BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Nonparametric preprocessing in system identification: a powerful t
 ool.
DTSTART:20101105T101500
DTSTAMP:20260407T051008Z
UID:ac3a4205b65962e0d7961f0b9d7e5e2382d1d162bc388c52c4f617dc
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. J. Schoukens\, Dept. of Fundamental Electricity and Inst
 rumentation\, Vrije Universiteit Brussel\, Belgium\nMy dream is to develop
  robust identification methods that provide good models to the user withou
 t requesting advanced user interactions. High quality nonparametric freque
 ncy response function measurements together with nonparametric measurement
 s of the disturbing noise power spectra turn out to be important steps to 
 make this dream comes through. Nonparametric models can not only be used t
 o initialize more advanced identification methods\, the results can also b
 e used directly in modern control design approaches.\n \nIn this presentat
 ion we study first the properties of the existing nonparametric methods fo
 r estimating the plant and noise transfer functions of a linear dynamic sy
 stem. The analysis is based on the recent insight that leakage errors in t
 he frequency domain have a smooth nature that is completely similar to the
  initial transients in the time domain. This not only allows us to underst
 and better the existing classic methods\, but also opens the road to new b
 etter performing algorithms. These will be presented in the second part of
  the presentation. The presentation includes the output error setup\, the 
 errors-in-variables setup\, and measurements under feedback conditions. Ev
 entually\, some of the methods are illustrated on experimental data. 
LOCATION:ME C2405
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR