Publication: Nonlinear system identification employing automatic differentiation
Introduction
Applications
Tools
Research Groups
Workshops
Publications
   List Publications
   Advanced Search
   Info
   Add Publications
My Account
About
Impress

Nonlinear system identification employing automatic differentiation

- Article in a journal -
 

Area
Data Assimilation

Author(s)
Jan Schumann-Bischoff , Stefan Luther , Ulrich Parlitz

Published in
Communications in Nonlinear Science and Numerical Simulation

Year
2013

Publisher
Elsevier

Abstract
An optimization based state and parameter estimation method is presented where the required Jacobian matrix of the cost function is computed via automatic differentiation. Automatic differentiation evaluates the programming code of the cost function and provides exact values of the derivatives. In contrast to numerical differentiation it is not suffering from approximation errors and compared to symbolic differentiation it is more convenient to use, because no closed analytic expressions are required. Furthermore, we demonstrate how to generalize the parameter estimation scheme to delay differential equations, where estimating the delay time requires attention.

AD Tools
ADOL-C

BibTeX
@ARTICLE{
         Schumann-Bischoff2013Nsi,
       title = "Nonlinear system identification employing automatic differentiation",
       author = "Jan Schumann-Bischoff and Stefan Luther and Ulrich Parlitz",
       publisher = "Elsevier",
       year = "2013",
       journal = "Communications in Nonlinear Science and Numerical Simulation",
       volume = "18",
       pages = "2733–2742",
       abstract = "An optimization based state and parameter estimation method is presented where the
         required Jacobian matrix of the cost function is computed via automatic differentiation. Automatic
         differentiation evaluates the programming code of the cost function and provides exact values of the
         derivatives. In contrast to numerical differentiation it is not suffering from approximation errors
         and compared to symbolic differentiation it is more convenient to use, because no closed analytic
         expressions are required. Furthermore, we demonstrate how to generalize the parameter estimation
         scheme to delay differential equations, where estimating the delay time requires attention.",
       ad_area = "Data Assimilation",
       ad_tools = "ADOL-C"
}


back
  

Contact:
autodiff.org
Username:
Password:
(lost password)