Author(s): Andreas Papadimitriou; Vasiliki Tsoukala
Linked Author(s): Vasiliki (Vicky) Tsoukala, Vasiliki Tsoukala
Keywords: No Keywords
Abstract: The morphological coastal bed evolution is of high interest to engineers, scientists and the public due to the vast number of activities concentrated near the shoreline. Traditionally, process-based models have been employed to predict bed level changes in time scales of 1-5 years, however they are associated with prohibitive computational restrictions. To reduce the computational burden, wave Input Reduction methods, aiming to reduce the forcing input and accelerate morphological simulations have been developed. The present paper aims at evaluating the K-Means algorithm as an alternative approach to select wave representatives for morphological simulations. Several alternative configurations were tested in order to enhance and coerce the algorithm to “smartly” select the representative waves. The examined configurations were implemented in the coastal area of Rethymno, Greece for an annual dataset of sea-state wave characteristics and the results were deemed very satisfactory compared to those obtained by the full wave climate, rendering the use of K-Means algorithm a valuable tool for coastal engineers and scientists.
Year: 2022