Vol. 7 No. 2
Hydrological processes modeling using RBNN -a neural computing technique
Abstract: Severe erosion in the watersheds under Damodar Valley Corporations (DVC), Hazaribagh, Jharkhand, India has been taking place for a long time and several soil and water conservation measures are being adopted by the Soil Conservation Department under DVC. For effective planning of soil conservation programs, hydrologic models can always be of help. Radial basis neural network (RBNN) model is neural network model which requires lesser data. In the present study, one of the watersheds under DVC named Nagwa was selected for simulating surface runoff and sediment yield. Maximum and minimum daily temperature and rainfall were used as input for RBNN model training and validation for surface runoff and runoff was included when simulating for sediment yield. The RBNN model was trained for the vear 1991-2000 and validated for the year 2005-2007. Results indicate that coefficient of determination (R2), Nash-Sutcliffe simulation efficiency (NSE) and root mean square error (RMSE) values for SWAT model were found to be 0.74, 0.74 and 0.41 mm during training and 0.65, 0.63 and 4.15 mm during validation period respectively. The model performed quite well for simulation of sediment yield with R2, NSE and RMSE values of 0.77, 0.69 and 0.24 t ha-1 during training period and 0.88, 0.82 and 0.65 t ha-1during validation period, respectively. It could be stated that RBNN model based on simple input could be used for estimation of monthly runoff, sediment yield, missing data, and testing the accuracy of other models.
Download Full Article: Click Here