Short-term load forecasting using artificial neural network / Suhana Shaari
This thesis presents a neural network based approach for short-term load forecasting that uses the most correlated weather data for training and testing the neural network of weather data determines the input parameters of the neural networks. Inputs to the ANN are past loads and the output of the A...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English |
Published: |
2012
|
Online Access: | https://ir.uitm.edu.my/id/eprint/84792/1/84792.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my-uitm-ir.84792 |
---|---|
record_format |
uketd_dc |
spelling |
my-uitm-ir.847922024-01-15T09:00:20Z Short-term load forecasting using artificial neural network / Suhana Shaari 2012 Shaari, Suhana This thesis presents a neural network based approach for short-term load forecasting that uses the most correlated weather data for training and testing the neural network of weather data determines the input parameters of the neural networks. Inputs to the ANN are past loads and the output of the ANN is the load forecast for a given day. The network with one or two hidden layers is tested with various combinations of neurons, and the results are compared in term of forecasting error. Historical load data and temperature observations for the year 2006 - 2010 obtained from the Australian Energy Market Operator (AEMO) & Bereau of Meteordology (BOM) for Sydney/NSW. The inputs used were the hourly load demand for the full day (24 hours), the weather, humidity and holiday for the state.-The network trained over 4 year's data. A mean average percent error (MAPE) of 1.99% was achieved when the trained network was tested on one year data. 2012 Thesis https://ir.uitm.edu.my/id/eprint/84792/ https://ir.uitm.edu.my/id/eprint/84792/1/84792.pdf text en public degree Universiti Teknologi MARA (UiTM) Faculty of Electrical Engineering Zakaria, Zuhaina |
institution |
Universiti Teknologi MARA |
collection |
UiTM Institutional Repository |
language |
English |
advisor |
Zakaria, Zuhaina |
description |
This thesis presents a neural network based approach for short-term load forecasting that uses the most correlated weather data for training and testing the neural network of weather data determines the input parameters of the neural networks. Inputs to the ANN are past loads and the output of the ANN is the load forecast for a given day. The network with one or two hidden layers is tested with various combinations of neurons, and the results are compared in term of forecasting error. Historical load data and temperature observations for the year 2006 - 2010 obtained from the Australian Energy Market Operator (AEMO) & Bereau of Meteordology (BOM) for Sydney/NSW. The inputs used were the hourly load demand for the full day (24 hours), the weather, humidity and holiday for the state.-The network trained over 4 year's data. A mean average percent error (MAPE) of 1.99% was achieved when the trained network was tested on one year data. |
format |
Thesis |
qualification_level |
Bachelor degree |
author |
Shaari, Suhana |
spellingShingle |
Shaari, Suhana Short-term load forecasting using artificial neural network / Suhana Shaari |
author_facet |
Shaari, Suhana |
author_sort |
Shaari, Suhana |
title |
Short-term load forecasting using artificial neural network / Suhana Shaari |
title_short |
Short-term load forecasting using artificial neural network / Suhana Shaari |
title_full |
Short-term load forecasting using artificial neural network / Suhana Shaari |
title_fullStr |
Short-term load forecasting using artificial neural network / Suhana Shaari |
title_full_unstemmed |
Short-term load forecasting using artificial neural network / Suhana Shaari |
title_sort |
short-term load forecasting using artificial neural network / suhana shaari |
granting_institution |
Universiti Teknologi MARA (UiTM) |
granting_department |
Faculty of Electrical Engineering |
publishDate |
2012 |
url |
https://ir.uitm.edu.my/id/eprint/84792/1/84792.pdf |
_version_ |
1794192051111198720 |