Signal Processing Methods for Flood Early Warning Systems

Authors
  • A.L. Pyayt
  • I.I. Mokhov
  • A.P. Kozionov
  • V.T. Kusherbaeva
Publication date 2012
Host editors
  • R. Hinkelmann
  • M.H. Nasermoaddeli
  • S.Y. Liong
  • D. Savic
  • P. Fröhle
  • K.F. Daemrich
Book title Proceedings of 10th International Conference on Hydroinformatics 2012: Hamburg, Germany
ISBN
  • 9783941492455
Event HIC 2012 - 10th International Conference on Hydroinformatics - "Understanding Changing Climate and Environment and Finding Solutions"
Publisher Hamburg: TuTech Innovation
Organisations
  • Faculty of Science (FNWI) - Informatics Institute (IVI)
Abstract We present in a data-driven approach for detection of anomalies in earthen dam (dike) behaviour that can indicate the onset of flood defence structure failure. This approach is implemented in the UrbanFlood early warning system's Artificial Intelligence component that processes dike measurements in on-line manner. Suggested approach successfully detected anomalies registered during the non-destructive piping experiment at the Zeeland dike (the Netherlands).
Document type Conference contribution
Language English
Permalink to this page
Back