Assessing survivability to support power grid investment decisions. (November 2016)

• Record Type: Journal Article
• Authors: Koziolek, Anne; Avritzer, Alberto; Suresh, Sindhu; Menasché, Daniel S.; Diniz, Morganna; de Souza e Silva, Edmundo; Leão, Rosa M.; Trivedi, Kishor; Happe, Lucia
• Is Part Of: Reliability engineering & system safety. Volume 155(2016:Nov.)
• Page Start: 30
• Page End: 43
• Publication Date: 2016-11
• Subjects: Power grids -- Optimization -- Reactive power -- Active power -- Survivability; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2016.05.015
• Languages: English
• ISSNs: 0951-8320

Abstract

Abstract: The reliability of power grids has been subject of study for the past few decades. Traditionally, detailed models are used to assess how the system behaves after failures. Such models, based on power flow analysis and detailed simulations, yield accurate characterizations of the system under study. However, they fall short on scalability. In this paper, we propose an efficient and scalable approach to assess the survivability of power systems. Our approach takes into account the phased-recovery of the system after a failure occurs. The proposed phased-recovery model yields metrics such as the expected accumulated energy not supplied between failure and full recovery. Leveraging the predictive power of the model, we use it as part of an optimization framework to assist in investment decisions. Given a budget and an initial circuit to be upgraded, we propose heuristics to sample the solution space in a principled way accounting for survivability-related metrics. We have evaluated the feasibility of this approach by applying it to the design of a benchmark distribution automation circuit. Our empirical results indicate that the combination of survivability and power flow analysis can provide meaningful investment decision support for power systems engineers. Abstract : Highlights: We propose metrics and models for scalable survivability analysis of power systems. The survivability model captures the system phased-recovery, from failure to repair. The survivability model is used as a building block of an optimization framework. Heuristics assist in investment options accounting for survivability-related metrics.