Research On Scanning Detection System For 3D Profile Of Non-contact Rail

In recent years,high-speed rail has become the most convenient way for people to travel.The problem with this phenomenon is that the wear rate of the rails will become faster and faster,and the quality of the rails is also increasingly demanding.At present,the detection of rail profile by various welding rail bases in China is mainly based on contact detection such as caliper and electronic ruler.Some developed countries have begun to use non-contact detection methods to detect rail contour parameters and other state parameters,but the accuracy is not enough.high.This paper proposes a detection device based on the laser profiler to collect the contour information of the rail.The laser profilometer measures with precise laser triangulation and is fast and accurate.However,in the actual rail inspection scene,there is a visual blind zone when measuring a laser profilometer,and it is impossible to detect all the data on the four sides of the rail.Therefore,the research content of this paper is how to apply the laser profilometer to the modern rail inspection field.Important,the main work is as follows:(1)The principle of measurement and measurement of laser profilometer is studied.The direct and oblique type of laser triangulation is studied in detail.This paper is based on four laser profilometers to scan the contours of the rails.Each laser profiler is mutual.Independent,so we studied the quadratic fitting of the center of the rail rail waist feature by the constrained least squares fitting method after transferring the four independent contours to the same coordinate system to complete the feature points on the rail contour.Extraction and matching to achieve three-dimensional stitching of the full contour of the rail.(2)Considering the influence of system vibration and external disturbance in the transport rail,this paper studies an adaptive covariance estimation algorithm for noise suppression.The system innovation is obtained by the established state transition equation and the measurement equation and the signal value measured based on the system.Thesystem noise and the measurement noise covariance matrix are adaptively adjusted according to these innovations.In the aspect of rail flatness and end face verticality detection,this paper designs an efficient geometric detection algorithm with high processing speed and high accuracy.(3)The dynamic scanning control method of the sliding scanning table in the rail detecting device is studied.The difficulty lies in the accuracy of the control.In order to improve the control precision of the device,the detecting device integrates a plurality of photoelectric switch sensors with a scale displacement sensor to generate a plurality of electrical signals for measurement control,and completes closed-loop control of the sliding scanning platform according to the real-time feedback value of the grating scale.Then,the positioning control of the sliding scanning platform is completed by the positioning module of QPLC.In this paper,based on the deficiency of the T acceleration-deceleration control method,an S acceleration and deceleration control method is designed to reduce the influence of the vibration of the sliding scanning platform on the laser measurement accuracy.By analyzing and processing the parameters of the rails measured by the equipment,and comparing the national standard values with manual measurement methods,it is concluded that the flatness error does not exceed 0.04 mm and the verticality error does not exceed 0.03 degree,and the device can be verified.The operation and measurement results are stable,and the test results are in line with national accuracy standards.The testing equipment integrates various existing flatness measuring scales and twisting measuring scales,and can simultaneously measure the various parameters of the rails,which greatly promotes the welding production of the rails.