Water Flows Modelling and Forecasting using a RBF Neural Network

Carlos Hernán Fajardo Toro, Daniel González Peña, Benedicto Soto González, Florentino Fernández Riverola


Hydrologic estimation model based on the utilisation of radial basis function neural networks is presented, in which the aim is to forecast stream flows in an automated fashion. The problem of river flow forecasting is a non-trivial task because (i) the various physical mechanisms governing the river flow dynamics act on a wide range of temporal and spatial scales and (ii) almost all mechanisms involved in the river flow process present some degree of nonlinearity. The proposed neural network was used to forecast daily river discharges in a river basin providing satisfactory results and outperforming previous successful techniques. The proposed model has been recently used to make hydrologic estimations in the Ulloa river, a river basin located in the north west of the Iberian Peninsula. The results obtained from the experiments are presented and discussed.


Radial Basis Function network; river flow forecasting; hydrologic models; black-box models; auto-regressive models.

DOI: http://dx.doi.org/10.18046/syt.v6i12.996


  • There are currently no refbacks.