Publication: Automatic differentiation for sensitivity calculation in electromagnetism: Application for optimization of a linear actuator
Introduction
Applications
Tools
Research Groups
Workshops
Publications
   List Publications
   Advanced Search
   Info
   Add Publications
My Account
About

Automatic differentiation for sensitivity calculation in electromagnetism: Application for optimization of a linear actuator

- Article in a journal -
 

Area
Electrical Engineering

Author(s)
Petre Enciu , Laurent Gerbaud , Frédéric Wurtz

Published in
IEEE Transactions on Magnetics

Year
2011

Publisher
IEEE Magnetics Society

Abstract
Automatic differentiation (ad) is introduced as a powerful technique to compute derivatives of functions given in the form of computer programs in high-level programming languages such as FORTRAN, C, or C++. This paper applies ad to compute error-free gradients of electromagnetic device sizing models. Then, the obtained gradients are exploited in optimization to size electromagnetic devices by means of minimizing a cost function with constrained parameters and performances. Often, the electromagnetic devices models have to be described not only by analytical formulas, but also by algorithms. This paper proposes an electromagnetic model of a linear actuator dealing with implicit equations solved by numerical algorithms. The ADOL-C package is considered for automatic differentiation.

AD Tools
ADOL-C

BibTeX
@ARTICLE{
         Enciu2011Adf,
       title = "Automatic differentiation for sensitivity calculation in electromagnetism: Application
         for optimization of a linear actuator",
       author = "Petre Enciu and Laurent Gerbaud and Fr{\'e}d{\'e}ric Wurtz",
       publisher = "IEEE Magnetics Society",
       year = "2011",
       journal = "IEEE Transactions on Magnetics",
       volume = "47",
       pages = "1238--1241",
       abstract = "Automatic differentiation (AD) is introduced as a powerful technique to compute
         derivatives of functions given in the form of computer programs in high-level programming languages
         such as FORTRAN, C, or C++. This paper applies AD to compute error-free gradients of electromagnetic
         device sizing models. Then, the obtained gradients are exploited in optimization to size
         electromagnetic devices by means of minimizing a cost function with constrained parameters and
         performances. Often, the electromagnetic devices models have to be described not only by analytical
         formulas, but also by algorithms. This paper proposes an electromagnetic model of a linear actuator
         dealing with implicit equations solved by numerical algorithms. The ADOL-C package is considered for
         automatic differentiation.",
       ad_area = "Electrical Engineering",
       ad_tools = "ADOL-C",
       doi = "10.1109/TMAG.2010.2073452"
}


back
  

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