Research Based On Piezoelectric Ceramics Hysteresis Nonlinear Modeling And Control

In order to improve space optical remote sensor’s control accuracy of imagestabilization during tracks movement in the imaging process, taking piezoelectricactuator as the research object, based on domestic and external piezoelectrichysteresis modeling and control methods, we make a deep research on systemidentification theory and non-linear system control methods. Piezoelectric ceramic hassmall size, large output force, high displacement resolution, high frequency response,no noise, no heat features, is one of the best actuator’s conmponent in small-scale andmicro-precision control systems. However, voltage driven piezoelectric ceramiccontain nonlinearity casued by hysteresis and creep, making it difficult into usage. Byway of solving these problems, using a combination of theory and practice, weanalyse piezoelectric ceramics hysteresis mechanism for the high precision and smallstroke piezoelectric ceramic, implement the device’s hysteresis nonlinearity, make adynamic hysteresis model of piezoelectric ceramics, control by inverse compensationand incremental PID, finally, achieve the piezoelectric system hysteresis nonlinearityerror correction.Analysing hysteresis characteristics of the piezoelectric ceramic,we give apiezoelectric ceramic hysteretic nonlinearity formation mechanism. Aimming atpiezoelectric ceramic’s buck and boost hysteresis curve not coincide, the actuator’soutput displacement of the creep characteristics with time hysteresis, the linearexpansion and the piezoelectric effect on the temperature characteristics, we study the pressure electric ceramic microscopic polarization mechanism. When thepiezolelectric ceramic electric field is large, from the microscopic realm, give thepiezoelectric ceramic hysteretic nonlinearity causes; indentify two internalpiezoelectric ceramic factors including electric dipole polarization and electric domainsteering, resulting in a large number of ferroelectric domain irreversible shift.Research based on hysteresis modeling methods of piezoelectric ceramic.Improve Duhem model parameter identification. For fast swing mirror in large spaceoptical remote sensor, we need to suppress the vibration distribution in a wide band,and choose Duhem dynamic hysteresisi model to finish nonlinearity correction, basedon system identification theory, improve the forgetting factor recursive least squaresmethod, complete Duhem model parameter identification, the recognition accuracycan reach5310-6.By feedforward inverse compensation and imcremental PID control method, westudy the piezoelectric ceramic hysteretic control method, the control absolute errorhas improved greatly. According to the inverse function theory, establish Duheminverse model, achieved inverse model compensation, given incremental PID controlof adjustable parameters, used deviation of three measurements getting controlincrement to further enhance the control precision. Matlab/Simulink simulationresults show: in the range of0~10μm, the control algorithem’s absolute error is lessthan3.5nm. For sake of the piezoelectric ceramic can be used in a reliable,high-precision control system, it lays a good technical foundation.Based on graphical language environment LabvieW，we finish the realizationbased on LabvieW FPGA algorithms and control system with NI Rseries IntelligentCapture Card7841, high-speed AD, DA acquisition system, algorithms built on theVirtex-5FPGA chip; Design bias voltage regulator circuit, complete the experimentalsystem’s building and testing. Algorithm makes the range of0~10μm, the controlsystem’s absolute error is less than90%, and implements dynamic performance in nslevel.Through simulation experiments, we verify the effectiveness of the algorithm to improve the piezoelectric ceramic system control precision and dynamiccharacteristics; the research and simulation results show that Piezoelectric devices inspace optical remote sensor fast swing mirror precision tracking and positioningsystems, have played an important application value.