Open AccessUsing artificial neural network in simulating of the streamflow in the Srepok watershed
Author(s) -
Khoi Nguyen Dao,
Phuong Ai Huynh
Publication year - 2016
Publication title -
khoa học công nghệ
Language(s) - English
Resource type - Journals
ISSN - 1859-0128
DOI - 10.32508/stdj.v19i2.796
Subject(s) - streamflow , watershed , precipitation , lag , artificial neural network , environmental science , correlation , lag time , series (stratigraphy) , meteorology , climatology , mathematics , computer science , machine learning , geology , geography , biological system , cartography , drainage basin , computer network , paleontology , geometry , biology
In this study, artificial neural network (ANN) model was used to simulate the streamflow in the Srepok watershed, Vietnam. Correlation analysis of time series for precipitation and streamflow was employed to determine input data for the ANN model. This result indicated a significant correlation up to 2 day time lag and 1 day time lag for the precipitation and streamflow series data, respectively. According to the correlation analysis, three ANN models including ANN1, ANN2, and ANN3 were investigated. A 3-year data record for the precipitation and streamflow was used for ANN training and testing. The result of ANN training and testing showed that the ANN2 with 3 input data (P(t), P(t-1), and Q(t- 1)) gave the best simulation (NSE = 0.95 for training period and NSE = 0.96 for testing period) comparing to those of ANN1 and ANN3. In addition, the comparison of ANNs showed that the increase of the input data did not offer the better result.