Capability of Adaptive Neuro-Fuzzy Inference System to Model Suspended Sediment Transport (A Case Study: Ajichay River)

Research areas:

• Volume 3,Issue 6, May 2015

Year:

2015

Type of Publication:

Article

Keywords:

Fuzzy Inference System, Suspended Sediment, Streamflow, Ajichay Watershed

Authors:

Vahid Nourani; Farhad Alizadeh; Kiyoumars Roshangar

Journal:

IJAIM

Volume:

3

Number:

6

Pages:

275-280

Month:

May

Abstract:

Suspended sediment (QS) transport has a great importance in understanding of river hydraulics and engineering and morphology, which is been a subject for engineers and geologists to study. In this research for the purpose of modeling QS of Ajichayriver, Adaptive Neuro-Fuzzy Inference System (ANFIS) was employed. Selected ANFIS model utilized 2 gauss membership function because of it’s fine capability in mapping. For prediction of one-step-ahead QS for three stations located in Ajichayriver, selected relevant inputs were exported to ANFIS in order to perform sensitivity analysis. Results demonstrated that input data combination of QS and streamflow (QW) of one month and twelve month ago (Comb.4) had the best performance. Results of RMSE and DC approved that ANFIS performance in temporal modeling of QS was appropriate.

Full text: IJAIM_441_Final.pdf