The ability of the artificial neural network to estimate the flow of the Minab River in Hormozgan Province

Accurate forecasting of effective parameters in water resources projects is one of the most important research topics for water engineers. One of these parameters is river flow. In this paper, the neural network is used in two ways: multi-layered perceptron network (MLP) network, with error post-propagation algorithm and radial base function (RBF) and using Matlab and SPSS 19 software to estimate Dubai in Minab watershed using data. Simultaneous daily rainfall and daily water supply have been used for Brentin, Glasgard and Faryab stations for a period of 25 years. In this study, in order to investigate and determine the highest rainfall effect of each of the stations on the Dubai River, the impact of 5 consecutive days of rainfall on Dubai on a daily basis (p0, p-1, p-2, p 3.3, p-4, p-5) For each station and jointly, the first and second to fifth day rains were performed for stations with two to two subscribers. The results showed that the multi-layered perceptron (MLP) method had the lowest error (RMSE) and the highest correlation (R^ 2) between daily precipitation at all three stations measured with Dubai. Also, the first day's rainfall in the couple's stations and the first and fifth day's rain in the single stations have the greatest impact on Dubai's output