Fatigue Curves in Steel Railway Generated by Genetic Algorithm

Resumo

Genetic Algorithms (GA) are global optimization, based on natural selections and genetic mechanisms
for which structured and random search strategies are geared by reinforcing the search for high amplitude points
in which the function to be minimized (maximized) has relatively low (high) values. This work deals with a study
on the Genetic Algorithm application to estimate parameters inherent to S-N-p Curves generated for typically
structural details of the metal railway bridges applying together with The Maximum Likelihood Method to infer
runout data, such as censored data for failure probability functions. MatLab software is employed to develop the
computational program. To compare the obtained results, the Interior Point Algorithm is also presented, due to its
wide use in problems involving linear and nonlinear quadratic programming.