Detalhes bibliográficos
| Ano de defesa: |
2013 |
| Autor(a) principal: |
Freitas, Patrícia Fernanda da Silva [UNESP] |
| Orientador(a): |
Não Informado pela instituição |
| Banca de defesa: |
Não Informado pela instituição |
| Tipo de documento: |
Dissertação
|
| Tipo de acesso: |
Acesso aberto |
| Idioma: |
por |
| Instituição de defesa: |
Universidade Estadual Paulista (Unesp)
|
| Programa de Pós-Graduação: |
Não Informado pela instituição
|
| Departamento: |
Não Informado pela instituição
|
| País: |
Não Informado pela instituição
|
| Palavras-chave em Português: |
|
| Link de acesso: |
http://hdl.handle.net/11449/110524
|
Resumo: |
After the electricity industry restructuring, electric energy has become a commodity that can be bought or sold in electricity markets. Within this context, new tools to predict loads and electricity prices are yet to be devised. Neural networks have been typically used for price forecasting, among other statistical-based techniques. This work presents a method for day-ahead price forecasting using a generalized regression neural network where nodal prices are determined by solving a DC optimal power flow. In order to train the neural network several load-price scenarios are generated by randomly varying the loads in each bus of the electric energy system. The results are validated by analyzing the mean absolute percentage error, a common measure adopted in the technical literature as well as the base case and the maximum error which are within those found in the literature. The proposed methodology has been successfully applied to the IEEE 24-RTS system. |
