Instability of offshore structures is primarily affected due to scouring phenomenon around their foundations. This significantly contributes to the failure of offshore wind turbines that serve as critical energy infrastructure units. In the present study, different machine-learning algorithms, viz. Adaptive Neuro Fuzzy Interface System (ANFIS), Artificial Neural Network (ANN), and ANN along with an optimization technique Particle Swarm Optimization (PSO) [1], have been implemented to predict the scour depth around the tripod foundations. In exploring the prediction models, various parameters influencing the scour depth in the marine environment have been considered, such as current velocity (U_c), wave height (H_w), wave period (T_w), Froude number (F_r) and Keulegan-Carpenter number (KC). For training, testing, and validating the ML model’s performance, 99 data points were collected from previously reported experimental studies [2]. The effectiveness of all three machine-learning schemes, ANFIS, ANN, and ANN-PSO has been evaluated using the statistical parameters namely, coefficient of determination (R^2), Root Mean Square Error (RMSE), Mean Absolute Error (MSE) and checked against those previously reported values in literature. Among all the machine learning models the ANN-PSO results in good agreement with the reported outcomes and has better efficiency (R^2=0.98) for predicting the scour depth followed by ANN (R^2=0.95) and ANFIS (R^2=0.85) machine learning algorithms.