Redes neurais artificiais e computação evolucionária aplicados em modelagem de um turbo-gerador

Abstract

This paper evaluates the application of the computational intelligence methodologies in a nonlinear process identification. The different intelligent methodologies utilized are evolutionary computation (hybrid genetic algorithms with simulated annealing) and artificial neural networks (feedforward and recurrent topologies). The simulations are realized in the identification of a turbogenerator mathematical model through a step signal, pseudo-random binary sequence, and white noise excitation signals. The performance of the techniques are presented and discussed.