Thesis
Spanish
ID: <
http://hdl.handle.net/10251/143524>
Abstract
[EN] Along the process of calculation using heuristic algorithms, establishing the optimization error is not easy if the current solution is close to the solution of the problem. Due to that, the convergence in the last steps may be quite slow and sometimes the objective function drops abruptly. On the other hand, the result obtained depends on the discretization error level, in shape optimization of structural systems using the finite element method (FEM) as a calculator. Managing the stoppage of the optimization algorithm would be easily done by controlling the convergence error of the optimization process. Moreover, different margins of discretization could be used throughout the optimization process (higher level at the beginning and lower when approaching the solution of the problem). In this work, it is proposed to implement an optimization procedure using evolutionary algorithms and the finite element method with control of the optimization error and adaptation of discretization error. TFGM