Modeling and Forecasting Renewable Energy

Location

CSU 255

Start Date

12-4-2022 1:30 PM

End Date

12-4-2022 2:30 PM

Student's Major

Mathematics and Statistics

Student's College

Science, Engineering and Technology

Mentor's Name

Iresha Premarathna

Mentor's Department

Mathematics and Statistics

Mentor's College

Science, Engineering and Technology

Description

Renewable energy sources are promising alternatives to fossil fuels (non-renewable energy). Renewable energy sources are clean, sustainable, safe, and environmentally friendly with little or no CO2 emissions. This reduces the dependence on fossil fuels, which in turn reduces environmental pollution. However, unpredictability in renewable energy sources such as solar and wind energy makes relying on these alternatives challenging. The data have been collected from www.eia.gov. It is based on monthly data from January 1973 to September 2021, and we obtain forecasts for the next 14 months (October 2021 to Nov 2022). This project aims to develop a model that can accurately forecast the future production and consumption of renewable energy in the United States. We utilize Autoregressive Integrated Moving Average (ARIMA) Model for modeling. Holt-Winters exponential method has been used for forecasting, found within the forecast package in R/RStudio. Finally, model validations have been tested using both visual (residual plots) and analytical (using p-values) methods.

This document is currently not available here.

Share

COinS
 
Apr 12th, 1:30 PM Apr 12th, 2:30 PM

Modeling and Forecasting Renewable Energy

CSU 255

Renewable energy sources are promising alternatives to fossil fuels (non-renewable energy). Renewable energy sources are clean, sustainable, safe, and environmentally friendly with little or no CO2 emissions. This reduces the dependence on fossil fuels, which in turn reduces environmental pollution. However, unpredictability in renewable energy sources such as solar and wind energy makes relying on these alternatives challenging. The data have been collected from www.eia.gov. It is based on monthly data from January 1973 to September 2021, and we obtain forecasts for the next 14 months (October 2021 to Nov 2022). This project aims to develop a model that can accurately forecast the future production and consumption of renewable energy in the United States. We utilize Autoregressive Integrated Moving Average (ARIMA) Model for modeling. Holt-Winters exponential method has been used for forecasting, found within the forecast package in R/RStudio. Finally, model validations have been tested using both visual (residual plots) and analytical (using p-values) methods.

Recommended Citation

Sanjel, Deepshikha. "Modeling and Forecasting Renewable Energy." Undergraduate Research Symposium, Mankato, MN, April 12, 2022.
https://cornerstone.lib.mnsu.edu/urs/2022/oral-session-04/3