Genetic Algorithms And Boundary Elements: Some Applications In Optimization Of 2d Potential Problems.

Abstract

The present work introduces the application of genetic algorithms (GA) to topology
optimizations of 2D potential problems using the boundary element method (BEM). Initially, a
brief description of GA basis is presented. Next, some elementary ideas on its application along
with BEM procedures are summarized. The proposed approach is able to create cavities where
they are less influent or simply move the existing ones, in order to, extremize a given cost
function. Therefore it is possible to optimize the location of key features of the domain, like
heating lines, as well as to generate the optimal topology of the problem. The performance of the
proposed algorithm is assessed by some examples and discussed.