An enhanced Bouc-Wen model for characterizing rate-dependent hysteresis of piezoelectric actuators

Abstract

A classical Bouc-Wen model is widely applied in hysteresis modeling and compensation for piezoelectric ceramic actuators. However, the classical Bouc-Wen model cannot characterize rate-dependent hysteresis under excitations at high frequencies precisely. In this paper, an enhanced Bouc-Wen model is developed by introducing the frequency of input voltage based on the classical Bouc-Wen model. A number of experiments were conducted to characterize the rate-dependent hysteresis of piezoelectric ceramic actuators under sinusoidal excitations at a range of 1–150 Hz. The measured data were used to demonstrate the validity of the developed model. A method of parameter estimation based on the Matlab/Simulink parameter estimation tool is adopted to identify the parameters of models. The comparisons of experiments and simulations show that the developed model can describe rate-dependent hysteresis more accurately than the classical Bouc-Wen model. The modeling errors of the developed model were decreased by nearly 75% compared with that of the classical Bouc-Wen model. The root-mean-square error of the developed model is controlled in 0.1719 μm.

Nonlinear Hysteresis Modeling of Piezoelectric Actuators Using a Generalized Bouc–Wen Model

“…The enhanced Bouc–Wen model in our previous work [22] is closely related to the input frequency f and cannot be used to describe the rate-dependent hysteresis behaviors when the frequency is unknown. However, the generalized Bouc–Wen model is closely related to the rate of input trueu˙false(tfalse) and can be widely used to describe the rate-dependent hysteresis behaviors without limitations.…”

“…In our previous work [22], we have developed an enhanced Bouc–Wen model by introducing input frequency. But there is a limitation that the developed model cannot be applied when the input frequency is unknown.…”

A Modified Duhem Model for Rate-Dependent Hysteresis Behaviors

“…Over the past decade, several methods for parameters identification of models [19,24,32] have been developed, but their identification processes are generally complex. In our previous work [15,33], the nonlinear least squares method is proposed to identify Bouc–Wen model. The nonlinear least squares method adopts the trust-region-reflective algorithm and take the nonlinear least squares function for optimization through the MATLAB/Simulink Optimization Toolbox.…”

“…The rate-independent hysteresis models include the Preisach model [6,7], Prandtle–Ishlinskii model [8,9], Maxwell-slip model [10,11] and polynomial-based hysteresis model [12,13], and can be used to describe the nonlinear relationship between the input voltage and the output displacement of PCAs. However, the input rates of these models are generally lower than that of the rate-dependent hysteresis models, such as the Bouc–Wen model [14,15] and Dahl model [16,17,18]. That is mainly because, unlike the rate-independent hysteresis model, the rate-dependent one can describe the dynamic relationship between the input rate and the output.…”

Hysteresis modeling of piezoelectric micro-positioning stage based on convolutional neural network

“…A large amount of literature is devoted to extending the PI model to describe rate-dependent hysteresis. 12–18 However, there are many parameters in these models and the identification process is complicated.…”