Machine Learning Predictions of Electricity Transfers Between Balancing Authorities in the Carolinas

Author

Victoria Groleau

Date of Award

3-2024

Document Type

Thesis

Degree Name

Master of Science in Engineering Management

Department

Department of Systems Engineering and Management

First Advisor

Daniel J. Weeks, PhD

Abstract

Climate change through reduced streamflow, increased temperatures, and other factors impacts the efficiency of energy generation systems. The United States electric grid is comprised of a large network of balancing authorities engaged in trading electricity to maintain balance between supply and demand. The generation of electricity, a pivotal component of this balance, is impacted by climate change and weather variability as well as the growing demand for energy. Several hydro climatological factors such as streamflow, air temperature, and wind speed significantly influence the efficiency of power plant electricity generation. Due to the exchange of electricity between balancing authorities, impacts to electricity generation in one region could have negative effects on other regions, creating additional vulnerabilities in the electric grid. This study analyzes how hourly electricity trade of balancing authorities in the Carolinas (CAR) region can be predicted using hourly wind speed, air temperature, and streamflow data. The study uses random forest machine learning models to determine electricity transfer patterns based on the balancing authority’s electricity generation and exports. The results imply that streamflow is a significant predictor in electricity exchange of this region, highlighting drought as a notable vulnerability in the regional electric grid. However, there are limited predictive capabilities for seasonal variables such as time-of-day and day-of-week. This research adds to the existing body of knowledge within the energy-water nexus to highlight the coupling of these resources in the electricity trade.

AFIT Designator

AFIT-ENV-MS-24-M-122

Comments

A 12-month embargo was observed for posting this work on AFIT Scholar.

Distribution Statement A, Approved for Public Release. PA case number on file.

Recommended Citation

Groleau, Victoria, "Machine Learning Predictions of Electricity Transfers Between Balancing Authorities in the Carolinas" (2024). Theses and Dissertations. 7747.
https://scholar.afit.edu/etd/7747