Prediction of Sediment Deposition in Reservoir Using Artificial Neural Networks

The use of art ificial neural network (ANN)modeling for prediction and forecast ing variables in water resources engineering are being increasing rapidly Infrastructural applicat ions of ANN in terms of selection of inputs, architecture of networks, training algorithms, and selection of training parameters indifferent types of neural networks used in water resources engineering have been reported. ANN is a vigorous technique to develop immense relat ionship between the input and output variables, and able to extract complex behaviour between the water resources variables such as rainfall, inflow, capacity of the reservoir and sediment deposition. An appropriate training based ANN model is able to adopt the physical understanding between the variables and may generate more effective results than conventional predict ion techniques. Annual rainfall; annual inflow and capacity of the reservoir were chosen as inputs. Fifty three years of data pertaining to Shivaji sagar Reservoir on the Koyna River in Maharashtra, India, were used in this study. The pattern of the sediment volume retained in this reservoir was well captured by the Multi-Layer Perceptron 3-10-5-1 ANN model using the back propagation algorithm. Based on several performance indices, it was found that the ANN model estimated the volume of sediment retained in the reservoir with better accuracy and less effort as compared to conventional regression analysis