Water Flows Modelling and Forecasting using a RBF Neural Network

  • Carlos Hernán Fajardo Toro Universidad de Vigo
  • Daniel González Peña Universidad de Vigo
  • Benedicto Soto González Universidad de Vigo
  • Florentino Fernández Riverola Universidad de Vigo
Keywords: Radial Basis Function network, river flow forecasting, hydrologic models, black-box models, auto-regressive models.

Abstract

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.

Downloads

Download data is not yet available.

Author Biographies

Carlos Hernán Fajardo Toro, Universidad de Vigo
Ph.D. from the University of Vigo, Spain. He was born in Cali, Colombia, in 1968. He works as an Associate and Full time Professor at the Icesi University in Cali – Colombia from 1994 to 2001 and from 2001 as an Associate Professor for the Com­puter Science Department of the University of Vigo, collaborating as investigator with the research group SING (Computer Systems of New Generation) belonging to the University of Vigo and group IREHISA belonging the Universi­dad del Valle in Colombia. Talking about his investigation field, he is cantered in the study of hybrid methods of Artificial Intelligence and their application to real prob­lems, although he has also worked in topics related with software quality and project management focused on software development. (http://sing.ei.uvigo.es/).
Daniel González Peña, Universidad de Vigo

Ph.D. stu­dent in the Computer Science Department at the University of Vigo, Spain. His research work is focused in the study of hybrid methods of Artificial Intelligence such as CBR (Case Based Rea­soning), and its application to real problems like spam filtering and bioinformatics. During the last years, he has been working as Web developer in some other projects and as Java developer in some Open Source projects such as GeneCBR (www.genecbr.org), AIBench, (www.aibench.org), and JavaTraceIt! (javatraceit.siteinteresa.org). He collaborates with the research group SING at the Computer Engineering School at University of Vigo. He is joint author of several books and book’s chapters about Web, Java and Java Enterprise devel­opment.

Benedicto Soto González, Universidad de Vigo
Ph.D. from the University of Santiago de Compostela, Spain. He was born in Pontevedra, Spain in 1963. He works as a Titular Professor for the Biology and soil sciences Department of the University of Vigo, collaborating as investigator with the Group of pedology and agricultural chemistry belonging to the University of Vigo. Talking about his investigation field, he is cantered in the study of soils and land management, although he has also worked in topics related with hydrology analysis. He is joint author of several books and book’s chapters, as well as author of different articles published by well-known editorials like Sprin­ger-Verlag, Ios Press, Kluwer, etc. (http://webs.uvigo.es/webc02/).
Florentino Fernández Riverola, Universidad de Vigo
Ph.D. from the University of Vigo, Spain. He was born in Langen-Hessen, Germany, in 1973. He works as an Associate Professor for the Computer Science Depar­tment of the University of Vigo, collaborating as investigator with the Group of Intelligent Computer Systems of the University of Sa­lamanca and the research group SING (Computer Systems of New Generation) belonging to the Uni­versity of Vigo. Talking about his investigation field, he is cantered in the study of hybrid methods of Artificial Intelligence and their application to real problems, although he has also worked in topics related with the develop­ment of communication protocols for wireless networks. He is joint author of several books and book’s chapters, as well as author of different articles published by well-known editorials like Sprin­ger-Verlag, Ios Press, Kluwer, etc. (http://sing.ei.uvigo.es/).
Published
2008-12-17
Section
Original Research