Hysteresis modelling and compensation for piezoelectric actuator using Jaya-BP neural network

Abstract

This paper proposes a new training algorithm using a hybrid Jaya-back propagation algorithm (called H-Jaya) to optimize the neural network weights, which is applied to identify the nonlinear hysteresis Piezoelectric actuator based on the experimental input-output data. The identified H-Jaya-neural model will be used to design an advanced feed-forward (FF) controller for compensating the hysteresis nonlinearity. Furthermore as to improve the tracking performance, a feed-forward-feedback control scheme is conducted. To evaluate the effectiveness of the proposed approach, firstly, it is tested through identifying the nonlinear hysteresis of Piezoelectric (PZT) actuator and compared with other meta-heuristic techniques, including differential evolution (DE), particle swarm optimization (PSO), and Jaya. Then, the accuracy of the hysteresis model-based compensator is evaluated under various control experiments using the piezoelectric actuator. The results of experiments executed on PZT actuator configured with a PZS001 from Thorlabs prove that the proposed approach obtains an excellent performance in hysteresis modeling and compensation.

An explainable neural network integrating Jiles-Atherton and nonlinear auto-regressive exogenous models for modeling universal hysteresis

“…Recognizing these challenges, recent decades have witnessed the integration of neural networks in hysteresis modeling of piezoelectric dynamic systems [13][14][15].…”

An Identification Method for Dynamic Systems with SCNs Hysteresis Compensator: Application to the Inchworm Piezo Motor

“…It is currently possible to classify hysteresis modeling into two main categories [19]. A hysteresis model-based approach is primarily used [20,21]. It has been shown that Prandtl-Ishlinskii(PI) hysteresis modeling is one method often used to describe hysteresis [22], and it typically describes a continuous loop of hysteresis that is rateindependent [23].…”

Modeling and identification of nonlinear hysteresis behavior of piezoelectric actuators using a computationally efficient phenomenological model and modified cuckoo search algorithm

“…Different from physical models, phenomenological hysteresis models employ numerical equations directly to characterize nonlinear input and output relationships of hysteresis, without regard to the natural physical properties. Accordingly, phenomenological models are extensively applied in nonlinear hysteresis modeling of PZT actuators, such as Preisach model [9][10][11][12], Maxwell slip model [13,14], Prandtl-Ishlinskii (PI) model [15][16][17], Rayleigh model [18], Dahl model [19], Duhem model [20,21], Jiles-Atherton (J-A) model [22], neural networks model [23], frictional model [24], Bouc-Wen (BW) model [25][26][27], etc. Xiao and Li [12] developed a modified inverse Preisach model to characterize hysteretic responses of PZT actuators at a wide frequency range, where µ-density functions and weights were optimized by fast Fourier transform to realize the online rate-dependent compensation of PZT hysteresis in real-time.…”

“…Gan et al [20] revised the Duhem model to portray rate-dependent hysteretic responses of PZT actuators, where the trigonometric function was included with nonlinear least square (LS)based parameter identification. Son and Anh [23] utilized a back-propagation neural network to develop a feed-forward controller to compensate for the hysteresis responses of PZT actuator, the performance of which has been validated by experimental data. Gan and Zhang [26] devised a general BW model with relaxation function, which well illustrates ratedependent hysteretic responses of PZT actuators.…”