Gated Graph Attention Networks for Predicting Duration of Large Scale Power Outages Induced by Natural Disasters

TL;DR

This paper introduces BiGGAT, a novel graph neural network model combining GAT and GRU, to accurately predict the duration of large-scale power outages caused by natural disasters, addressing complex spatial dependencies.

Contribution

The paper presents BiGGAT, a new graph-based neural network that effectively models spatial and temporal dependencies for outage duration prediction under real-world challenges.

Findings

BiGGAT outperforms benchmark models in outage duration prediction.

The model effectively captures high-order spatial dependencies.

Experimental results validate the approach's superior performance.

Abstract

The occurrence of large-scale power outages induced by natural disasters has been on the rise in a changing climate. Such power outages often last extended durations, causing substantial financial losses and socioeconomic impacts to customers. Accurate estimation of outage duration is thus critical for enhancing the resilience of energy infrastructure under severe weather. We formulate such a task as a machine learning (ML) problem with focus on unique real-world challenges: high-order spatial dependency in the data, a moderate number of large-scale outage events, heterogeneous types of such events, and different impacts in a region within each event. To address these challenges, we develop a Bimodal Gated Graph Attention Network (BiGGAT), a graph-based neural network model, that integrates a Graph Attention Network (GAT) with a Gated Recurrent Unit (GRU) to capture the complex spatial characteristics. We evaluate the approach in a setting of inductive learning, using large-scale power outage data from six major hurricanes in the Southeastern United States. Experimental results demonstrate that BiGGAT achieves a superior performance compared to benchmark models.