Modeling the Effect of Breakwater Layer Porosity on Wave Overtopping Relationships Using Machine Learning Algorithm

Accurate and reliable estimation of wave overtopping in coastal structures is crucial for their design and safety assessment.In this study, a new relationship for estimating wave overtopping in rubble mound breakwaters with Xbloc armor units is developed using numerical modeling and machine learning.The proposed method is based on supervised learning and nonlinear regression, following the general form of the Hibbertsgaard et al. model.To validate the model, the results of the proposed relationship were compared with experimental data as well as the Owen and Van der Meer formulas.The coefficient of determination (R 2 ), root mean square error (RMSE) and the Index of Agreement (d) indicated that the proposed relationship showed a better correlation with experimental data.Numerical modeling was performed for eight different conditions, and in all cases, the proposed relationship demonstrated a better agreement with numerical outputs compared to previous empirical formulas.These results suggest that numerical modeling and machine learning can provide an accurate and cost-effective alternative to expensive and time-consuming physical experiments for deriving wave overtopping relationships.This study also specifically examines the effect of breakwater porosity, which has received less attention in previous research, and incorporates it into the proposed relationship.The findings indicate that the proposed method can serve as an efficient tool for developing more accurate wave overtopping relationships without the need for extensive physical modeling.