Surrogate models for uncertainty analysis of micro-actuator

Intelligent materials such as shape memory alloy have attracted the attention of scientists to innovate applications in micromachining technology. It is worth mentioning that the numerical study of such technology without considering uncertainties in material parameters has shown a great attention. However, material parameters show variability due to the experiment measurement of these parameters. This paper focuses on uncertainty analysis of a shape memory alloy micro actuator with taking into account uncertainties in material parameters. An uncertainty analysis approach combining the finite element method, metamodeling and Monte Carlo simulation is presented in this work. The constructed metamodels are validated and compared by errors measures and cross validation. After that, Monte Carlo simulation is conducted by the approximations provided by the metamodel. The metamodel based probabilistic method used in this paper is considered as an approach with high efficiency for uncertainty analysis in micro actuator.