Development Of Wire Rope Defect Detection System For Mine Hoist

The wire rope of mine hoist is an important component of the hoisting system.During the service of the wire rope of mine hoist,regular defect detection of the wire rope is related to the safe production of coal mine.Because the mine hoist wire rope is in a dark,humid and dusty environment for a long time,it is easy to produce broken wires caused by corrosion.Frequent start-stop of wire rope is affected by variable load,fatigue,wear and so on,resulting in irreversible mechanical loss of wire rope,which is easy to cause downhole safety hazards.The health status of mine hoist wire rope directly affects the normal operation of the hoisting system.Based on the defect detection of mine hoist wire rope,this paper studies the image acquisition method of wire rope,the defect detection method of wire rope and the practicability verification of detection system.Complete the hardware test platform and write the software of wire rope defect detection system.The main work of this paper is as follows :(1)According to the defect detection requirements of mine hoist wire rope,the composition structure of wire rope,the classification of wire rope defects and the methods and standards of scrapping are analyzed and introduced.A wire rope defect detection system for mine hoist based on machine vision and deep learning was designed and developed.The image acquisition unit is composed of high-speed camera and double telecentric lens,and the image of wire rope under normal operation is successfully obtained.(2)The image collected by the system is detected by wire rope geometry.The Vision Assistant is used to preprocess the collected wire rope image,then the template matching and positioning of the wire rope are performed,and finally the edge detection of the wire rope is performed to complete the measurement of the wire rope geometry.(3)Due to the particularity of wire rope structure,it is difficult to detect the broken wire of wire rope by traditional visual detection.The design is based on YOLOV5 to realize the function of wire rope broken wire detection.From the production of data sets,setting training parameters for training,to the analysis of results to verify its effectiveness,the detection of broken wire defects on the surface of wire rope is completed.(4)The system is tested and analyzed by using the wire rope visual inspection test bench built in the laboratory.The results show that the system can detect the wire rope defects well and verify the feasibility of the system.Figure 58 Table 12 Reference 81...