Titulo del Seminario: Logica difusa y redes neuronales para la modelizacion de sistemas no lineales. Conferencista: Dr. Wen Yu, Departamento de Control, CINVESTAV-IPN. Abstract: Fuzzy systems can approximate any continuous nonlinear function to arbitrary accuracy, provided that suitable fuzzy rules are available.Fuzzy modeling includes two stages: structure identification (fuzzy rules) and parameter identification (membership functions). Online clustering, support vector machines, hierarchical and recurrent techniques will be applied to determine the structure of the fuzzy system. Input-to-state stability and ellipsoid methods will be used to update the membership functions. An application on modeling of crude oil blending is proposed.
