GMM-based hysteresis modeling and high-performance disturbance rejection tracking control of a piezoelectric actuator
The piezoactuated nanopositioning stage eliminates low-frequency errors in fly-cutting.Herein,a novel system modeling strategy is proposed for piezo-actuating units in the stage,and a disturbance rejection tracking control scheme is designed accordingly.First,a Gaussian mixture model(GMM)is used for modeling the inherent hysteresis characteristics of a piezoelectric actuator.Based on this model,a feedforward compensation is performed to eliminate the influence of hysteresis nonlinearity on control accuracy.Second,an extended state observer is established to observe and compensate for all external disturbances and unmodeled errors to enhance the robustness of the control system.For improving the tracking performance and increasing the control bandwidth,state feedback and zero-phase tracking error control strategy are implemented to optimize the characteristics of a closed-loop system.The experimental results confirmed the effectiveness of the disturbance rejection tracking control method based on the proposed system model.While the input signal frequency ranges between 0-50 Hz,this strategy attains a system tracking error of less than 2.2％,thereby satisfying the requirements of high-precision tracking within the desired control bandwidth.
piezoelectric actuatorhysteresis nonlinearityGaussian mixture modelextended state observerdisturbance-rejection
万方数据
维普
