Research On Intelligent Detection Device And Motion System For High-speed Railway Wheels

The defect detection of high-speed rail wheels in the manufacturing process is the most important part to ensure the safe driving of high-speed trains.In the past,the detection method relied on simple mechanical devices to fix laser ultrasonic probes and rotate high-speed rail wheels to detect internal defects,or be carried by workers.Testing equipment,handheld ultrasonic testing probe to scan,by observing the display screen to artificially judge whether there are defects.Conventional ultrasonic inspection methods are restricted by the probe,and the scanning angle is limited.At the same time,the coverage angle of the ultrasonic beam emitted by the probe is small,and the scanning efficiency is low,which cannot meet the needs of fast scanning and accurate flaw detection in wheel production.In recent years,with the continuous development of robotics technology,automated and intelligent detection systems have gradually become one of the research hotspots.At present,there are still relatively few objects that can realize automatic non-destructive testing,mainly some simple-shaped workpieces such as planes and cylinders or components of a certain specific shape,and some high-end automatic testing equipment mainly depends on imports,which greatly increases the production cost.Therefore,this article combines robotics,ultrasonic phased array inspection technology and visual recognition technology to research and analyze the intelligent inspection system in the manufacturing process of high-speed rail wheels:(1)According to the inspection requirements of high-speed rail wheels,the inspection system was designed with a six-axis robotic arm,AOS ultrasonic phased array platform,and Azure kinect camera,and the imaging principle of the depth camera was studied and the camera was calibrated.(2)Based on the point cloud template matching method,the identification and positioning of high-speed rail wheels and the extraction of the scanning path of the probe are realized.This research first extracts and processes the point cloud of the three-dimensional image of high-speed rail wheels,saves it as a template,and extracts the target scanning path.Then the Azure kinect camera was used to reconstruct the three-dimensional point cloud of the high-speed rail wheels to match the template point cloud.(3)A virtual environment for automatic detection of high-speed rail wheels is built based on Coppeliasim.Through the API communication interface between MATLAB and CoppeliaSim,the detection manipulator model in CoppeliaSim can respond to the external control of MATLAB to simulate the scanning trajectory and scan process motion analysis.The results show that this method can realize the automatic detection of high-speed rail wheels,intuitively display the movement of the detection robot arm probe,and achieve a reasonable trajectory,which provides an experimental basis and theoretical basis for the follow-up research on the detection of the robot arm control problem.(4)The Azure kinect camera stereo vision error measurement experiment designed in this research shows that the camera has the smallest error within the working range of 0.6-1.6m,which can reach about 1mm.This research is based on the ultrasonic phased array automatic scanning experiment designed based on the scanning path extracted from the high-speed rail wheels,and the results verify the feasibility of the system.