# Exploratory analysis of high-resolution power interruption data reveals spatial and temporal heterogeneity in electric grid reliability

## TL;DR

This study analyzes three years of high-resolution power interruption data revealing significant spatial and temporal variability in grid reliability, highlighting the limitations of aggregated metrics and advocating for more granular data access to improve policy analysis.

## Contribution

It demonstrates the heterogeneity in grid performance hidden by aggregated metrics and advocates for increased access to detailed data for better policy insights.

## Key findings

- Metrics vary significantly across locations and time
- Granular data reveals heterogeneity not seen in public reports
- Limited data access hampers detailed policy analysis

## Abstract

Modern grid monitoring equipment enables utilities to collect detailed records of power interruptions. These data are aggregated to compute publicly reported metrics describing high-level characteristics of grid performance. The current work explores the depth of insights that can be gained from public data, and the implications of losing visibility into heterogeneity in grid performance through aggregation. We present an exploratory analysis examining three years of high-resolution power interruption data collected by archiving information posted in real-time on the public-facing website of a utility in the Western United States. We report on the size, frequency and duration of individual power interruptions, and on spatio-temporal variability in aggregate reliability metrics. Our results show that metrics of grid performance can vary spatially and temporally by orders of magnitude, revealing heterogeneity that is not evidenced in publicly reported metrics. We show that limited access to granular information presents a substantive barrier to conducting detailed policy analysis, and discuss how more widespread data access could help to answer questions that remain unanswered in the literature to date. Given open questions about whether grid performance is adequate to support societal needs, we recommend establishing pathways to make high-resolution power interruption data available to support policy research.