| As a non-contact dimension measuring instrument,video measuring machine is widely used in industry.In recent years,customers’ requirements for measuring accuracy of video measuring machine have been increasing.The perpendicular length measurement error is an important technical index of video measuring machine.The measuring technology has become a research topic in the industry.The aim of this thesis is to improve the perpendicular length measurement accuracy of MM420 video measuring machine,so that the error of perpendicular length measurement can be less than(8+L/100)μm.The perpendicular length measurement method of the video measuring machine is to automatically focus on the surface of the parts to be measured at different heights and read the measured values of z-axis grating ruler,and the difference is the perpendicular length measurement result.The auto-focusing technology of video measuring machine and the measurement error compensation technology of Z-axis grating scale are the key technologies that affect the perpendicular length measurement result.In this thesis,the auto-focusing technology of video measuring machine is studied from three aspects: definition evaluation function and its performance index,focusing window and search strategy.This thesis proposes a new performance index called "resolution",which unifies software and hardware performance to achieve optimal overall performance.In this thesis,the auto-focusing technology scheme of video measuring machine is as follows: definition evaluation function is selected as improved Tenengrad gradient evaluation function,focusing window is selected as full window,binary mountain climbing search algorithm is used for coarse focusing,and Gaussian curve fitting method is used for fine focusing.Binary mountain climbing search algorithm pioneered in this thesis is an improved mountain climbing search algorithm whose search efficiency is the highest compared with the four traditional methods.In the auto-focusing experiment,100 auto-focusing experiments were carried out on steel gauge block,ceramic gauge block,printed circuit board and plastic parts,respectively.The autofocusing accuracy was 5μm(5% depth of field),auto-focusing repeatability was 2.6μm and auto-focusing time was 10.19 s.In addition,auto-focusing experiments were carried out on different surface roughness samples,and the auto-focusing accuracy reached 6μm(6% depth of field).The measurement error compensation method of z-axis grating ruler of video measuring machine can be divided into three steps: firstly,the appropriate etalon should be selected,and the appropriate method is used to record the values of the etalon and the measured values of grating ruler respectively;Secondly,the mathematical model of error compensation is established.Finally,error compensation is made by software according to mathematical model.In this thesis,the glass linear scale is selected as the etalon,the linear scale positioning tool is designed,the standard value and the measurement value are collected,and the error compensation of the measurement results is carried out by using the piecewise linear compensation mathematical model.Experimental results show that the compensation method is feasible.The maximum error of the Z-axis grating ruler decreases from 55μm to 3μm,and the measuring accuracy of the Z-axis grating ruler is greatly improved.Finally,the experimental study on the perpendicular length measurement of video measuring machine was carried out.Using 20 mm steel gauge block and special step gauge as the etalons,the perpendicular length measurement error of standard MM420 video measuring machine is(-4.5±5.88)μm and(-9.9±7.22)μm respectively.The measurement error of the proposed method on the experimental platform is(3.6±2.66)μm and(-3.7±5.07)μm.After the verification of the two measurement methods,the measurement accuracy of the proposed method is improved obviously on the experimental platform,and the measurement error is less than(8+L/100)μm.The research results of this thesis have important application value in improving the perpendicular length measurement accuracy of video measuring machine. |