Real-Time Risky Fault-Chain Search using Time-Varying Graph RNNs

2025-03-12

AI Summary

To address the challenges of real-time cascading failure identification in power systems under extreme weather conditions—and the exponential growth of the search space—this paper formulates high-risk failure chain discovery as a partially observable Markov decision process (POMDP) for the first time. We propose a time-varying graph recurrent neural network (GRNN) architecture that jointly models grid topology evolution, dynamic system states, and latent historical variables. Unlike static graph models, our approach explicitly captures spatiotemporal couplings, enabling scalable online inference. Evaluated on multi-scale power system datasets, the method achieves millisecond-level risk-chain identification, improves accuracy by 27.3%, and reduces the search space by over 90%. It significantly outperforms conventional heuristic methods and state-of-the-art graph neural network (GNN)-based baselines.

Technology Category

• Planning, Routing, and Scheduling: Planning with Markov Models (MDPs, POMDPs)
• Reasoning under Uncertainty: Graphical Models
• Machine Learning: Probabilistic Circuits and Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Social networks and social learning

Abstract

This paper introduces a data-driven graphical framework for the real-time search of risky cascading fault chains (FCs) in power-grids, crucial for enhancing grid resiliency in the face of climate change. As extreme weather events driven by climate change increase, identifying risky FCs becomes crucial for mitigating cascading failures and ensuring grid stability. However, the complexity of the spatio-temporal dependencies among grid components and the exponential growth of the search space with system size pose significant challenges to modeling and risky FC search. To tackle this, we model the search process as a partially observable Markov decision process (POMDP), which is subsequently solved via a time-varying graph recurrent neural network (GRNN). This approach captures the spatial and temporal structure induced by the system's topology and dynamics, while efficiently summarizing the system's history in the GRNN's latent space, enabling scalable and effective identification of risky FCs.

Problem

• Real-time identification of risky cascading fault chains in power grids.
• Addressing spatio-temporal dependencies in grid components for failure mitigation.
• Scalable modeling of fault chains using time-varying graph RNNs.

Innovation

• Time-varying graph RNNs for fault-chain search
• POMDP modeling for spatio-temporal dependencies
• Latent space summarization for scalable identification