Paper Details

[ Downloads: 0 | Abstract Viewed: 85 | Pages: 9 ]

Abstract of the Article

Note: English CIVILICA is in its Trial Period so Full Texts can not be provided! Persian users can download it here

Comparison of Sediment Rating Curve with MLP and GRNN models in sediment estimation of Zayanderood River

Authors:
[ Najafi ] - MSc Student of Hydraulic Structures, Department of Water Engineering, Mazandaran University
[ Heidarpour ] - Assistant Professor, Department of Water Engineering, Isfahan University of Technology
[ Abas Asadi ] - Espahan Regional Water Organization

Abstract:

This study investigates transported sediment in the main branch of the Zayanderood River. For doing it, sediment transportation rate was modeled using statistical regression and two methods of artificial neural networks, in three situations. In the 1 th situation the entire measured data set was modeled without any dividing into the sub-set. That's clear that this model has a great error. In the 2 nd situation the data set was divided into two sub-set, floody and unfloody sub-set, and also in the 3 rd situation, into two sub-set, low and high sediment concentration. Then each sub-set was simulated with each method, separately. The results illustrate that, MLP have more agreement with the measured ones, compared with the GRNN and statistical regression methods, in terms of the selected performance criteria. On the other hand, dividing the data set based on the sediment concentration has more accurate responses.

Keywords: