Author(s): Andreas Papadimitriou; Michalis Chondros; Anastasios Metallinos; Vasiliki Tsoukala
Linked Author(s): Vasiliki (Vicky) Tsoukala, Vasiliki Tsoukala
Keywords: No Keywords
Abstract: Process-based models have been employed extensively in the last decades for the prediction of coastal bed evolution in the medium term (1-5 years), under the combined action of waves and currents, due to their ability to resolve the dominant coastal processes. Despite their widespread application, they are associated with a high demand in computational resources, rendering the annual prediction of the coastal bed evolution a tedious task. To combat this, various accelaration techniques such as wave input reduction or elimination of lowly-energetic sea-states have been implemented in many practical applications. The purpose of this research is to further expand on the concept of accelerating morpholgical simulations by developing a methodology centered around employing an Artificial Neural Network (ANN), tasked with eliminating wave records unable to initiate sediment motion and hence further reduce computational times. The ANN has been trained on a 2DH idealized plane sloping beach with a robust dataset produced by simulations of three sophisticated numerical models. The proposed methodology has been implemented for a real field case in the coastal area of Rethymno, Greece and the obtained results were deemed very satisfactory by maintaining a balance between accuracy of results and computational efficiency, having strong implications for practical coastal engineering purposes.
Year: 2022