中文摘要：

超磁致伸缩驱动器（GMA）虽然具有很多优点，但是超磁致伸缩材料（GMM）在磁化过程中存在磁滞非线性，磁滞误差可达20％，要解决这一问题，必须对GMA采用精确有效的方法实现建模，并用于GMA驱动位移精密控制。研究中采用LMS算法对研制的GMA进行自适应系统模型辨识，用不同频率的正弦信号和方波信号作为输入，辨识模型都能精确逼近GMA输出信号，辨识精度高达0．069um；最后采用Fx-LMS算法对GMA进行驱动位移控制实验，通过在线辨识有效减小磁滞误差，提高控制精度。

英文摘要：

Giant magnetostrictive actuator （GMA） has many applications in the area of precision driving technology. However, giant magnetostrictive material （GMM） has its inherent nonlinearity behavior such as hysteresis during the magnetization process, which can induce the error up to 20 % in displacement control. To solve this problem, the GMA needs to be modeled by a precise and effective method, so that the model can be applied to precise driving control of GMA. In this research, LMS algorithm was used for adaptive identification of the GMA system, by which the GMA system could be efficiently modeled online. The identification model, capable of achieving 0.069 μm precision, can accurately approach the output displacement of GMA with different types of input current signals and frequencies. Finally, the Fx-LMS algorithm was employed to execute the displacement control of the actuator. In the experiment, an online identification procedure was employed to reduce the hysteresis influence and improve the control precision of the displacement driving of GMA.