Stream flow forecasting by artificial neural network (ANN) model trained by real coded genetic algorithm (GA); A Case study when role of groundwater flow component in surface runoff is small

Sohail, A. R.*; Watanabe, Kunio*; Takeuchi, Shinji

Runoff analysis for precise prediction of discharge was carried out by artificial neural network model with real coded genetic algorithm (GAANN), back propagation artificial neural network model (BPANN) and multivariate autoregressive moving average model (MARMA). It was found that for very small catchments seasonal effect on the runoff is dominant. It was also found that estimation by ANN models was better than MARMA model for analyzing the responses to intense rainfalls in summer. The accuracy of the forecasts after several time periods in future was also investigated and found to decrease as time period is increased.