Recently, an analysis method using machine learning for solving problems in coastal and ocean engineering has been highlighted. Machine learning models are effective modeling tools for predicting specific parameters by learning complex relationships based on a specified dataset. In coastal and ocean engineering, various studies have been conducted to predict dependent variables such as wave parameters, tides, storm surges, design parameters, and shoreline fluctuations. Herein, we introduce and describe the application trend of machine learning models in coastal and ocean engineering. Based on the results of various studies, machine learning models are an effective alternative to approaches involving data requirements, time-consuming fluid dynamics, and numerical models. In addition, machine learning can be successfully applied for solving various problems in coastal and ocean engineering. However, to achieve accurate predictions, model development should be conducted in addition to data preprocessing and cost calculation. Furthermore, applicability to various systems and quantifiable evaluations of uncertainty should be considered.

Coastal erosion, a worldwide social issue, has garnered substantial attention. Numerous methods have been implemented to control coastal erosion problems; however, the presence of rigid structures limits erosion mitigation, thereby causing various challenges. For instance, in the case of submerged breakwaters, local scour in front of the structure and scour caused by the flow occurring in open inlets affect the subsidence and stability of the structure and can also cause structural failure. To solve these problems, this paper proposes a hybrid method of using a submerged breakwater with an artificial coral reef installation; further, this study evaluates the attenuation of waves and mitigation of sediment transportation through large-scale 3D hydraulic experiments. We found that the hybrid method with an artificial coral reef installed in the open inlet shows excellent wave control and plays a clearly beneficial role in the advancement of the shoreline. The artificial coral reef method reduced the return flow generated by the drag force at the breakwater shoulder and open inlet. In addition, scour at the breakwater shoulder was inhibited by collecting the sand escaping offshore. Simultaneously, scour at the open inlet was also mitigated. The application of the hybrid method compensated for the problems caused by local scour and erosion in the submerged breakwater, thereby leading to the improvement of its function. Therefore, the hybrid method proposed in this paper was determined to be applicable not only for submerged breakwaters, but also for various structures for controlling coastal erosion.

The adoption of low-crested and submerged structures (LCS) reduces the wave behind a structure, depending on the changes in the freeboard, and induces stable waves in the offshore. We aimed to estimate the wave transmission coefficient behind LCS structures to determine the feasible characteristics of wave mitigation. In addition, various empirical formulas based on regression analysis were proposed to quantitatively predict wave attenuation characteristics for field applications. However, inherent variability of wave attenuation causes the limitation of linear statistical approaches, such as linear regression analysis. Herein, to develop an optimization model for the hydrodynamic behavior of the LCS, we performed a comprehensive analysis of 10 types of machine learning models, which were compared and reviewed on the prediction accuracy with existing empirical formulas. We found that, among the 10 models, the gradient boosting model showed the highest prediction accuracy with MSE of 1.0 × 10⁻³, an index of agreement of 0.996, a scatter index of 0.065, and a correlation coefficient of 0.983, which indicates a performance improvement over the existing empirical formulas. In addition, based on a variable importance analysis using explainable artificial intelligence, we determined the significant importance of the input variable for the relative freeboard (R_C/H₀) and the relative freeboard to water depth ratio (R_C/h), which confirms that the relative freeboard was the most dominant factor for influencing wave attenuation in the hydraulic behavior around the LCS. Thus, we concluded that the performance prediction method using a machine learning model can be applied to various predictive studies in the field of coastal engineering, deviating from existing empirical-based research.

Wave absorbers installed in front of revetments are effective in reducing wave overtopping and inundation caused by periodic waves. The wave absorbers’ mechanism of reducing wave overtopping and inundation caused by long-period waves such as tsunamis and storm surges is not clearly understood. This study conducted a physical modeling test and numerical analysis based on a large eddy simulation model using in-house code to examine the characteristics of wave overtopping and inundation according to the revetment type for solitary waves. In a vertical revetment (VR), the dominant vertical velocity of the solitary wave cannot bend at a right angle during overtopping, causing flow separation to occur at the crest, which leads to increased drag and vorticity. In a wave absorbing revetment (WAR), the flow cross-sectional area decreases along the slope of the wave absorber, causing the flow velocity of the solitary wave to increase and the horizontal velocity to be dominant during the overtopping and inundation process. In contrast with the general wave overtopping characteristics of periodic waves, the maximum overtopping water surface elevation in front of the vertical wall is higher in a VR than in a WAR. However, the order of maximum inundation heights reverses as the wave propagates inland.