Daily sea level forecast at tide gauge Burgas, Bulgaria using artificial neural networks
Citations
43 citations
30 citations
28 citations
27 citations
20 citations
References
7,720 citations
1,522 citations
"Daily sea level forecast at tide ga..." refers background or methods in this paper
...Following routine procedures for the selection of the best ANN suited to the daily sea level data, different activation function options and network architectures are compared (Cichocki & Unbehauen, 1993; Demuth et al., 2008)....
[...]
...The General Regression Neural Network (GRNN)model is a kind of the radial basis network that is often used for any regression problem (Cichocki and Unbehauen, 1993)....
[...]
...Any nonlinear mathematical function of input–output relations among the experimental data can be approximated by artificial neural networks (Cichocki and Unbehauen, 1993)....
[...]
...Conclusions are reported in the last section....
[...]
...To forecast the daily sea levels the following models are developed: (a) ordinary time-dependent function with harmonic terms; (b) multilayer Feed-Forward (FF); (c) Cascade-Feed-Forward (CFF); (d) Feed-Forward Time-Delay (FFTD); (e) Radial Basis Function (RBF); (f) Generalized Regression (GR)…...
[...]
987 citations
"Daily sea level forecast at tide ga..." refers background or methods in this paper
...Traditionally, a tidal harmonic analysis is used for the sea-level prediction (Pugh, 1987), but byneglecting thehydro-meteorological effects theprediction error can reach 30% (Ghorbani et al....
[...]
...The observed sea level can be represented as a sum of a relative mean level at a given location, an astronomical tidal component and a non-tidal residual sum (Pugh, 1987)....
[...]
...Traditionally, a tidal harmonic analysis is used for the sea-level prediction (Pugh, 1987), but byneglecting thehydro-meteorological effects theprediction error can reach 30% (Ghorbani et al., 2010)....
[...]