TITLE:
Streamflow Decomposition Based Integrated ANN Model
AUTHORS:
Nikhil Bhatia, Laksha Sharma, Shreya Srivastava, Nidhish Katyal, Roshan Srivastav
KEYWORDS:
Artificial Neural Network; Rainfall-Runoff Modeling; Streamflow Decomposing; Black Box Modelling
JOURNAL NAME:
Open Journal of Modern Hydrology,
Vol.3 No.1,
January
28,
2013
ABSTRACT:
The prediction of riverflows requires the understanding of rainfall-runoff process which is highly nonlinear, dynamic and complex in nature. In this research streamflow decomposition based integrated ANN (SD-ANN) model is developed to improve the efficacy rather than using a single ANN model for the flow hydrograph. The streamflows are decomposed into two states namely 1) the rise state and 2) the fall state. The rainfall-runoff data obtained from the Kolar River basin is used to test the efficacy of the proposed model when compared to feed-forward ANN model (FF-ANN). The results obtained in this study indicate that the proposed SD-ANN model outperforms the single ANN model in terms of both the statistical indices and the prediction of high flows.