Adaptive Neuro-Fuzzy Inference System for Prediction of Bed- Load Sediment Transport in the Swash Zone

Fluid and sediment interactions occurring in the swash zone determine the erosion or accretion of the beach and act as boundary conditions for morphodynamic models. Therefore, it is necessary to predict the sediment transport in this area. In this paper, the abilities of Fuzzy Inference System (FIS) and Adaptive-Network- Based Fuzzy Inference System (ANFIS) methods are used to predict and modeling bed-load sediment transport in the swash zone. The ANFIS and FIS are established using the free stream velocity time series and antecedent sediment data. Statistic measures were used to evaluate the performance of the models. The crossshore sediment transport rate and swash velocity time series for the swash experiments of Masselink and Hughes (1998) were used as case studies. Based on comparison of the results, it is found that the ANFISbased predictions are slightly superior to the FIS-based predictions. Also, results indicate that using neurofuzzy approach in sediment transport modeling have a sufficient prediction accuracy.

کلیدواژه ها:

ANFIS ، Sediment transport ، Shields parameter ، Swash zone