Seismic fragility estimation of electrical substations accounting for component damage and short circuit faults

Modern society relies heavily on electricity, which is transmitted from generating stations to final consumers through an electrical power grid. Electrical substations are key components of these grids. Previous earthquakes have heavily damaged some of these substations, affecting their functionality and leading to service interruptions. Functionality losses are usually modeled using fragility functions, which in general terms relate a seismic intensity measure with the probability of failure. Most previous studies use generic substation fragility functions that are not specific to the modeled substations. Indeed, power substations are composed of several internal components laid out in a wide range of different configurations, which cannot be accurately represented by these generic models. This study proposes a method to construct fragility functions based on the internal configuration of substation components and accounts for faults to individual lines within the substation and short circuit faults that render all the substation nonfunctional. The proposed method was applied to Chilean substations, resulting in fragility functions that vary significantly depending on their voltage level and their internal configuration. On average, the resulting fragility functions are fairly similar to the generic functions provided by HAZUS. However, fragility functions of individual substation archetypes can differ significantly between each other and with those of HAZUS. Thus, using fragility functions that consider a more realistic internal configuration of electrical components instead of generic functions can improve estimations of seismic performance, risk, and resilience of electric power grids, and hence help in providing better tools to prepare and mitigate earthquake effects.