The Evaluation of the Cprecip’s Parameter Ability on Appropriation of Snow’s Effect in River Daily Discharge Prediction by Neural  Network and Fuzzy Neural Network

Seyyed Abbasi, S.M.; Goran Orimi, M.; Farid Hosseini, A.; Sharifi, M.B.

The Evaluation of the Cprecip’s Parameter Ability on Appropriation of Snow’s Effect in River Daily Discharge Prediction by Neural Network and Fuzzy Neural Network

Document Type : Research Paper

Authors

S.M. Seyyed Abbasi ¹

M. Goran Orimi ²

A. Farid Hosseini ³

M.B. Sharifi ⁴

¹ Graduate Student of Electrical Control, Ferdowsi University of Mashhad

² MS in Civil Hydraulic Structures, Ferdowsi University of Mashhad

³ Assistant Professor, Department of Water Engineering, Ferdowsi University of Mashhad

⁴ Assistant Professor, Department of Civil Engineering, Ferdowsi University of Mashhad

Abstract

One of the most effective parameters in discharge prediction especially in snowy basins, is the snow parameter. Snow water equivalent (SWE) is the most common parameter used in modeling river flow to take the effect of the snow on the model into account. In this research, according to unavailability of the SWE parameter in most of the basins, we tried to offer Cprecip and MAZ-Cprecip instead of SWE. Cprecip is cumulative precipitation from November 1^st to April 1^st indicating the snowpack amount melting in the spring. MAZ-Cprecip is received by giving some changes on the Cprecip (to be suitable with Mazandaran basins). The results showed that the Cprecip parameter can be replaced with SWE and that MAZ-Cprecip parameter is more efficient than Cprecip parameter in Mazandaran basins, due to its most conformity with the basins in this region.

Keywords

Discharge prediction

Snow

Artificial neural network

Fuzzy neural network