Graduate Project
 

Power Electric System Load and Generation Forecasting via ARMA, ARIMA, and K-Medoids Methods for Grid Expansion Purposes: A Comparison

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_projects/fn107676n

Descriptions

Attribute NameValues
Creator
Abstract
  • The electric power grids of countries across the globe rely on load and generation forecasting to know when, where, and how much resources need to be dispatched to sustain proper grid operation. Because of this, forecasting needs to be highly accurate to avoid unnecessary resource dispatch which can be costly. Forecasting has become more complex over the years as grid expansion efforts have started to see new generation resources enter the energy mix, new transmission networks being built, smart grid integration, and energy storage technologies. These new systems have complicated once-standard forecasting techniques to a point that alternate techniques should be evaluated. This project describes, analyzes, and compares three statistical modeling methods (ARMA, ARIMA, and K-Medoids) that are commonly used in forecasting to determine their accuracy and ability to consistently predict future load and generation demand. These methods will be used to determine day-ahead forecasting and week-ahead forecasting using California Independent System Operator (CAISO) datasets from April 2023 as a baseline. The results are analyzed to deduce the usability and accuracy of the three methods and compared to determine whether they are adequate for forecasting load and generation in an energy future upswept by grid expansion.
License
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Rights Statement
Publisher
Peer Reviewed
Language

Relationships

Parents:

This work has no parents.

Items

Thumbnail Title Date Uploaded Visibility Actions
file details McLaughlinZacheryD2023.pdf 2023-06-20 Public Download