Artificial neural network approach for assessing harbor tranquility: The case of Trabzon Yacht Harbor, Turkey

Abstract The basic functions of a harbor are to provide safe anchorage for vessels and to facilitate the smooth and unhindered transfer of passengers and cargo between vessels and land. To perform these functions, harbor basins must be tranquil. Traditionally, the tranquility level can be determined by physical and numerical model studies. In this study, physical and artificial neural network (ANN) models of wave height within a harbor were performed, and their results were compared for Trabzon Yacht Harbor, Turkey. Physical model studies were carried out in the Karadeniz Technical University Civil Engineering Department Hydraulics Laboratory wave basin. Models were simulated for 180 cases with various wave and breakwater conditions. Wave heights were measured at 24 points in the harbor basin. Experimental data were divided into 144 training, 24 testing, and 12 validation patterns in the ANN model. Comparison of the results from the physical and ANN models revealed that the maximum average and average relative errors computed for the validation data set were 19.8% and 15.9%, respectively. The ANN model was separately simulated for artificial input values of different wave and breakwater conditions.