Cascading power outages propagate locally in an influence graph that is not the actual grid topology

Abstract

In a cascading power transmission outage, component outages propagate nonlocally; after one component outages, the next failure may be very distant, both topologically and geographically. As a result, simple models of topological contagion do not accurately represent the propagation of cascades in power systems. However, cascading power outages do follow patterns, some of which are useful in understanding and reducing blackout risk. This paper describes a method by which the data from many cascading failure simulations can be transformed into a graph-based model of influences that provides actionable information about the many ways that cascades propagate in a particular system. The resulting "influence graph" model is Markovian, in that component outage probabilities depend only on the outages that occurred in the prior generation. To validate the model, we compare the distribution of cascade sizes resulting from n-2 contingencies in a 2896 branch test case to cascade sizes in the influence graph. The two distributions are remarkably similar. In addition, we derive an equation with which one can quickly identify modifications to the proposed system that will substantially reduce cascade propagation. With this equation, one can quickly identify critical components that can be improved to substantially reduce the risk of large cascading blackouts. KeywordsBig data, cascading failure, complex networks Abstract-In a cascading power transmission outage, component outages propagate nonlocally; after one component outages, the next failure may be very distant, both topologically and geographically. As a result, simple models of topological contagion do not accurately represent the propagation of cascades in power systems. However, cascading power outages do follow patterns, some of which are useful in understanding and reducing blackout risk. This paper describes a method by which the data from many cascading failure simulations can be transformed into a graph-based model of influences that provides actionable information about the many ways that cascades propagate in a particular system. The resulting "influence graph" model is Markovian, in that component outage probabilities depend only on the outages that occurred in the prior generation. To validate the model, we compare the distribution of cascade sizes resulting from n − 2 contingencies in a 2896 branch test case to cascade sizes in the influence graph. The two distributions are remarkably similar. In addition, we derive an equation with which one can quickly identify modifications to the proposed system that will substantially reduce cascade propagation. With this equation, one can quickly identify critical components that can be improved to substantially reduce the risk of large cascading blackouts.

Cascading blackout overall structure and some implications for sampling and mitigation

“…This is useful in mitigating the triggers associated with cascading but root cause analysis does not address the causes or mitigation of propagation. However, it is becoming feasible to relate candidate mitigations such as line upgrades to reductions in propagation or large blackout risk [25][26][27]. ... Fig.…”

Estimation for Expected Energy Not Served of Power Systems Using the Screening Methodology of Cascading Outages in South Korea

“…As a result, it can be expressed as Equation (2) [27,28]. Where Z m is the number of outages in m generation and Z m+1 means the number in m + 1 generation.…”

Failure Localization in Power Systems via Tree Partitions

“…Because of the intricate interactions among power system components, outages may cascade and propagate in a very complicated, non-local manner [3]- [5], exhibiting very different patterns for different networks [6]. Such complexity originates from the interplay between network topology and power flow physics, and is aggravated by possible hidden failures [7] and human errors [8].…”

“…There are three traditional approaches for characterizing the behavior of cascades in the literature: (i) using simulation models [9] that rely on Monte-Carlo approaches to account for the steady state power flow redistribution on DC [5], [8], [10], [11] or AC [12]- [14] models; (b) studying purely topological models that impose certain assumptions on the cascading dynamics (e.g., failures propagate to adjacent lines with high probability) and infer component failure propagation patterns from graph-theoretic properties [16]- [18]; (c) investigating simplified or statistical cascading failure dynamics [3], [21], [22], [24]. In each of these approaches, it is typically challenging to make general inferences across different scenarios due to the lack of structural understanding of power redistribution after line failures.…”