Abstract

This study explores the potential of hybrid solar and wind renewable energy systems in the United States, focusing on two specific locations in the Southwestern region of the country: Vaughn, New Mexico and Fort Stockton, Texas. Using historical wind speed and solar irradiance (GHI) data spanning 2018-2020, I used Python to apply machine learning models and estimate energy output at each site. After testing multiple methods, Random Forest regression emerged as the most reliable approach for capturing and handling the complex seasonal patterns of weather data such as this. The results indicate that both locations demonstrate strong potential for hybrid renewable energy, but Vaughn specifically displays complementary seasonal performance between solar and wind energy output. This research highlights the value of data-driven site evaluation and supports the notion that integrating predictive models can inform more efficient renewable energy investment decisions.

Advisor

Kelvey, Robert

Department

Statistical and Data Sciences

Disciplines

Applied Statistics | Longitudinal Data Analysis and Time Series | Statistical Models

Keywords

Renewable Energy, Hybrid, Solar Energy, Wind Energy, Random Forest, Modeling

Publication Date

2025

Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis

Share

COinS
 

© Copyright 2025 Andy Heintz