Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Dec 15;24(24):8006.
doi: 10.3390/s24248006.

A Dual Filter Based on Radial Basis Function Neural Networks and Kalman Filters with Application to Numerical Wave Prediction Models

Athanasios Donas  1 , Ioannis Kordatos  1 , Alex Alexandridis  1 , George Galanis  2 , Ioannis Th Famelis  1
Affiliations

A Dual Filter Based on Radial Basis Function Neural Networks and Kalman Filters with Application to Numerical Wave Prediction Models

Athanasios Donas et al. Sensors (Basel). .

Abstract

The aim of this study is to introduce and evaluate a dual filter that combines Radial Basis Function neural networks and Kalman filters to enhance the accuracy of numerical wave prediction models. Unlike the existing methods, which focus solely on systematic errors, the proposed framework concurrently targets both systematic and non-systematic parts of forecast errors, significantly reducing the bias and variability in significant wave height predictions. The produced filter is self-adaptive, identifying optimal Radial Basis Function network configurations through an automated process involving various network parameters tuning. The produced computational system is assessed using a time-window procedure applied across divergent time periods and regions in the Aegean Sea and the Pacific Ocean. The results reveal a consistent performance, outperforming classic Kalman filters with an average reduction of 53% in bias and 28% in RMSE, underlining the dual filter's potential as a robust post-processing tool for environmental simulations.

Keywords: Kalman filters; WAM; post-process algorithms; radial basis function neural networks; significant wave height.

PubMed Disclaimer

Conflict of interest statement

The authors report no conflicts of interest.

Figures

Figure A1
Figure A1
Number of clusters. Total results for 30 time windows.
Figure A2
Figure A2
Regularization parameter. Total results for 30 time windows.
Figure A3
Figure A3
Activation functions. Total results for the 30 time windows.
Figure 1
Figure 1
Method’s Diagram.
Figure 2
Figure 2
Locations of Aegean Stations (“https://poseidon.hcmr.gr/ (accessed on 11 December 2024)”).
Figure 3
Figure 3
Location of Station 46002 (“https://www.ndbc.noaa.gov/ (accessed on 11 December 2024)”). Red squares indicate Stations with no data during the last 8 hours, while yellow squares indicate Stations with recent data.
Figure 4
Figure 4
A standard Radial Basis Function neural network.
Figure 5
Figure 5
Time Series Diagram. Mykonos 2007.
Figure 6
Figure 6
Time Series Diagram. Mykonos 2008.
Figure 7
Figure 7
Time Series Diagram. Heraklion 2007.
Figure 8
Figure 8
Time Series Diagram. Heraklion 2009.
Figure 9
Figure 9
Time Series Diagram. 46002 2012.
Figure 10
Figure 10
Time Series Diagram. 46002 2013.
See this image and copyright information in PMC

References

    1. Takahashi K., Miyoshi Y. Introduction to Wave-Particle Interactions and Their Impact on Energetic Particles in Geospace. In: Balasis G., Daglis I.A., Mann I.R., editors. Waves, Particles, and Storms in Geospace. Oxford University Press; Oxford, UK: 2016. pp. 35–50. - DOI
    1. Galanis G., Emmanouil G., Chu P.C., Kallos G. A New Methodology for the Extension of the Impact of Data Assimilation on Ocean Wave Prediction. Ocean. Dyn. 2009;59:523–535. doi: 10.1007/s10236-009-0191-8. - DOI
    1. Famelis I., Galanis G., Ehrhardt M., Triantafyllou D. Classical and Quasi-Newton Methods for a Meteorological Parameters Prediction Boundary Value Problem. Appl. Math. Inf. Sci. 2014;8:2683–2693. doi: 10.12785/amis/080604. - DOI
    1. Famelis I.T., Tsitouras C. Quadratic shooting solution for an environmental parameter prediction problem. FJAM. 2015;91:81–98. doi: 10.17654/FJAMMay2015_081_098. - DOI
    1. Dong R., Leng H., Zhao C., Song J., Zhao J., Cao X. A Hybrid Data Assimilation System Based on Machine Learning. Front. Earth Sci. 2023;10:1012165. doi: 10.3389/feart.2022.1012165. - DOI

LinkOut - more resources