Design And Drive Control Of MEMS Self-sensing Deformable Mirror

With the development of modern laser communication,medical imaging,and other beam deflection technologies,the demand for new beam deflection devices has increased.Based on the flexible characteristics of new functional materials,combined with existing mature MEMS technology,a new type of beam deflection device is designed and manufactured that integrates driving and beam deflection functions based on functional materials.It is expected to solve the problems of mechanical inertia in traditional mechanical deflection devices.Therefore,this article designs a self sensing deformable mirror based on ionic polymer materials(IPMC)to integrate its sensing and execution functions,in order to solve the problem of external sensing systems required for beam deflection devices and facilitate better control.The main content of this article is as follows:Introduce the driving mechanism and mathematical model of IPMC deformable mirror,and measure the mechanical characteristics of the deformable mirror under voltage driving through an experimental platform.Subsequently,mechanical analysis was conducted on the designed IPMC cantilever beam structure to determine the relationship between deflection,strain,and rotation angle with input voltage.The driving equation of voltage displacement angle was established,and the proposed equivalent beam model was verified through COMSOL finite element analysis and simulation.The self sensing scheme is designed.The electrode is patterned by designing a mask.Using the strain resistance effect of the electrode,the bridge circuit is used to measure its change value and realize the temperature compensation of the resistance.Determine the optimal position for electrode patterning through mechanical analysis and COMSOL strain simulation diagram,and design sensing calibration experiments.Design a CMAC neural network expert PID composite control scheme to achieve tracking control of IPMC deformable mirrors.The CMAC neural network has the characteristics of high fitting,fast convergence speed,and online learning.Therefore,the CMAC neural network is used to establish the hysteresis inverse model of the IPMC deformable mirror as a feedforward inverse compensation controller.Secondly,design expert control rules and use expert experience to adjust PID parameters to achieve closed-loop control.To verify the effectiveness of the proposed modeling and control methods,MATLAB simulation experiments were conducted.The results show that using CMAC neural network to approximate the hysteresis model of IPMC has high accuracy;CMAC inverse compensation expert PID compound control method has good tracking control effect,can meet the control requirements of complex dynamics,and is suitable for the control of strong disturbance,time-varying,nonlinear systems,with good robustness and adaptability.Finally,the control algorithm was validated through an experimental platform.