Wave height predictions in complex sea flows through soft-computing models: Case study of Persian Gulf

The present study case examined the capability of Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), M5P, and Random Forest (RF) soft-computing approaches for the prediction of wave height over Persian Gulf. Observed data consisted of 8001 wave height data: 5601 data were selected for each model development, while the remaining 2400 data were used for testing the soft-computing models applied here. Six statistical indices were considered here for evaluating the accuracies of the tested models: Coefficient of Determination (R2), Correlation Coefficient (CC), Root Mean Squared Error (RMSE), Bias, Scatter Index (SI), and Mean Absolute Error (MAE). The outcomes of this study revealed that the M5P model outperformed all the other models in terms of predictive performance for wave heights, and Taylor diagram and single-factor Analysis of variance (ANOVA) results confirmed that M5P and ANFIS-trimf models are highly more suitable for retrieving wave height estimations compared to the other soft-computing models. The findings of this study case represent a useful tool for ecohydraulic, sea, and coastal engineers in obtaining accurate wave height predictions when dealing with the design of engineering structures and water resource management in an extremely tangled fluid dynamic and climate context as Persian Gulf area.