扫描探针显微镜中压电执行器回滞非线性软计算建模及其实时补偿

A method to model hysteresis for piezoelectric actuators of SPM by soft computing and its real-time compensation

本文提出了自适应范围DNA(AdaptiveRangeDNA ,简称ARDNA)软计算模型框架以及扫描探针显微镜 (简称SPM)中压电执行器回滞非线性ARDNA建模及其补偿原理。基于基本回滞算子的线性叠加构建压电执行器回滞非线性模型结构 ,采用ARDNA软计算方法 ,同时优化模型结构 (指简单回滞算子的个数及其阈值 )和参数 (指每个回滞算子的加权值 ) ,建立最优的压电执行器回滞非线性模型。通过设计单位反馈控制器构建其逆模型 ,而压电执行器与其逆模型串联 ,由逆模型获取参考位移所对应的理想激励电压 ,从而确保压电执行器输出跟踪参考位移 ,实现回滞非线性实时补偿。数字仿真结果表明了压电执行器回滞非线性建模及其补偿方法的有效性。本文的方法对形状记忆合金SMA(ShapeMemoryAlloy)、磁流变液体 (MagnetorheologicalFluids)

In this paper, a novel method to model hysteresis for piezoelectric actuator of scanning probe microscope (Abbr. SPM) and the real time hysteresis compensation principle were proposed. The structure of model was provided by the summation of weighted elementary hysteresis operators. The adaptive range DNA (Abbr. ARDNA) soft computing method was introduced to optimize the structure and parameters of the model simultaneously. Based on the model, the feedback controller was designed to obtain the inverse model. The piezoelectric actuator is in series with its inverse model to implement real time hysteresis compensation. Finally, the simulation results demonstrate that ARDNA is of strong ability of converging globally and finding the optimal hysteresis model for piezoelectric actuator and that the actuator can track the reference displacements because its hysteresis is compensated. And the method also can be used to model the Shape Memory Alloy (Abbr. SMA) and Magnetorheological Fluids etc., which all show hysteresis characteristic.