Domain Knowledge Enhanced Maximum Flow Algorithms for FACTS Device Control
Department
Computer Science
Major
Computer Science
Research Advisor
Tauritz, Daniel R.
Advisor's Department
Computer Science
Abstract
One solution to the problem of preventing cascading failures in the electrical power system is the use of power electronics, such as Flexible AC Transmission System (FACTS) devices. In order to effectively employ them we must be able to calculate control settings quickly enough to mitigate a cascading failure. In this paper, we abstract away the complexity of the power grid by modeling it as a directed graph, which allows us to use the maximum flow algorithm to determine control settings. We investigate two heuristic methods for incorporating domain knowledge into the maximum flow algorithm as well as a generalized maximum flow algorithm that can be used to model line losses. The heuristic methods failed to significantly reduce overloads, but the generalized maximum flow algorithm did manage to accomplish this through more accurate control of the system.
Biography
Evan Wright is a sophomore attending the University of Missouri--Rolla, and is pursuing a dual major in Computer Science and Applied Mathematics.
Research Category
Engineering
Presentation Type
Oral Presentation
Document Type
Presentation
Presentation Date
12 Apr 2006, 10:30 am
Domain Knowledge Enhanced Maximum Flow Algorithms for FACTS Device Control
One solution to the problem of preventing cascading failures in the electrical power system is the use of power electronics, such as Flexible AC Transmission System (FACTS) devices. In order to effectively employ them we must be able to calculate control settings quickly enough to mitigate a cascading failure. In this paper, we abstract away the complexity of the power grid by modeling it as a directed graph, which allows us to use the maximum flow algorithm to determine control settings. We investigate two heuristic methods for incorporating domain knowledge into the maximum flow algorithm as well as a generalized maximum flow algorithm that can be used to model line losses. The heuristic methods failed to significantly reduce overloads, but the generalized maximum flow algorithm did manage to accomplish this through more accurate control of the system.
