Results 1 - 1 of 1
Results 1 - 1 of 1. Search took: 0.016 seconds
[en] Highlights: • It solves the problem of maritime spatio-temporal forecasting for the first time. • A new method EEMD-SOM-BP is proposed for maritime forecasting of solar irradiation. • An asymmetric four-parallel structure of SOM is proposed to mine data features. • Three experiments are performed to determine the optimal settings of EEMD-SOM-BP. - Abstract: Owing to a shortage of fossil fuels and environmental pollution, renewable energy is gradually replacing fossil fuels in the power systems of hybrid ships. To exploit fully solar energy by the successful day-ahead scheduling of ships, this work proposes a new day-ahead spatio-temporal forecasting method. Ensemble empirical mode decomposition (EEMD) is used to extract data features and decompose original historical data into several frequency bands. After the original data are processed, data from the four land weather stations that are closest to the ship and self-organizing map-back propagation (SOM-BP) hybrid neural networks are used to forecast the solar radiation received by the ship in the next 24 h. Multiple comparative experiments are implemented. The results show that the EEMD-SOM-BP spatio-temporal forecasting method can accurately forecast the solar radiation on a ship that is sailing along a navigation route.