A hysteresis functional link artificial neural network for identification and model predictive control of SMA actuator

Tai, Nguyen Trong; Ahn, Kyoung Kwan

doi:10.1016/j.jprocont.2012.02.007

Cited by 48 publications

(19 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Combining Equations (18) and (19), it can be concluded that when the identified parameters of PMPI model meet the condition (16), I-M compensator is globally stable.…”

Section: Stability Analysismentioning

confidence: 99%

“…Because of the high accuracy and flexibility, the phenomenological model is more popular in hysteresis modeling. The phenomenological models include Preisach model [11,12], polynomial model [13,14], Bouc-Wen model [15,16], Duhem model [17,18], neural network model [19,20], Prandtl-Ishlinskii (PI) model [21][22][23], and etc. Among them, because of simple expression and analytical inverse model, the PI model is the most widely used in hysteresis modeling and compensation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Research on Asymmetric Hysteresis Modeling and Compensation of Piezoelectric Actuators with PMPI Model

Wang

Chen

et al. 2020

Micromachines

View full text Add to dashboard Cite

Because of fast frequency response, high stiffness, and displacement resolution, the piezoelectric actuators (PEAs) are widely used in micro/nano driving field. However, the hysteresis nonlinearity behavior of the PEAs affects seriously the further improvement of manufacturing accuracy. In this paper, we focus on the modeling of asymmetric hysteresis behavior and compensation of PEAs. First, a polynomial-modified Prandtl–Ishlinskii (PMPI) model is proposed for the asymmetric hysteresis behavior. Compared with classical Prandtl–Ishlinskii (PI) model, the PMPI model can be used to describe both symmetric and asymmetric hysteresis. Then, the congruency property of PMPI model is analyzed and verified. Next, based on the PMPI model, the inverse model (I-M) compensator is designed for hysteresis compensation. The stability of the I-M compensator is analyzed. Finally, the simulation and experiment are carried out to verify the accuracy of the PMPI model and the I-M compensator. The results implied that the PMPI model can effectively describe the asymmetric hysteresis, and the I-M compensator can well suppress the hysteresis characteristics of PEAs.

show abstract

“…Combining Equations (18) and (19), it can be concluded that when the identified parameters of PMPI model meet the condition (16), I-M compensator is globally stable.…”

Section: Stability Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Research on Asymmetric Hysteresis Modeling and Compensation of Piezoelectric Actuators with PMPI Model

Wang

Chen

et al. 2020

Micromachines

View full text Add to dashboard Cite

show abstract

“…The phenomenological models, such as Preisach model [11], [12], Krasnoselskii-Pokrovskii model [13], Prandtl-Ishlinskii model [14], etc., describe the hysteresis by a collection of weighted elementary functions. The complex and generally multidimensional structure of this method makes it difficult to tune the parameters in real time to meet the change of the operating environment of SMA actuators [15], [16]. The physical models, on the other hand, including Duhem model [17], Bouc-Wen model [18], Liang model [19], [20], and other empirical models [21], [22], describe the SMA behavior based on its physical properties.…”

Section: Introductionmentioning

confidence: 99%

Active Modeling and Control for Shape Memory Alloy Actuators

et al. 2019

View full text Add to dashboard Cite

In this paper, an active modeling and control scheme is developed for Shape Memory Alloy (SMA) actuators to eliminate the negative influences caused by the uncertainties in its dynamics. First, a nonlinear SMA dynamic model based on Liang model and the empirical models is built and linearized, and all the uncertainties due to time-varying parameters, external disturbances, as well as the linearization, are considered as model error of the linearized model. Secondly, an active modeling based on Kalman filter is constructed to estimate the model error in real time, which intends to improve the model accuracy actively. Finally, an active modeling based control method is proposed to compensate the model error in order to improve control performance of SMA actuators. Experiments are conducted on a one degree-of-freedom (DOF) testbed actuated by a SMA wire. The experimental results of the active model error estimation, and the control performance with and without the active model based compensation are presented and compared to demonstrate the improvements of the proposed scheme. INDEX TERMS Shape memory alloy (SMA), model error, active modeling, active compensation control.

show abstract

“…To overcome this problem, proportional-integral-derivate (PID) controllers with hysteresis models have been implemented [18,19,20]; for position control of SMA actuators using the generalized Prandtl–Ishlinskii inverse model [18], for micro-positioning control of SMA actuators by modeling the hysteresis using NNs [19], and for magnetic SMA actuators by using a radial basis function NN to obtain the Jacobian information of the system in order to adjust the controller parameters [20]. In addition, neural network controllers, previously trained to identify the system were proposed to control SMA actuators [21,22,23,24]; using the inverse of the ANN that replicate the dynamics of the SMA force actuator to implement the controller [21], implementing a model predictive controller based on a functional link ANN to control the linear memory metal actuator displacement [22], and realizing a recurrent neural model predictive, variable structure, controller designed to control a one degree of freedom (1-DOF) rotary manipulator actuated by an SMA wire [23,24]. The main disadvantages of these recurrent neural network controllers are the complexity of the control system, and the requirement of training the neural network, to identify the system, prior to the implementation of the control.…”

Section: Introductionmentioning

confidence: 99%

Neural Network Direct Control with Online Learning for Shape Memory Alloy Manipulators

Gómez-Espinosa

Sundin

Eguren

et al. 2019

Sensors

View full text Add to dashboard Cite

New actuators and materials are constantly incorporated into industrial processes, and additional challenges are posed by their complex behavior. Nonlinear hysteresis is commonly found in shape memory alloys, and the inclusion of a suitable hysteresis model in the control system allows the controller to achieve a better performance, although a major drawback is that each system responds in a unique way. In this work, a neural network direct control, with online learning, is developed for position control of shape memory alloy manipulators. Neural network weight coefficients are updated online by using the actuator position data while the controller is applied to the system, without previous training of the neural network weights, nor the inclusion of a hysteresis model. A real-time, low computational cost control system was implemented; experimental evaluation was performed on a 1-DOF manipulator system actuated by a shape memory alloy wire. Test results verified the effectiveness of the proposed control scheme to control the system angular position, compensating for the hysteretic behavior of the shape memory alloy actuator. Using a learning algorithm with a sine wave as reference signal, a maximum static error of 0.83° was achieved when validated against several set-points within the possible range.

show abstract

A hysteresis functional link artificial neural network for identification and model predictive control of SMA actuator

Cited by 48 publications

References 27 publications

Research on Asymmetric Hysteresis Modeling and Compensation of Piezoelectric Actuators with PMPI Model

Research on Asymmetric Hysteresis Modeling and Compensation of Piezoelectric Actuators with PMPI Model

Active Modeling and Control for Shape Memory Alloy Actuators

Neural Network Direct Control with Online Learning for Shape Memory Alloy Manipulators

Contact Info

Product

Resources

About