Optimizing vegetation management to improve the resilience of the electrical power system to extreme weather

The reliability of the electrical grid is a matter of both national security and qualify of life. The frequency of weather-related outages are expected to increase due to climate change. Electrical utilities in the United States spend over $70 billion annually on storm restoration. The number of outages and cost can be significantly decreased with better information on vegetation risk near critical infrastructure and assessing key vulnerabilities. StormImpact Inc. will use NASA Earth Observation data to build machine learning (ML) vegetation growth and risk models that will allow electrical utilities to proactively prepare for storms and improve their infrastructure resilience. We anticipate cost savings of 10% to 20% for utility customers due to improved vegetation management efficiency. We will leverage these data to provide enhanced StormImpact prediction models and a new national power outage forecast model. These innovations will greatly improve storm preparedness, optimize vegetation management, and improve the resiliance of the electrical grid in a changing climate.