Elicitation of Structured Expert Judgement to estimate the probability of a major power system unreliability event

Resilience of a power system is concerned with preventing adverse outcomes from disturbances to the system, containing them when they do occur and recovering from them as quickly and safely as possible.
Major unreliability events worldwide over many years have shown a number of common phenomena, e.g. cascades of outages and frequency or voltage instability. However, the precise pathway is different every time and dependent on complex and uncertain system behaviour. Events also differ based on the characteristics of each system and social factors such as the priorities and judgements of individuals such as key control room staff. Moreover, major
unreliability events remain very rare. The use of modelling to estimate the probability of a regional or whole system shutdown is therefore extremely difficult.
This paper concerns an approach to the estimation of the probability of a major power system unreliability event. The general approach – use of Structured Expert Judgement (SEJ) – has been used in many fields in which major risks need to be understood in respect of phenomena that are not readily amenable to modelling. This paper describes what the authors believe to be the world’s first application of the approach to understanding and enhancement of power system
resilience. It outlines a conceptual model of how a system blackout might happen and describes how it is used with a group of experts to provide estimates of the likelihood of a system disturbance propagating along different paths.