Optimal passivity enforcement of state-space models via localization methods

Author

Mahmood, Zohaib ; Grivet-Talocia, Stefano ; Calafiore, Giuseppe C. ; Daniel, Luca

Author_Institution

Dept. Electr. Eng. & Comput. Sci., MIT, Cambridge, MA, USA

fYear

2014

Firstpage

89

Lastpage

92

Abstract

This work presents a novel approach for passivity enforcement of state-space macromodels, based on nonsmooth localization algorithms applied to a convex formulation of the passivity constraints. The main advantages of proposed scheme are guaranteed optimality and limited required computational resources. Compared to convex formulations based on a direct implementation of Bounded Real Lemma we are able to reduce both the memory and time requirements by orders of magnitude.

Keywords

convex programming; integrated circuit modelling; state-space methods; bounded real Lemma; computational resources; convex formulations; localization methods; memory requirements; nonsmooth localization; optimal passivity enforcement; passivity constraints; state-space macromodels; time requirements; Computational modeling; Ellipsoids; Integrated circuit interconnections; Integrated circuit modeling; Mathematical model; Memory management; Solid modeling;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical Performance of Electronic Packaging and Systems (EPEPS), 2014 IEEE 23rd Conference on

Print_ISBN

978-1-4799-3641-0

Type

conf

DOI

10.1109/EPEPS.2014.7103602

Filename

7103602