Application of Artificial Neural Networks (ANN) for unit commitment prediction / Robert Engkiau
This report presents an application Artificial Neural Network (ANN) in MATLAB for predicting a unit commitment in power system. Presented here is a design framework parallel training process over the unit commitment data. Dedicated artificial neural networks can handle a large number of inequality c...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | en |
| Published: |
2003
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/79852/1/79852.pdf https://ir.uitm.edu.my/id/eprint/79852/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This report presents an application Artificial Neural Network (ANN) in MATLAB for predicting a unit commitment in power system. Presented here is a design framework parallel training process over the unit commitment data. Dedicated artificial neural networks can handle a large number of inequality constraints included in unit commitment. Results from existing Genetic Algorithm (GA) program were used as the NN training and testing data set. The minimum operating cost from that results as a input and targeted output of neural network (NN). Stage of scheduling, temperature and day were added to the training input for speeding up the convergence process. The developed ANN is capable to predict a unit commitment when an unseen data fed to the network using MATLAB ANN toolbox, Version 6.5. |
|---|