FORECASTING OF MONTHLY AVERAGED STREAMFLOW DATA USING ANN AND SVM

Abstract

River flow forecasting is required to provide important information on a wide range of cases related to design and operation of river systems. Since there are a lot of parameters with uncertainties and non-linear relationships, the calibration of conceptual or physically-based models is often a difficult and time consuming procedure. So it is preferred to implement a heuristic black box model to perform a non-linear mapping between the input and output spaces without detailed consideration of the internal structure of the physical process. The present study comprises development of Artificial Neural Network and Support Vector Machine approach to forecast the monthly averaged stream flow data of seven rivers. The Feed Forward Neural Network with Back propagation algorithm and Support Vector Regression methods are used in this study. Trial and error method is adopted for training the Artificial Neural Network. Trained model was used to forecast the monthly averaged data for particular river. A comparative study of both networks indicates that both models can be used for hydrologic time series forecasting, if sufficient past record is available without many gaps