基于超磁致伸缩材料的折弯型压曲放大机构设计、分析与控制

为满足微位移定位系统的应用要求，设计由超磁致伸缩材料驱动的新型折弯型柔性压曲放大机构。在其他微位移放大机构的基础上，设计便于制作和使用的二维平面型折弯型机构，该机构在实现微位移放大的同时也使位移的方向产生90°转变。通过有限元计算分析，优化折弯屈曲板的结构参数。针对超磁致伸缩材料输出位移与输入电流间磁滞非线性与其负载和频率的关系，采用BP神经网络分别构造磁滞上升半环和下降半环的逆模型，并通过输入信号对其进行选择从而进行磁滞补偿。仿真试验结果验证了该方法的有效性。构建基于超磁致伸缩材料驱动的压曲放大机构试验装置，采用该装置，分别对单环磁滞和多环磁滞进行补偿控制试验，获得了满意的位移控制效果。

DESIGN, ANALYSIS AND CONTROL OF NOVEL PRESSING AND BENDING MAGNIFYING MECHANISM DRIVEN BY GIANT MAGNETOSTRICTIVE MATERIAL

A novel pressing and bending flexible magnifying structure for micro-positioning system, which is driven by a giant magnetostrictive material based actuator, is introduced. Inspired by several other micro-displacement magnifying units, the mechanism is designed with a two-dimensional plane structure, which is both easy-manufacture and easy-using. It not only amplifies a micro displacement but also changes the direction of a translation motion by 90~. The parameters of the structure of the bended plate are optimized numerically by finite element analysis. It is observed that the hysteresis in a giant magnetostrictive material (GMM) actuator between the current or the magnetic field input and the output displacement changes with the frequencies of the input signals as well as the loads of the actuator. Two BP-neural networks are adopted to construct the inverse models of the upper and lower branches of the hysteresis loop, separately. And the two inverse models are selected by the input signals in real time control. Simulation results prove the correctness of the neural network based inverses in hysteresis compensations. An experiment setup of GMM based pressing and bending mechanism is developed. Experiments performed on the setup show the effectiveness of the inverse based compensation method for hysteresis with both main and minor loops.