An application of artificial neural network on short term load forecasting using back propagation algorithm / Elia Erwani Hassan
This study is covered a new approach to load forecasting using Artificial Neural Network (ANNs). Improving accuracy of load forecast by Back Propagation Algorithm is the main objective for this project. This accuracy is dependent on several ANN parameters such as learning rate and momentum rate. The...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
1998
|
| Online Access: | https://ir.uitm.edu.my/id/eprint/101732/1/101732.pdf https://ir.uitm.edu.my/id/eprint/101732/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This study is covered a new approach to load forecasting using Artificial Neural Network (ANNs). Improving accuracy of load forecast by Back Propagation Algorithm is the main objective for this project. This accuracy is dependent on several ANN parameters such as learning rate and momentum rate. The Back Propagation Algorithm, which consists of the multi-layered perception model, makes possible to train the ANN training pattems. As an input, we look at the past 24 hours load data with the type of days as weekdays, Sunday and public holidays. The next 24 hours load patters are considered as outputs. By using Back Propagation Algorithm with 25 hidden nodes, 0.7 learning rate and 0.7 momentum rate have been found to give faster result than other conventional techniques. |
|---|