Assessing Hydro Power System Relevant Variables: a Comparison Between a Neural Network and Different Machine Learning approaches *
| Citations: | 1 - 1 self |
BibTeX
@MISC{Lounis_assessinghydro,
author = {Hakim Lounis and Mounir Boukadoum and Vincent Siveton},
title = {Assessing Hydro Power System Relevant Variables: a Comparison Between a Neural Network and Different Machine Learning approaches *},
year = {}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
This work investigates the performance of five pattern classification algorithms in predicting natural contributions flow in a hydropower generation network, using a database of historical hydrological data. We compared the use of a multilayer perceptron, trained with the Resilient Backpropagation (RPROP) algorithm, vs. four well-known machine-learning techniques, as part of a software framework adapted to hydropower system assessment. The framework uses variable prediction algorithms to support rule-based decision processes. Our results are that the use of a neural network far outweighs
