Neural Network Self-Tuning Control for a Piezoelectric Actuator

Abstract

Piezoelectric actuators (PEA) have been widely used in the ultra-precision manufacturing fields. However, the hysteresis nonlinearity between the input voltage and the output displacement, which possesses the properties of rate dependency and multivalued mapping, seriously impedes the positioning accuracy of the PEA. This paper investigates a control methodology without the hysteresis model for PEA actuated nanopositioning systems, in which the inherent drawback generated by the hysteresis nonlinearity aggregates the control accuracy of the PEA. To address this problem, a neural network self-tuning control approach is proposed to realize the high accuracy tracking with respect to the system uncertainties and hysteresis nonlinearity of the PEA. First, the PEA is described as a nonlinear equation with two variables, which are unknown. Then, using the capabilities of super approximation and adaptive parameter adjustment, the neural network identifiers are used to approximate the two unknown variables automatically updated without any off-line identification, respectively. To verify the validity and effectiveness of the proposed control methodology, a series of experiments is executed on a commercial PEA product. The experimental results illustrate that the established neural network self-tuning control method is efficient in damping the hysteresis nonlinearity and enhancing the trajectory tracking property.

Iterative Learning and Fractional Order PID Hybrid Control for a Piezoelectric Micro-Positioning Platform

“…The modeling method based on neural network not only possesses evident dominant in precision, but also possesses a good prospect in the application of production process, so it is often used in the modeling of hysteresis nonlinear system [36]- [39]. ALNN is a typical feedforward neural network comprised of an input layer and an output layer.…”

Feedforward Control of Piezoelectric Ceramic Actuators Based on PEA-RNN

“…Therefore, there has been no widely accepted general model. More and more researchers have introduced other new piezoelectric ceramic feedforward control methods, such as using the Radial basis function(RBF) network [ 14 ] and neural network self-turning control [ 15 ].…”

Micrometer Level Control Design of Piezoelectric Actuators: Fuzzy Approach

“…These linear control designs possess the advantages of an easy to implement control structure and a low calculation consumption. Recently, nonlinear control designs for piezoelectric actuators were proposed [9,10]. [10] delivered a neural network-based control design which requires a huge convergence time for adjusting weights of the backpropagation neural network, and convergence of the displacement tracking error cannot be guaranteed and proven mathematically.…”

“…Recently, nonlinear control designs for piezoelectric actuators were proposed [9,10]. [10] delivered a neural network-based control design which requires a huge convergence time for adjusting weights of the backpropagation neural network, and convergence of the displacement tracking error cannot be guaranteed and proven mathematically. A nonlinear robust fuzzy eliminator with two approximators for learning the disturbed piezoelectric actuator model and eliminating hysteresis is investigated in [9].…”