Hierarchical estimation for complex multi-domain dynamical systems

Pamela J. Tannous, Donald J. Docimo, Herschel C. Pangborn, Andrew G. Alleyne

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Complex power systems have dynamics spanning multiple energy domains and operating at multiple time scales. Hierarchical control has been proven to guarantee successful management of the coupling between the resulting fast transients and slow dynamics. It is usually prohibitively expensive or even infeasible to measure every signal in the system. Therefore, a reliable estimation framework that provides accurate estimates is vital to the success of the control design. This paper proposes a multi-level hierarchical estimation approach that can be used to supply reliable estimates to hierarchical controllers of complex multi-domain power systems. Models of complex multi-domain power systems can be accurately represented using graphs. System decomposition can be then achieved using clustering algorithms from graph theory. In this work, local estimates at each level of the hierarchical estimator are obtained using extended Kalman filters. A hierarchical estimator-controller is designed for an automotive electric vehicle as an illustrative example.

Original languageEnglish (US)
Title of host publication2019 American Control Conference, ACC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages909-915
Number of pages7
ISBN (Electronic)9781538679265
DOIs
StatePublished - Jul 2019
Event2019 American Control Conference, ACC 2019 - Philadelphia, United States
Duration: Jul 10 2019Jul 12 2019

Publication series

NameProceedings of the American Control Conference
Volume2019-July
ISSN (Print)0743-1619

Conference

Conference2019 American Control Conference, ACC 2019
CountryUnited States
CityPhiladelphia
Period7/10/197/12/19

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Hierarchical estimation for complex multi-domain dynamical systems'. Together they form a unique fingerprint.

Cite this