Identificación robusta de un Proceso Biomédico mediante Algoritmos Evolutivos

In this paper, a non-linear robust identification (RI) methodology to characterize the feasible parameter set (FPS), when the identification error is unknown but bounded simultaneously by several norms, is presented. For that, the Robust Identification (RI) problem is transformed into a multimodal optimization one with an infinite number of global minima which constitute the FPS. For the optimization task a special Genetic Algorithm (e-GA), inspired by Multiobjective Evolutionary Algorithms (MOEA), is presented, which characterizes the FPS by means of a discrete set of models (FPS*) well distributed along the FPS. An application example to a biomedical model which shows the blockage that produces a given drug on the ionic currents of a cardiac cell is presented to illustrate the methodology.