| The mine lifting system plays a vital role in connecting the underground and ground,and the rigid cage guide serves as its primary component.However,due to complex environmental factors such as impact vibration and geological changes,misalignment,clearance,and tilt are common defects that occur during service,severely affecting the smoothness and reliability of container operation.Thus,effectively detecting the status of the rigid cage guide is essential for the safe operation of the lifting system.This article aims to address the current issues regarding the low intelligent detection level,poor inspection efficiency and accuracy,and inability to fully reflect the defect status of the mine rigid cage guide.To this end,a lightweight autonomous inspection robot has been developed,and a defect detection method for the rigid cage guide based on multi-sensor collaboration has been proposed.A research on the status detection system for the mine rigid cage guide based on inspection robots has been carried out,with the following main tasks completed:(1)Design the overall scheme of the detection system.Based on the analysis of the working condition and inspection requirements of the rigid cage guide,the overall scheme of the rigid cage guide status detection system based on the inspection robot is designed,including the body structure,hardware framework and software architecture of the inspection robot.Based on the performance indicators of the robot,the key hardware of the robot was selected,and the software control architecture of the robot was designed based on the motion control requirements of the inspection.(2)Design and analysis of inspection robot system.The statics characteristics of the robot are analyzed,and the mechanical structure of the robot is designed in detail,including the adsorption device,the guiding mechanism,the drive transmission device,the explosionproof structure,and the obstacle avoidance device.Through ANSYS finite element analysis,the mechanical properties of key components in the robot structure were analyzed,including the strength of the transmission bevel gear and the explosion-proof characteristics of each cavity.The influence of the size of the magnetic wheel ring core on magnetic force was analyzed using COMSOL software,and the correctness of the static analysis results was verified.(3)Research on a method for detecting the state of a rigid cage guide using multisensor collaboration.Analyzed various defect states of the rigid cage guide and classified them according to structural defects(local defects,overall defects)and surface defects.On this basis,the attitude detection method and the visual detection method of the rigid cage guide are studied respectively to achieve efficient detection of various defects of the rigid cage guide.In the aspect of attitude detection,the kinematics process of the robot wheel going up and down the steps is analyzed,so as to establish a mathematical model to realize the characterization of local structural defects,and the overall structural defect detection method based on track estimation is studied.In terms of visual detection,multiple object detection algorithms including SSD,YOLOv4,and YOLOv5 were compared and optimized.The results showed that YOLOv5 has better detection accuracy and efficiency and is suitable for visual detection of local structural and surface defects in the rigid cage guide in this article.A multi sensor collaborative inspection strategy for rigid cage guide based on inspection robots is proposed,combining pose detection and visual detection methods.(4)Development and experimental research of prototype inspection robot system.The prototype of the inspection robot is developed,and the upper computer of the system is developed using Qt.The experimental platform is built in the laboratory environment,and the functional test of the detection system is carried out through it.Besides,the mine field experiment is carried out.The results show that the inclination detection results in the laboratory posture experiment can accurately reflect the defect characteristics of the tested rigid cage guide.The visual detection method based on YOLOv5 can accurately identify joint gaps joint misalignment and surface defects,and in on-site experiments,the inspection robot can accurately reflect the health status of the rigid cage guide.There are 76 pictures,17 tables and 98 references. |
PDF Full Text Request