Precision Measuring Method Research Of Linear Motor Mover Position Based On Stripe Image EPCA

Permanent magnet synchronous linear motor(PMSLM)servo system with high-precision,high speed,zero transmission and high dynamic characteristics,etc.,to achieve fast and high-precision drive,is the core drive components of high-precision machine tool,and have broad prospects in the field of precision machining.The real-time precision measurement of PMSLM mover position is very important for high-precision control of linear servo feedback system,which directly affects the quality of high-precision machining.The traditional position measurement method of PMSLM mover mostly adopts some position sensors,which are easily disturbed by the external environment,and have poor anti-noise capability and anti-vibration performance.There are some shortcomings in the practical application.Digital image measurement technology has the advantages of non-contact,fast,high precision full-field measurement and small environmental impact,which is the new direction and hotspot in the field of modern measurement.Digital image measurement method is applied to the real-time feedback precision measurement of the PMSLM's mover position to overcome the deficiencies of the traditional mover position detection method,which is of great significance for improving the control precision of the PMSLM servo system and further improving the product machining precision,In order to realize the real-time,high-precision and robustness of PMSLM's position measurement,this subject introduces the image measurement technology into the position measurement of PMSLM's mover in combination with the advantages of image measurement method,and then a precise position measurement method based on 1-D extended phase correlation algorithm(EPCA)of fence image features is proposed to realize the stable and efficient position measurement of PMSLM mover.This subject mainly carries on the following several aspects research:1.The research background and significance of the subject are expatiated.Development status of the mover position measurement technology and the advantages of the image measurement method are analyzed.Especially,the application and research status of digital image measurement method in the position detection of mover,as well as the research status quo of digital image correlation measurement(DIC)methods are introduced in detail,and the main contents and framework of this thesis are given.2.According to the moving characteristics of linear motor,the image measurement technology is introduced into the position measurement of linear motor mover,and the fence image measurement model of PMSLM's mover position is established.Combined with phase correlation algorithm(PCA),it has strong robustness,high precision(integer pixel)and insensitivity to image gray-scale change.Based on the fact that the phase correlation peak matches the zero-mean normal distribution,this thesis proposes a sub-pixel displacement measurement method based on multi-peak fitting PCA to realize the sub-pixel displacement measurement of the PMSLM mover position.The experimental verification and analysis are also carried out.3.Aiming at the accuracy and robustness of the linear motor mover position detection,singular value decomposition(SVD)is introduced to improve the structure of PCA.The decomposition of two-dimensional image phase correlation matrix(PCM)into two independent singular vectors in the horizontal and vertical directions not only reduces the complexity and improves the anti-interference capability of the algorithm,but also the matching of two-dimensional image parameters is reduced to one-dimensional phase unwrapping,and then the linear phase difference between the images is obtained.Then,the least square method(LSM)was used to perform phase-difference linear fitting,and the exact sub-pixel offset between the images was obtained.The stable and accurate sub-pixel displacement measurement of the mover position was achieved and verified by experiments anti-interference analysis,the accuracy of 0.01 pixels.4.In order to improve the real-time performance of PMSLM's mover position detection,multi-line sampling and parallel computing methods are used to directly convert the motion parameter estimation between two-dimensional fence images into the simpler one-dimensional fringe signal correlation calculation.Based on the 1-D PCA technique,LSM linear fitting was introduced to achieve real-time high-precision measurement of the fence image sequences.Meanwhile,to suppress the spectrum leakage of discrete Fourier transform(DFT)and ensure the accuracy of the displacement detection,the Hanning window function is utilized to preprocess the sampled one-dimensional fringe signals.The accuracy and anti-jamming ability of the proposed algorithm were verified.5.Finally,based on the principle of image measurement,an image position test platform of PMSLM mover is established and the real-time performance,robustness and measuring accuracy of the proposed motion detection method based on 1-D EPCA are analyzed by comparative experiments,and the detection accuracy is about up to 0.02mm.Therefore,the research on the precision measurement of linear motor mover based on the fence image EPCA can provide a new reference method for the fast and accurate position detection of PMSLM mover.