NHESS - Probabilistic and machine learning methods for uncertainty quantification in power outage prediction due to extreme events

Abstract. Strong hurricane winds damage power grids and cause cascading power failures. Statistical and machine learning models have been proposed to predict the extent of power disruptions due to hurricanes. Existing outage models use inputs including power system information, environmental parameters, and demographic parameters. This paper reviews the existing power outage models, highlighting their strengths and limitations. Existing models were developed and validated with data from a few utility companies and regions, limiting the extent of their applicability across geographies and hurricane events. Instead, we train and validate these existing outage models using power outages from multiple regions and hurricanes, including hurricanes Harvey (2017), Michael (2018), and Isaias (2020), in 1910 US cities. The dataset includes outages from 39 utility companies in Texas, 5 in Florida, 5 in New Jersey, and 11 in New York. We discuss the limited ability of state-of-the-art machine learning models to (1) make bounded outage predictions, (2) extrapolate predictions to high winds, and (3) account for physics-informed outage uncertainties at low and high winds. For example, we observe that existing models can predict outages higher than the number of customers (in 19.8 % of cities with an average overprediction ratio of 5.2) and cannot capture well the outage variance for high winds, especially above 70 m s−1. Our findings suggest that further developments are needed for power outage models for proper representation of hurricane-induced outages.

NHESS - Probabilistic and machine learning methods for uncertainty quantification in power outage prediction due to extreme events Articles | Volume 23, issue 5 Article Assets Peer review Metrics Related articles Articles | Volume 23, issue 5 https://doi.org/10.5194/nhess-23-1665-2023 © Author(s) 2023. This work is distributed under the Creative Commons Attribution 4.0 License. Special issue: Advances in machine learning for natural hazards risk... https://doi.org/10.5194/nhess-23-1665-2023 © Author(s) 2023. This work is distributed under the Creative Commons Attribution 4.0 License.

saved by

• David A. Lee

related reading

• The Unreasonable Difficulty of Time Series Forecastingsuzyahyah.github.io
• GraphCast: AI model for faster and more accurate global weather forecastingdeepmind.google
• [2510.13786] The Art of Scaling Reinforcement Learning Compute for LLMsarxiv.org
• Neural general circulation models for weather and climatenature.com
• [2201.02177] Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasetsarxiv.org
• [2606.25076] Machine learning is revolutionizing weather forecasting -- the next step is a change in how we workarxiv.org
• There Is No Power - by Strategist and Architect - TSCStscsw.substack.com
• Bracketing Inference with Uncertainty Quantification: A Reliability Pipeline for Neural Aerodynamic Surrogatesresearchsquare.com
• Modeling the uncertainty on the covariance matrix for probabilistic forecast reconciliationarxiv.org
• [2310.03775] Hidden Markov Models for Stock Market Predictionarxiv.org
• Large Language Models Must Be Taught to Know What They Don't Knowarxiv.org
• Predicting LLM Safety Before Release by Simulating Deploymentcdn.openai.com