Networked System Identification

theory and applications

Master thesis (2022)

Authors

S. FANG Electrical Engineering, Mathematics and Computer Science

Contributors

Robert Kooij Network Architectures and Services - EEMCS (mentor)
I. Jokic Network Architectures and Services - EEMCS (mentor)
Faculty
Electrical Engineering, Mathematics and Computer Science, Electrical Engineering, Mathematics and Computer Science
Copyright: © 2022 SONGLEI FANG
More Info
expand_more

Published Date

29-08-2022

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Abstract

A complex network consists of the underlying topology, defined by a graph and the dynamical processes taking place on a network, defined by a set of governing equations. In this thesis, we deploy the discrete-time linear state-space (DLSS) model to identify the dynamical processes taking place on a complex network. Unlike the black-box identification approach, we split the network into dynamical units and identify the dynamics of each dynamical unit independently. Next, we relate input/output vectors of individual dynamical units, based on the underlying topology and provide the model for the dynamics of the entire network. Because we use a linear model, by scaling the model to the entire network, no information is lost about dynamical processes of individual units.
In this thesis, we apply this new networked system identification solution to two real-world complex networks, water and road networks, and find this identification approach to successfully improve the identification performance compared to the common black-box identification approach.