N-k Failure Analysis Algorithm for Identification of Extreme Events for Cascading Outage Pre-screening process

The electric power grid is prone to severe contingencies which degrade the reliability, resilience and security of the power system. Current practices assess the risk posed by contingencies though steady-state methods like N-1, N-2 and N-1-1 analyses. While these methods have been tested and used for decades, the increasing frequency of blackouts and major outages demonstrate that these methods do not fully capture the contingency size and consequences such as load loss. In other words, many contingencies exhibit phenomena, such as cascading failures, that go beyond conventional steady-state models. Unfortunately, the computational requirements of full contingency assessment with detailed models remains out of reach for the modern grid operators. To address this challenge, this paper explores the suitability of using steady-state, N-k analysis to identify a smaller set of contingencies that require detailed cascading analysis, including protection system operation and dynamic simulation. The method is demonstrated on the IEEE 300-bus test system for N-1 and N-2 contingencies and a detailed analysis is provided to determine situations where N-k analysis correctly (or incorrectly) identifies the contingencies that require further analysis. The reasons for discrepancies of results of conventional steady state and advanced analyses has been identified.