Amirkabir University of Technology Amirkabir Journal of Civil Engineering 2588-297X 46 1 2014 09 23 Synthetic Streamflow Generation using Artificial Neural Network Models Synthetic Streamflow Generation using Artificial Neural Network Models 1 10 326 10.22060/ceej.2014.326 FA Mostafa Yeganehfard M.Sc. Department of Civil & Environmental Engineering, Iran University of Science and Technology, Tehran, Iran Bagher Zahabioun Associate Professor, Department of Civil & Environmental Engineering, Iran University of Science & Technology, Tehran, Iran Journal Article 2010 01 11 In this study, capability of Artificial Neural Network (ANN) models for synthetic streamflow generation is evaluated. The used generating model compouned from ANN model and a random component with normal distribution. In model developing, the multilayer feedforward neural networks and back propagation learning algorithm has been used. Then long term synthetic streamflow series up tp 300 years of daily streamflow, using only observed daily streamflow in Khersan River has been generated. For model assessment, The comparison carried out in respect of different statistics of the historical data and synthetically generated data such as Basic Statistics like mean, standard deviation and skewness and series Persistence Statistics like autoregressives that finaly has shown model’s ability for Synthetic daily streamflow Generation. In this study, capability of Artificial Neural Network (ANN) models for synthetic streamflow generation is evaluated. The used generating model compouned from ANN model and a random component with normal distribution. In model developing, the multilayer feedforward neural networks and back propagation learning algorithm has been used. Then long term synthetic streamflow series up tp 300 years of daily streamflow, using only observed daily streamflow in Khersan River has been generated. For model assessment, The comparison carried out in respect of different statistics of the historical data and synthetically generated data such as Basic Statistics like mean, standard deviation and skewness and series Persistence Statistics like autoregressives that finaly has shown model’s ability for Synthetic daily streamflow Generation. https://ceej.aut.ac.ir/article_326_a666587afda6e89aec274a3657558a27.pdf