Charles Darwin University

CDU eSpace
Institutional Repository

 
CDU Staff and Student only
 

Application of neural networks to short term load forecasting for the Northern Territory power system

Egan, Sean (2015). Application of neural networks to short term load forecasting for the Northern Territory power system. Bachelor of Engineering (4th Year Project) Thesis, Charles Darwin University.

Document type: Thesis
Citation counts: Google Scholar Search Google Scholar
Attached Files (Some files may be inaccessible until you login with your CDU eSpace credentials)
Name Description MIMEType Size Downloads
Download this reading Thesis_CDU_59878_Egan_S.pdf PDF version generated by student application/pdf 1.50MB 30
Reading the attached file works best in Firefox, Chrome and IE 9 or later.

Author Egan, Sean
Title Application of neural networks to short term load forecasting for the Northern Territory power system
Institution Charles Darwin University
Publication Date 2015-10
Thesis Type Bachelor of Engineering (4th Year Project)
Supervisor Hill, Damien J.
Au-Yeung, Tat
Subjects ENGINEERING
0906 - Electrical and Electronic Engineering
Abstract Short term load forecasting is the prediction of power consumed in a region or entirety of a power system. It is typically used for the purpose of predicting market costs and ensuring operational system security requirements are met. Issues with short term load forecasting occur for smaller power systems due to higher variability from unpredictable events. This is problematic for with traditional load forecasting methods.

Traditional forecasting methods require a full understanding of all factors influencing the load of the power system and are less capable of adapting to changes in the power system. An artificial neural network is able to recognize patterns and learn the extent of the impact for each factor influencing the load continuously, this allows for continuous adaptation. This thesis describes the development of Network X, an artificial neural network to short term load forecast for the Northern Territory power system which is highly variable and thus challenging for traditional forecasting models.

A quantitative evaluation of Network X against other artificial neural network short term load forecasting models and manual forecasting methods indicates Network X performs on an equally high level, with mean absolute percentage error for day ahead predictions of 4.7%. Due to the high load variability of the Northern Territory power system, this result is particularly significant.
Keyword electrical
power system
artificial neural network (ANN)
short term load forecast (STLF)
Northern Territory power system (NTPS)


© copyright

Every reasonable effort has been made to ensure that permission has been obtained for items included in CDU eSpace. If you believe that your rights have been infringed by this repository, please contact digitisation@cdu.edu.au.

 
Versions
Version Filter Type
Access Statistics: 14 Abstract Views, 30 File Downloads  -  Detailed Statistics
Created: Tue, 13 Dec 2016, 12:16:25 CST by Jessie Ng