Neural network based iterative learning control for magnetic shape memory alloy actuator with iteration-dependent uncertainties

The hysteresis of the piezoelectric actuator exhibits obvious rate dependence, especially in fast scanning or tracking. This article proposes an adaptive hysteresis compensation method that integrates the direct inverse modeling (DIM) method and an adaptive Kalman filter (AKF). DIM is adopted to directly obtain the inverse hysteresis model and use it as the hysteresis compensator. AKF is utilized to dynamically update the weights of the inverse hysteresis model so as to eliminate the manual tuning of control parameters. Further, a forgetting factor is adopted in AKF to reduce the influence of the old data. Experimental results on tracking of step, sinusoidal, and triangular trajectories demonstrate the effectiveness of DIM + AKF in hysteresis compensation. The transient and steady state performances of the closed-loop system are significantly improved. A closed-loop bandwidth higher than 200 Hz is achieved. The robustness of DIM + AKF against the rate dependence is further verified using a 0–200 Hz swept sinusoidal trajectory. As the onsite tuning is successfully eliminated, DIM + AKF is easy to follow for inexperienced users. 

Direct Inverse Hysteresis Compensation of Piezoelectric Actuators Using Adaptive Kalman Filter

This paper proposes a feedforward and feedback combined hysteresis compensation method for a piezoelectric actuator (PEA) based on the multi-layer feedforward neural network (MFNN) inverse model. Under the scheme of direct inverse modeling, the MFNN is utilized as the feedforward hysteresis compensator, which can be directly identified from the measurements. The high modeling accuracy and high robustness of the MFNN help to increase the bandwidth of the closed-loop system. Experiments are conducted on a commercial PEA so as to verify the effectiveness of the proposed method. The superimposition of two sinusoidal signals is found to be efficient for the training of the MFNN. Closed-loop trajectory tracking experiments demonstrate that the bandwidth can be increased up to 1000 Hz and the maximum deviation can be maintained closed to the noise level. Meanwhile, there are only two parameters to be tuned in the proposed method, which guarantees ease of use for the inexperienced users. The proposed method successfully realizes high-precision hysteresis compensation performance across a wider frequency range.

High-Bandwidth Hysteresis Compensation of Piezoelectric Actuators via Multilayer Feedforward Neural Network Based Inverse Hysteresis Modeling.

Piezoelectric actuator (PEA) is widely applied in the field of micro/nano high precision positioning. However PEA has the phenomenon of hysteresis non-linearity between input voltage and output displacement, due to the natural property of piezoelectric materials. The PEA hysteresis can be compensated by hysteresis models, which makes the input voltage and output displacement more linearity. The research work on compensation of PEA hysteresis by using various hysteresis models has been being a hot topic. This paper presents a modified direct inverse rate-independent Prandtl–Ishlinskii (PI) (MDIPI) model for compensating the hysteresis of PEA. The proposed MDIPI model has two different sets of operators compared with classical PI (CPI) model having one set of operators. For the two sets operators in MDIPI model one is rate operators and the other is modified classical operators. By combining the two sets operators, the MDIPI model has the properties of the adaption and accuracy in hysteresis compensation. The MDIPI model can be used as feedforward controller to compensate different reference trajectories. Parameters of MDIPI model are calculated by matlab optimization tool box. The experiments of compensating the complex displacement trajectory and sinusoidal trajectory are validated on a platform of commercial PEA. The MDIPI model has achieved more accurate results than the Krasnosel’skii–Pokrovkii (KP), Preisach and CPI models. It is effective in improving the accuracy of PEA hysteresis compensation. 

A modified direct inverse Prandtl–Ishlinskii model based on two sets of operators for the piezoelectric actuator hysteresis compensation

Direct inverse linearization of piezoelectric actuator’s initial loading curve and its applications in full-field optical coherence tomography (FF-OCT)

Rurui Jia

Papers

Direct inverse linearization of piezoelectric actuator’s initial loading curve and its applications in full-field optical coherence tomography (FF-OCT)