Demystifying surrogate modeling for circuits and systems

Mustafa Berke Yelten*, Ting Zhu, Slawomir Koziel, Paul D. Franzon, Michael B. Steer

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

148 Citations (Scopus)

Abstract

Surrogate models are used in grey-box or black-box modeling of a wide variety of systems including electromagnetic modeling of complex structures, geological distributions of minerals, interaction of airflows with airfoils, chemical processes, to name just a few. Only recently have surrogate models been used explicitly in electronics. Surrogate modeling is a macromodeling technique which has solid mathematical foundations and most often, has a strong intuitive linkage to the physical world. So when did surrogate modeling start? One could say that it has always been with us but the term in its current usage was coined by Osio and Amon in 1996 [1]. This work was preceded in 1994 in a paper by Bandler et al. in which design tools, is a surrogate model that is continuously refined.

Original languageEnglish
Article number6155115
Pages (from-to)45-63
Number of pages19
JournalIEEE Circuits and Systems Magazine
Volume12
Issue number1
DOIs
Publication statusPublished - Mar 2012
Externally publishedYes

Funding

The paper has the distribution statement “A” (Approved for Public Release, Distribution Unlimited). The views expressed are those of the authors and do not reflect the official policy or position of the Department of Defense or the U.S. Government. This work was supported by the Self-HEALing mixed-signal Integrated Circuits (HEALICs) program of the Defense Advanced Research Projects Agency (DARPA) and the prime contractor Raytheon Company (Contract number: FA8650-09-C-7925).

FundersFunder number
Self-HEALing mixed-signal Integrated Circuits
Defense Advanced Research Projects Agency
Raytheon CompanyFA8650-09-C-7925

    Fingerprint

    Dive into the research topics of 'Demystifying surrogate modeling for circuits and systems'. Together they form a unique fingerprint.

    Cite this