Resilience of Power Grid Infrastructure with Renewables to Extreme Events in a Changing Climate

Prateek Arora
Ph.D. Candidate, New York University, USA

Research supported by:
Dr. Luis Ceferino
(Assistant Professor, Civil and Urban Engineering & Center for Urban Science and Progress, New York University, USA)

Abstract: Strong hurricane winds damage power grids and cause cascading power failures, resulting in critical service disruption and major economic losses. A resilient power system should minimize the extent of prolonged power disruptions. In this project, two different aspects of enhancing power grid resilience against power disruptions from hurricanes have been presented: (a) predicting the extent of power outages before the arrival of a hurricane to allow utilities to prepare for emergency response and rapidly recover, and develop a probabilistic method to understand the expected performance of power systems to future hurricanes, and (b) investigating the potential of solar power to sustain electricity supply amid power outages during a hurricane.

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, and environmental and demographic parameters. 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. These existing outage models were trained and validated using power outages from multiple regions and hurricanes, including Hurricanes Harvey (2017), Michael (2018) and Isaias (2020), in 1910 US cities.

The project discusses 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. The findings suggest that further development is needed for power outage models for the proper representation of hurricane-induced outages. Additionally, utilities must make risk-informed decisions to prioritize their limited resources (e.g. for grid hardening) in cities expected to experience larger and longer hurricane-induced outages.