Nonlinear robust identification using multiobjective evolutionary algorithms

Abstract

In this article, a procedure to estimate a nonlinear models set (Θp) in a robust identification context, is presented. The estimated models are Pareto optimal when several identification error norms are considered simultaneously. A new multiobjective evolutionary algorithm ε↗-MOEA has been designed to converge towards ΘP *, a reduced but well distributed representation of ΘP since the algorithm achieves good convergence and distribution of the Pareto front J(Θ). Finally, an experimental application of the ε↗-MOEA algorithm to the nonlinear robust identification of a scale furnace is presented. The model has three unknown parameters and ℓ∞ and ℓ1 norms are been taken into account. © Springer-Verlag Berlin Heidelberg 2005.