The determination of operational security for an electrical grid, for given operating conditions, has been traditionally done using deterministic criteria, meaning that an operating condition is identified as secure if it can withstand the effects (i.e. if no grid constraints are violated) of every contingency in a pre-specified contingency set. Although such a deterministic framework is conceptually straightforward, it does not directly and explicitly mirror the level of operational risk that the power system actually faces.
It is thus questionable if this approach is appropriate for evaluating the short-, mid-, and long-term reliability and resiliency of a system. An adequate measure of risk should rather consider both the quantified consequences of an undesirable outcome as well as its probability of occurrence, and all the more so in the presence of stochastically varying RES injections and within a liberalised and competitive market environment, which will prompt the need for more formally advanced approaches that yield insight into system behaviour and deliver reliable countermeasures. To deal with such issues the recent past has witnessed a substantial amount of work devoted to the development of security assessment methods explicitly based on probabilistic and stochastic methods, but not all approaches appear to be equally viable in terms of modelling assumptions and analytical complexity, necessitating further work to render them suitable for the effective implementation and deployment in a real-life industrial setting.