Modeling and Control of a Magnetostrictive System for High Precision Actuation at a Particular Frequency

modeling;control;actuator;Magnetostrictive

Mou, Gang ; Dr. Gregory D. Buckner, Committee Member,Dr. Fen Wu, Committee Member,Dr. Paul I. Ro, Committee Chair,Mou, Gang ; Dr. Gregory D. Buckner ; Committee Member ; Dr. Fen Wu ; Committee Member ; Dr. Paul I. Ro ; Committee Chair

University:North Carolina State University

关键词: modeling;    control;    actuator;    Magnetostrictive;   

Others  :  https://repository.lib.ncsu.edu/bitstream/handle/1840.16/147/etd.pdf?sequence=1&isAllowed=y

美国|英语

来源: null

【 摘 要 】

A magnetostrictive actuator made of Terfenol-D alloy can generate high mechanical strains with broadband response and provide accurate positioning. These characteristics have been employed as controllers and vibration absorbers in industrial and heavy structural applications, such as fast tool servo systems and precision micropositioners. Full utilization of magnetostrictive transducers in these applications requires a suitable controller as well as quantification of the transducer dynamics in response to various inputs.However, at moderate to high drive levels, the output from a magnetostrictive actuator is highly nonlinear and contains significant magnetic and magnetomechanical hysteresis. The control of this nonlinear system is a challenge. In order to simplify this problem, 50Hz is chosen as the working frequency for the actuator in the experiments since it shows near linear property at 50Hz and the approach used at 50Hz could be extended to a broader frequency range in the applications.First, with an optical sensor, the dynamics of the actuator are measured under voltage inputs at different frequencies and amplitudes. Using SAS System V8, a second order dynamic model is obtained at one frequency (50Hz). This model matches the open loop behavior very well. A PID controller is then developed. The control command signal generated through the DSP board is directed to the actuator. A close loop control system is thus formed.As a nonlinear control approach, sliding mode control can offer some ideal properties, such as insensitivity to parameter variations or uncertainties, external disturbance rejection, and fast dynamic response. In order to obtain better tracking performance and robustness, a sliding mode control algorithm is introduced into the system.The experiment results from the sliding mode controller are compared with those from the open loop and PID control. The comparison shows improvement in the displacement tracking performance at this frequency. Further work will involve the modification of the sliding mode controller using a time-varying switching gain and improvement in modeling of the actuator over a broader frequency range.