Prof. Alain Vande Wouwer
Universidad de Mons (Bélgica)
Días 20 y 21 de febrero de 16 a 19 horas.
Lugar: Aula 009 el día 20 y Aula 008 el día 21.
Recomendable traer ordenador con Matlab.
Resumen: Dynamic models of physical systems usually consist of sets of (non)linear differential (or difference) equations. These equations contain unknown parameters whose values have to be inferred from experimental data by minimizing the distance between the model prediction and the measured data. This course introduces various concepts that are useful in order to achieve a successful identification procedure. In particular, attention will be focused on
a) distance between model prediction and data
b) basic notions and tools for the minimization of this distance
c) parametric sensitivity analysis
d) identifiability and the Fisher Information Matrix
e) linear and nonlinear least squares
f) maximum likelihood estimation
g) sparse identification
The course is illustrated with several Matlab examples.
An identification project is also proposed to the course participants.
Breve CV: Alain Vande Wouwer received his master degree in electrical engineering from Faculté Polytechnique de Mons, Belgium, in 1988, and his Ph.D. degree in applied sciences from the same university (together with a European doctorate degree jointly with Stuttgart Universität, Germany) in 1994. In 1994, he achieved a postdoctoral research stay in the department of mechanical engineering at Laval University, Quebec. Since 2009, he has been a full professor in the department of electrical engineering of the University of Mons, Belgium, and is currently the head of the Systems, Estimation, Control, and Optimization group (SECO). He is the co-author of two books on numerical analysis of partial differential equations, and of more than 250 conference and journal articles on various aspects of process modeling and control. His research interests include nonlinear dynamics, parameter and state estimation, dynamic optimization, and process control, with applications to biochemical processes and unmanned aerial vehicles.