• Energy Research
• other engineering and technologies close

Abstract The term resilience describes the ability to survive and quickly recover from extreme and unexpected disruptions. A high energy system resilience is of utmost importance to modern societies that are highly dependent on continued access to energy services. This review covers the terminology of energy system resilience and the assessment of a broad landscape of threats mapped with the proposed framework. A more detailed discussion on two specific threats are given: extreme weather, which is the cause for most of the energy supply disruptions, and cyberattacks, which still are a minor, but rapidly increasing concern. The framework integrates various perspectives on energy system threats by showcasing interactions between the parts of the energy system and its environment. Weather-related threats are discussed distinguishing relevant meteorological parameters and different durations of disruptions, increasingly related to the impacts of the climate change. Extremes in space weather caused by solar activity are very rare, but are nonetheless considered due to their potentially catastrophic impacts on a global scale. Digitalization of energy systems, e.g. through smart grids important to renewable electricity utilization, may as such improve resilience from traditional weather and technical failure threats, but it also introduces new vulnerabilities to cyberattacks. Major differences between the internet and smart grids limit the applicability of existing cybersecurity solutions to the energy sector. Other structural energy system changes will likely bring new threats, which call for updating the threat landscape for expected system development scenarios.

Abstract Increasing reliance on uninterrupted electricity supply against emerging threats such as climate change and cyberattacks calls for higher resilience of societies against power disruptions. A better understanding of social and economic impacts during these disruptions would be important for planning of resilience improvements. However, traditional energy system models rarely include these aspects. This paper presents an integrated framework containing a geospatial power system operation model, capable of emulating system component failures and restoration according to environmental conditions, with a link to spatial social and economic values such as population, economic activity, critical services and facilities. The framework was applied for analyzing the effects of uncontrolled and controlled power outages for two windy winter weeks in Finland. This case illustrated how controlled optimization could reduce the societal costs of such outage by shifting power shortage to regions where such costs are lower and in part by shifting the costs to other factors.

A small number of the strongest windstorms can account for a major share of power interruptions. In addition to large direct costs, such windstorms may shift the strategic direction of the energy system development. Thus, future system planning may benefit from a better understanding of the energy system resilience against the largest historical and historically unprecedented, but meteorologically plausible windstorms. Present analysis tools handling power disruptions tend to focus either on transmission grids or small distribution grids. However, neither focus captures well impacts concentrated on the distribution level over large area that includes high variety of grids and environments. This paper presents a fragility-based power disruption model against windstorms on a national scale with details on the local medium voltage grid. The model integrates synthetically generated power grids and consumption profiles with fragility functions and windstorm severity dependent fixing times. Grids are generated by spatial mapping of the grid component totals for each distribution grid operator onto municipalities and assuming symmetrical grid topologies within the municipalities. The model is applied to reproduce the national lost load profile for year 2011 winter windstorms in Finland. The modeled profile reproduces the historical reference data with a RMSE of 9% of the outage peak. Peer reviewed