Dissertation
CONSEQUENCES OF TOPOLOGICAL CHANGES IN CYBER-PHYSICAL NETWORKS: A system-theoretic perspective
Doctor of Philosophy (PhD), Washington State University
01/2014
Handle:
https://hdl.handle.net/2376/112105
Abstract
Motivated by emerging needs in network controls, we explore the implications of local topological changes to dynamic networks, broadly defined, on their behavior. Specifically, we characterize the impacts of 1) changes to network structure, 2) local actuation, and 3) monitoring and observation capabilities, on the operators' ability to achieve control and monitoring tasks on the network. Particularly, we study these changes from an application-agnostic perspective, in the context of canonical models used in the network-control literature. The results developed in the dissertation provide us with graph-theoretic insights into network modification and control that may have application across domains.
A key motivation for the thesis work is to contribute to a network control theory for cyber-physical infrastructures. In contrast with many multi-agent systems like robot teams, infrastructure networks typically have a very large scale, with complex multi-faceted dynamics spread over wide geographical regions at several time horizons. The scale of these systems limits stakeholders' observational and actuational abilities: often only limited local observation/actuation is possible. Likewise, natural events typically will only directly impact a portion of the network. Nevertheless, these local inputs and outputs may invoke complex propagative impacts across the networks. This dissertation seeks to understand these consequences of local changes (including actuations and probes) on the global dynamics of networked systems, using several canonical case studies. These results highlight several crucial links between network structures of the cyber-physical infrastructural systems and the operators' ability to achieve control and monitoring tasks.
The dissertation also pursues a particular infrastructure-network engineering problem: Air Traffic Flow Management (ATFM). Specifically, the dissertation contributes to an effort to strategize traffic management initiatives across the United States National Airspace System using model-driven automation. This dissertation contributes to the research effort on ATFM by pursuing modeling of weather impacts on network flows, particularly airport arrival and departure flows. The dissertation also phrases ATFM within the context of dynamical-network analysis and control. The models that we develop in this context may open a future pathway for bringing tools from networks control domain to the ATFM problem.
Metrics
4 File views/ downloads
14 Record Views
Details
- Title
- CONSEQUENCES OF TOPOLOGICAL CHANGES IN CYBER-PHYSICAL NETWORKS: A system-theoretic perspective
- Creators
- Rahul Dhal
- Contributors
- Sandip Roy (Advisor)Ali Saberi (Committee Member)Thomas R. Fischer (Committee Member)Yan Wan (Committee Member)Sean Warnick (Committee Member)Christine Taylor (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 307
- Identifiers
- 99900581643901842
- Language
- English
- Resource Type
- Dissertation