Time-variation of scour depth around abutments is more complicated due to the three-dimensional vortex flow. In this study, variation of the scour depth with time around abutments was investigated experimentally under clear-water conditions. The Nonlinear Regression (NLR), neuro-fuzzy inference system (ANFIS) and artificial neural networks (ANN) were employed to predict the time variation of scour depth at intermediate and short abutment. Statistical analysis showed that, although the presented regression method (R2=0.957, RMSE=0.049 and MAE=0.035) provided a better prediction compared to previous equations but the ANFIS (R2=0.961, RMSE=0.041and MAE=0.025) produces a more accurate result. Statistical measurements showed that ANFIS as well as ANNs can be considered as a robust technique to predict the temporal scour depth at the short and intermediate abutment.
)
