Study On Dynamic Measurement Of MEMS Microstructures Based On Electro-optical Effect

Three dimensional dynamic performance test of MEMS,which is used to characterize the stability and reliability of MEMS device in high speed motion state,is very important in the development of MEMS product.With the development of the times,the measurement of dynamic mechanical properties of MEMS micro structure is becoming more and more important.However,because the MEMS device is usually in the millimeter micron level,the size is very small,and the range of the measuring instrument is limited.Traditional methods such as Atomic Force Microscopy and Electron Scanning Microscopy can not measure the dynamic mechanical properties of MEMS microstructure.Therefore,in this paper the Stroboscopic Microscopic Vision and Stroboscopic Microscopic Interference are studied in detail,a based on electro-optic effect of MEMS Micro structure dynamic 3D measurement system is designed,for in-plane motion and out of plane motion measurement of MEMS,and implement the corresponding 3D display and algorithm processing.In this paper,for the hardware design part of the three-dimensional dynamic measurement synchronization control system of the MEMS micro structure,there are some improvements compared with the previous.The one is to change the acoustic optic modulator to an electro optic modulator.The frequency of the laser signal can be increased from 1MHz to 10 MHz,the time of exposure is changed from 1us to 0.1us,and the sampling rate of the system is improved.The other one is to add lithium niobate phase shifter before the reference mirror,because of the principle of electro optic effect.Because there is no mechanical moving parts,so the repeatability is improved.The paper uses MCU and FPGA to control the laser strobe and CCD cameras.The software part of this paper,the computer program based on MFC.The upper computer provides the operating interface of the measuring system,and realizes the function of controlling data input,image acquisition and preservation and processing.In this paper,we study the Micro Vision and Micro Interference technology,and the motion of the MEMS surface and the motion correlation algorithm.First,the sub-pixel step method,the normalized covariance method,the two surface fitting method and the normalized covariance correlation method are used to realize the sub pixel level template matching.Secondly,the simplex search method and the penetration similarity detection algorithm are introduced to improve the matching speed.In the end,the traditional path to wrap the algorithm,and the path independent of the path and the path following algorithm and the path following algorithm are summarized.Finally,A new method for predicting the path of the path of the quality map is presented in this paper.The new algorithm can unwrap the whole interference continuously to the package and no discontinuity or intermittent small area,to avoid the traditional path growth forecast algorithm without good balance residuals so as to lead to discontinuities,algorithm has higher accuracy,shorter running time.