Resilient control system metrics

Abstract

Resilience of a system, particularly critical infrastructure, is of great interest to utilities and stakeholders. Consequences of natural or man-made events (e.g., Superstorm Sandy, the sequence of storms affecting the Caribbean and coast of the United States in 2017, and the Ukrainian power grid attack) have led to emphasis and increased interest in improving resilience. However, measurement of resilience in an absolute or relative manner has been achieved in a patchwork manner. This chapter will provide the description of metric development for an electricity distribution network and begin with a definition of resilience. That definition must meet the expectation of being understandable in its linguistic form that sets a goal that would lead to systems that would stand up to large disturbances of many possible types in a manner that is quantifiable by a set of metrics. The metrics that measure a system’s absolute or relative improvement is then developed for that purpose. Many notional definitions and attempts at resilience metrics formation have been coined by venerable organizations, such as the EPRI, ICS-CERT, DHS, National Academy of Sciences, and others. Common words among these sources are the ability to withstand or resist, survive, and respond expeditiously such that operations or life returns to normal as quickly as possible. To make resilience quantifiable, a proper definition of resilience and the relationship between it and reliability will facilitate development of useful resilience metrics and resilient grid architectures. The definition that is used as the basis has been developed by researchers in the community studying resilient controls over the past decade and can be found in Rieger and Rieger, Gertman, and McQueen: