the magnitude relation between the first linear slope corresponding to each optimized point cloud set and the