| With the progress of China’s advanced manufacturing industry,CNC machine tools have been applied in large-scale,especially when used in high-speed cutting occasions and precision manufacturing occasions,so the requirements for high-speed CNC machine tools are getting higher and higher,requiring accurate positioning accuracy,also have a strong driving force and fast feed system.For the CNC machine tool system,the feed device is mainly driven by the motor,so that the position of the motor is particularly important,the performance of the pros and cons will affect the entire CNC machine tool system.PMSLM(permanent magnet synchronous linear motor,PMSLM)is a"zero-drive" type device with the advantages of unlimited stroke length,high efficiency,no contact and slight noise.In ultra-precision CNC machine tool manufacturing and processing system,the linear servo system control accuracy determines the manufacturing accuracy,and the accuracy and stability of the position measurement directly affect the system’s control accuracy.Therefore,the servo system in the position of the mover high precision,high stability means of detection is crucial.Aiming at the shortcomings of various traditional linear mover position measuring sensors,the image measuring method is applied to the detection of motor mover.A digital speckle image based on improved PSNR(Peak Signal-to-noise Ratio,PSNR)is adopted.A digital speckle image with motion blur is proposed.An algorithm for accurately identifying motion blur parameters is proposed to recover the degraded image.Combined with the sub-pixel measurement algorithm based on Kalman filter and fused least square cubic curve fitting,a straight line High accuracy,high stability measurement of motor mover position.Specific work as follows:1、In view of the shortcomings of various traditional linear mover position measuring sensors,the image measuring method is used to detect the movcer,the application background and working principle of the image measuring method are analyzed,and how to use the image method to realize thc moving Sub-location measurement,which leads to the research background and the main content of this topic.2.The sub-pixel measurement algorithm based on Kalman filter and moving least square cubic surface fitting fusion is introduced to realize the high precision and stability measurement of the image.Focusing on the selection of the target source image,Aiming at the problem of the motion blur of the collected images,a method of fuzzy parameters identification and fuzzy image restoration with high precision and high efficiency is put forward.The theoretical method is verified by the experiment platform.3.A sub-pixel measurement algorithm based on fusion of Kalman filter and moving least square cubic surface fitting is studied to realize the high precision and stability measurement of the image.Aiming at the shortcomings of the traditional phase correlation algorithms and some surface fitting algorithms,this paper introduces Kalman filtering technology,combines it with moving least square surface fitting,and then combines the fitting weight factors to obtain the sub-pixel displacement The process of filtering,reducing filter pretreatment,so that greatly increased efficiency.Finally,the simulation of the classic "House" image shows that the sub-pixel measurement algorithm can meet the requirements of high accuracy and strong stability.4.The factors which affect the digital speckle image generation are introduced.According to the actual requirement of edge detection and the limitation of the test platform,a new PSNR based evaluation index of no reference image is proposed to optimize digital speckle image,and a reasonable interval of speckle image structure parameters is given.The experimental results show that the speckle images constructed in this interval have high PSNR and strong robustness.5.In view of the ambiguity of the speckle images captured in the actual shooting process,the limitations of the Radon algorithm are analyzed,and a fuzzy parameter identification algorithm with high accuracy and high accuracy is proposed.Firstly,the SVD reconstruction of the degraded image is carried out to reduce the interfrence of the noise on the identification accuracy of the motion parameters.Then,the fuzzy parameters are accurately calculated by the two minimum point coordinates of the principal component analysis and the negative peak of the cepstrum three-Degenerate image of blurred parameters restoration.Finally,the correctness of theoretical results is verified through simulation experiments. |
