scholarly journals A hopfield neural network-based Bouc-Wen model for magnetic shape memory alloy actuator

AIP Advances ◽  
2020 ◽  
Vol 10 (1) ◽  
pp. 015212 ◽  
Author(s):  
Yifan Wang ◽  
Chen Zhang ◽  
Zhongshi Wu ◽  
Wei Gao ◽  
Miaolei Zhou
Keyword(s):  
Neural Network ◽  
Shape Memory Alloy ◽  
Shape Memory ◽  
Hopfield Neural Network ◽  
Magnetic Shape Memory ◽  
Shape Memory Alloy Actuator ◽  
Magnetic Shape Memory Alloy

Experimental characterization and control of a magnetic shape memory alloy actuator using the modified generalized rate-dependent Prandtl–Ishlinskii hysteresis model

2018 ◽  
Vol 232 (5) ◽  
pp. 506-518 ◽  
Author(s):  
Saeid Shakiba ◽  
Mohammad Reza Zakerzadeh ◽  
Moosa Ayati
Keyword(s):  
Shape Memory Alloy ◽  
Shape Memory ◽  
Excitation Frequency ◽  
Hysteresis Loops ◽  
Experimental Results ◽  
Hysteretic Behavior ◽  
Magnetic Shape Memory ◽  
Shape Memory Alloy Actuator ◽  
Rate Dependent ◽  
Magnetic Shape Memory Alloy

In this article, two models are used, namely rate-independent and rate-dependent generalized Prandtl–Ishlinskii, to characterize a magnetic shape memory alloy actuator. The results show that the rate-independent model cannot consider the effect of input excitation frequency, while the rate-dependent model omits this drawback by defining a time-dependent operator. For the first time, the effects of excitation frequency on the hysteretic behavior of magnetic shape memory alloy actuator are investigated. In this study, five excitation voltages with different frequencies in the range of 0.05–0.4 Hz are utilized as inputs to the magnetic shape memory alloy actuator and the displacement outputs are measured. Experimental results indicate that, with increasing the excitation frequency, the size of the hysteresis loops changes. Since the generalized rate-dependent Prandtl–Ishlinskii model cannot consider the asymmetric hysteresis loops, in the developed model, a tangent hyperbolic function is applied as an envelope function in order to improve the capability of the model in characterizing the asymmetric behavior of magnetic shape memory alloy actuator. The parameters of both rate-dependent and rate-independent models are identified by genetic algorithm optimization. The results reveal that the rate-independent form is not capable of accurately describing the hysteretic behavior of magnetic shape memory alloy actuator for different input frequencies. Simulation and experimental results also demonstrate the proficiency of the developed model for precise characterization of the saturated rate-dependent hysteresis loops of magnetic shape memory alloy actuator. In addition, the proposed model is utilized for determining a proper range for controller coefficients during controller design.

Sign in / Sign up

Export Citation Format

Share Document