The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here: https://www.microsoft.com/en-us/microsoft-365/windows/end-of-ie-support).

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

Evaluation of artificial neural network techniques for river flow forecasting

På svenska

Author

Summary, in English

River runoff forecasting is one of the most complex areas of research in hydrology because of the uncertainty of hydrological and meteorological parameters and scarcity of adequate records. Artificial neural networks (ANN) can be an efficient way of modeling stream flow processes as it is capable of controlling and modelling nonlinear and complex systems and does not require describing the complex nature of the hydrological processes. In this study, daily river flow is forecasted using two ANN models: a Multi Layer Perceptron (MLP) network and a Radial Basis Function (RBF) Network. The ANN technique was applied to predict runoff in three mountain rivers in Georgia. The results show that ANNs can be successfully applied to forecast runoff using rainfall time series for the studied sub-catchments. A comparative study of both networks indicates that RBF models require little background knowledge of ANNs and need less time for development.

Publishing year

2007

Language

English

Pages

37-43

Publication/Series

Environmental Engineering and Management Journal

Volume

6

Issue

1

Document type

Conference paper

Publisher

Gh. Asachi Technical University of Iasi, Romania

Topic

  • Water Engineering

Keywords

  • radial basis function
  • modelling
  • rainfall-runoff
  • artificial neural network
  • multi layer perceptron
  • river flow forecasting

Conference name

3rd International Conference on Environmental Engineering and Management

Conference date

2006-09-23

Conference place

Iasi, Romania

Status

Published

ISBN/ISSN/Other

  • ISSN: 1843-3707
  • ISSN: 1582-9596