Adaptive control for piezoelectric actuator using direct inverse modeling approach

Abstract

Piezoelectric actuators with their high stiffness, fast response and nearly unlimited displacement resolution have been widely used in precision positioning systems. But, the existence of the inherent hysteretic behavior considerably degrades the accuracy of the actuators. To compensate for hysteresis, one of the suggested ways is to have a feedforward controller with an inverse model to linearize the relationship between the input and output. There are two approaches getting an inverse model: one is model-inversion approach and the other is direct inverse modeling approach. It is quite difficult to get inverse models for most hysteresis models except for classical Prandtl-Ishlinskii model (P-I) based on model-inversion approach. In this paper, the direct inverse modeling approach was adopted to get an inverse model based on a modified P-I model. Meanwhile, an adaptive control for piezoelectric actuator using direct inverse modeling approach was proposed. The parameters of the inverse P-I model can be identified online. The experimental results show that the tracking performance of the adaptive inverse control is much better than the tracking performance of a single inverse control.