Optimal design of linear oscillatory actuator using genetic algorithm

This paper presents the optimal design method of a linear oscillatory actuator (LOA) using the genetic algorithm (GA). In this method, the GA is coupled with a 2D-dynamic analysis code based on the finite element method (FEM) to obtain the optimal geometry which can give a satisfactory dynamic performance. In order to take the motion of the LOA into consideration, the finite element meshes are automatically regenerated using Delaunay triangulation. The utility of this method is verified through the comparison of the performances of the optimal geometry and those of the initial geometry.