Font Size: a A A

Research On Gear Vision Measurement System And Tooth Profile Measurement Technology

Posted on:2018-08-15Degree:DoctorType:Dissertation
Country:ChinaCandidate:N WangFull Text:PDF
GTID:1311330545457550Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
Gear is an important part of the mechanical transmission system,its main function is to convey the movement and power,and the application of spur gear is particularly widespread.Gear precision directly affects transmission quality of the system,so the requirement of gear precision is increasing high,accordingly,the requirement of gear measurement also stricter,measurement accuracy has become one of critical link to improve gear manufacturing technology.Tooth profile total deviation is a mandatory detection accuracy indicator to evaluate gear transmission stability,it directly reflects the transient contact of gear working flank,therefore,the tooth profile total deviation is particularly important in the precision measurement of gear.The current gear vision measurement system requires full gear located within the field of view,and the measurement accuracy is low.For the above constraints,a gear profile vision measurement system is developed,and the tooth profile measurement technology is researched deeply to achieve the purpose of measuring the small and medium module standard involute cylinder gear in 5 level precision.The main contributions of this dissertation include:(1)In order to achieve rapid measurement of high precision spur gears,a vision measurement system applicable to small and medium module standard involute cylinder gear in 5 level precision is developed.The mechanical movement module of the system is designed,and a backlight vision measurement system is set up,the system software is modular designed,and the development of submodular function is realized.(2)Based on the imaging model of the vision measurement system,the camera coordinate system,image coordinate system and measurement coordinate system are established to determine the measurement benchmark and calibrate the system.The pixel equivalent calibration is completed by calculating the ratio of physical value and pixel value of center distance of calibration circle in calibration plate.Considering various optical distortion errors,the error calibration model of the system is established by the theory center coordinates and actual center coordinates of calibration circle.The edge position error compensation model is established based on the measured diameter of the calibration circle according to the intensity of different light source.The calibration experiment adopts calibration plate and gauge block to verify the compensation accuracy of the calibration model.(3)Based on the study of existed sub-pixel edge location algorithms,according to the edge characteristic of tangential smoothing and normal steepness of the backlight image,two sub-pixel edge location algorithms are proposed: sub-pixel edge location algorithm based on Gaussian integral curve fitting and sub-pixel edge location algorithm based on Bertrand surface model.The image obtained is denoised by Gaussian filter,and the continuous single-pixel edge is determined by the eight neighborhood edge tracking algorithm,the edge tangential and normal information are obtained by fitting the pixel edge curve.In first algorithm,the gray value of points in the pixel edge normal section line are determines by the Bezier surface interpolation,and the Gaussian integral curve is fitted to the gray value to obtain the mean point coordinates and realize the sub-pixel edge location.In second algorithm,the edge gray value surface is regarded as a Bertrand surface that the generatrix is Gaussian integral curve and directrix is sub-pixel edge curve.Pixel edge neighborhood is selected as fitting area,the pixel points coordinates in fitting area are transformed into frame coordinate of edge curve,and the coordinates are fitted as Bertrand surface model,the directrix of surface is sub-pixel edge.The two algorithms separate the edge tangential and normal,and solve the problem that the edge tangential smoothing and the normal steep can not be taken into account.(4)According to the geometric characteristics of the involute gear,the measuring points of tooth profile can be seen as the points of same base circle with different different initial phase angle on standard involute,the measurement model of tooth profile total deviation based on the initial phase angle is established.Due to the center of the medium and small module gear is beyond the scope of vision measurement system,the calibration disc is used to locate external gear center,in the case of invariable relative position of the mandrel,lens and light source,the calibration disc image is obtained,and the radius constraint least square circle fitting method is used to fit the outer circumference edge of the calibration disc to obtain the gear center coordinates in the image coordinate system.In order to reduce the influence of the calculation error and other factors on edge location,the first order continuous segment curve fitting is performed on the sub pixel edge of the tooth profile,and the measuring points of tooth profile is determined.Based on the initial phase angle of each measuring points,the tooth profile total deviation is calculated by using the measurement model of tooth profile total deviation based on the initial phase angle.(5)The measurement of the tooth profile total deviation is carried out by applying the gear profile vision measurement system,the influence of repeatability precision of the measurement system and the errors of optical system,gear center location,system manufacturing,oil stains and burr on gear surface are analyzed.The measurement results of this system and gear measurement center are compared to verify that the system can realize the measurement of small and medium module standard involute cylinder gear in 5 level precision.
Keywords/Search Tags:Vision measurement, Involute gear, System calibration, Edge location, Tooth profile total deviation
PDF Full Text Request
Related items