Optimal design of Al/Si bimorph electro-thermal microactuator by integrating finite-element code with optimization method

Abstract

The present study is aimed to develop an optimization approach which is employed to design an electro-thermal microactuator based on numerical simulation of the device performance. The microactuator basically adopts a bimorph structure which consists of a P-type silicon layer and an aluminum layer. A finite-element multi-physics code (ANSYS 10.0) is used to develop the direct problem solver for predicting the stress/stain, electric, and thermal fields associated with the varying material layers thicknesses during the iterative optimization process. The simplified conjugate-gradient method (SCGM) is used as the optimization method. Regarded as a test case, the modifications of a conventional design, which suffered from high temperature, excessive power consumption, and oxidization of aluminum layer, are attempted. In this study, firstly the direct problem solver is used to predict the performance of various design models so as to select the best one with highest performance. Parametric study of the best design model is then performed, and based on the solutions yielded from the parametric study, the subtle influence of thicknesses of the P-type silicon and the aluminum layers ( t S and t AL ) on the performance of the microactuator is observed. Finally, the developed optimization approach is applied to predict the optimal combination of the geometrical parameters. The optimal thicknesses of the material layers used in the microactuator are successfully determined, and the optimization process leads to a great improvement in the performance of the microactuator. Results also show that the approach is robust and the obtained optimal combination of the geometrical parameters is independent of the initial guess for the particular case.