Development of metrics for resilience quantification in energy systems

Abstract

The design of energy systems usually requires technical, economical and environmental analysis. However, the growth of systems failure due to unpredictable low-probability external events makes the consideration of resilience in this design also important. Although there is no standard metric for resilience quantification yet, it is known that it should consider system configuration, operation time and total or partial energy generation during and after the event, as well as the components repair probability and time. A proposal for resilience quantification in four cogeneration plants was previously developed based on components stochastic failures and verification of their consequences in the plant energy generation. The present work aims to continue the development of this metric by including in its calculation the repair probability of the components, their repairing time and the plant downtime during the repair, essential parameters for resilience quantification. Two new metrics are proposed and simulations with 0, 50% and 75% of repair probability of the components are made in software CLIPS. One of the metrics is able to evaluate the influence of repairment in system resilience, while the other one predicts plant downtime during operation. The metrics point to S#2 as the most resilient system and S#3 as the most affect by repairing.