Portal de Periódicos da CAPES

Artigo Produção Nacional

BIBTEX RIS

A new approach to river flow forecasting: LSTM and GRU multivariate models

2019; Institute of Electrical and Electronics Engineers; Volume: 17; Issue: 12 Linguagem: Inglês

10.1109/tla.2019.9011542

1548-0992

Gabriel Adriano de Melo, Dylan Nakandakari Sugimoto, Paulo Marcelo Tasinaffo, Afonso Henriques Moreira Santos, Adilson Marques da Cunha, Luiz Alberto Vieira Dias,

Stock Market Forecasting Methods

Hydroelectric power stations are responsible for renewable energy generation, especially in countries with many rivers such as Brazil. It is very important to have good estimates of the hydrological flow in order to determine whether thermoelectric power plants should begin operation, an event that would increase the costs of electricity and also have aterrible environmental impact. The monthly flow of a river was estimated using two recurrent neural networks techniques: Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU). The results were compared with other articles that had the same structure and used the same data: the Rio Grande river in the Furnas and Camargos dam.