Purpose–The purpose of the current investigation is to design a robust and reliable computational framework to effectively identify the nonlinear behavior of shape memory alloy(SMA)actuators,as one of the most applic...Purpose–The purpose of the current investigation is to design a robust and reliable computational framework to effectively identify the nonlinear behavior of shape memory alloy(SMA)actuators,as one of the most applicable types of actuators in engineering and industry.The motivation of proposing such an intelligent paradigm emanates in the pursuit of fulfilling the necessity of devising a simple yet effective identification system capable of modeling the hysteric dynamical respond of SMA actuators.Design/methodology/approach–To address the requirements of designing a pragmatic identification system,the authors integrate a set of fast yet reliable intelligent methodologies and provide a predictive tool capable of realizing the nonlinear hysteric behavior of SMA actuators in a computationally efficient fashion.First,the authors utilize the governing equations to design a gray box Hammerstein-Wiener identifier model.At the next step,they adopt a computationally efficient metaheuristic algorithm to elicit the optimum operating parameters of the gray box identifier.Findings–Applying the proposed hybrid identifier framework allows the authors to find out its advantages in modeling the behavior of SMA actuator.Through different experiments,the authors conclude that the proposed identifier can be used for identification of highly nonlinear dynamic behavior of SMA actuators.Furthermore,by extending the conclusions and expounding the obtained results,one can easily infer that such a hybrid method may be conveniently applied to model other engineering phenomena that possess dynamic nonlinear reactions.Based on the exerted experiments and implementing the method,the authors come to the conclusion that integrating the power of metaheuristic exploration/exploitation with gray box identifier results a predictive paradigm that much more computationally efficient as compared with black box identifiers such as neural networks.Additionally,the derived gray box method has a higher degree of preference over the black box identifiers,as it allows a manipulated expert to extract the knowledge of the system at hand.Originality/value–The originality of the research paper is twofold.From the practical(engineering)point of view,the authors built a prototype biased-spring SMA actuator and carried out several experiments to ascertain and validate the parameters of the model.From the computational point of view,the authors seek for designing a novel identifier that overcomes the main flaws associated with the performance of black-box identifiers that are the lack of a mean for extracting the governing knowledge of the system at hand,and high computational expense pertinent to the structure of black-box identifiers.展开更多
文摘Purpose–The purpose of the current investigation is to design a robust and reliable computational framework to effectively identify the nonlinear behavior of shape memory alloy(SMA)actuators,as one of the most applicable types of actuators in engineering and industry.The motivation of proposing such an intelligent paradigm emanates in the pursuit of fulfilling the necessity of devising a simple yet effective identification system capable of modeling the hysteric dynamical respond of SMA actuators.Design/methodology/approach–To address the requirements of designing a pragmatic identification system,the authors integrate a set of fast yet reliable intelligent methodologies and provide a predictive tool capable of realizing the nonlinear hysteric behavior of SMA actuators in a computationally efficient fashion.First,the authors utilize the governing equations to design a gray box Hammerstein-Wiener identifier model.At the next step,they adopt a computationally efficient metaheuristic algorithm to elicit the optimum operating parameters of the gray box identifier.Findings–Applying the proposed hybrid identifier framework allows the authors to find out its advantages in modeling the behavior of SMA actuator.Through different experiments,the authors conclude that the proposed identifier can be used for identification of highly nonlinear dynamic behavior of SMA actuators.Furthermore,by extending the conclusions and expounding the obtained results,one can easily infer that such a hybrid method may be conveniently applied to model other engineering phenomena that possess dynamic nonlinear reactions.Based on the exerted experiments and implementing the method,the authors come to the conclusion that integrating the power of metaheuristic exploration/exploitation with gray box identifier results a predictive paradigm that much more computationally efficient as compared with black box identifiers such as neural networks.Additionally,the derived gray box method has a higher degree of preference over the black box identifiers,as it allows a manipulated expert to extract the knowledge of the system at hand.Originality/value–The originality of the research paper is twofold.From the practical(engineering)point of view,the authors built a prototype biased-spring SMA actuator and carried out several experiments to ascertain and validate the parameters of the model.From the computational point of view,the authors seek for designing a novel identifier that overcomes the main flaws associated with the performance of black-box identifiers that are the lack of a mean for extracting the governing knowledge of the system at hand,and high computational expense pertinent to the structure of black-box identifiers.
