| Autor(es): Mostrar menos - |
Moreno, Felipe M.
;
Coelho, Jefferson F.
;
Mathias, Marlon S.
;
de Barros, Marcel R.
;
Netto, Caio F. D.
;
de Freitas, Lucas P.
;
Dottori, Marcelo
;
Cozman, Fabio G.
;
Costa, Anna H. R.
;
Gomi, Edson S.
;
Tannuri, Eduardo A.
Número total de Autores: 11
|
| Tipo de documento: | Artigo Científico |
| Fonte: | PROCEEDINGS OF ASME 2023 42ND INTERNATIONAL CONFERENCE ON OCEAN, OFFSHORE & ARCTIC ENGINEERING, OMAE2023, VOL 5; v. N/A, p. 8-pg., 2023-01-01. |
| Resumo | |
This work employs Echo State Networks (ESN) to improve one day ahead forecasts of water currents in a port access channel entrance, with the aim to provide a more accurate method that can be used to alert the port authority about unsafe conditions for navigation in the near future. For this task Leaky integrator Echo State Networks (LiESNs) are applied to predict one day ahead currents in the Santos channel, Brazil; The method has been trained with data obtained from an Acoustic Doppler Current Profiler installed in the channel, with data available between the years of 2019-2020. Results have been compared with the Santos Operational Forecasting System (SOFS), a numerical model already in use for the same region. Experiments show that the LiESN outperforms the numerical model by the metrics of Index of Agreement (0.922 versus 0.812) and RMSE (0.152 ms(-1) versus 0.220 ms(-1)). (AU) | |
| Processo FAPESP: | 19/07665-4 - Centro de Inteligência Artificial |
| Beneficiário: | Fabio Gagliardi Cozman |
| Modalidade de apoio: | Auxílio à Pesquisa - Programa eScience e Data Science - Centros de Pesquisa em Engenharia |
| Processo FAPESP: | 20/16746-5 - Physics-informed machine learning aplicado para previsões de condições metoceânicas |
| Beneficiário: | Felipe Marino Moreno |
| Modalidade de apoio: | Bolsas no Brasil - Doutorado |