Earthquake Damage Detection in Water Distribution System

In a destructive earthquake considerable damage occurs in buried structures such as urban water supply systems. The preliminary action after an earthquake happens is the repair of the network supplying water to the damaged area. Since pipes are mostly buried the damages cannot be found, even by digging a long path. The most difficult part of the work is to find the damaged locations. In this paper an artificial neural network (ANN) is used to detect pipeline damage in water distribution systems due to an earthquake. The probable failure states have to be computed. The amount of output discharge from the tanks is obtained by direct analysis of a mathematical model for different states of pipe failure. The failure points are obtained by using the amount of water discharge from the tanks. The damaged pipes could be detected through a back analysis. Since there are various types of failure, the ANN is applied. Through a parametric study, different geometry, shape, diameter and pressure of the water network are surveyed and the best network architecture for each case is obtained. The peak responses and phase delays are assumed to be the network outputs. The network is trained in a supervised manner by the data obtained from direct analyses. The study shows the efficiency and capability of the ANN for modeling the observed nonlinear behavior.

Language

  • English

Media Info

  • Media Type: CD-ROM
  • Features: Figures; References; Tables;
  • Pagination: 10p
  • Monograph Title: Pipeline Engineering and Construction: What's on the Horizon? Proceedings of the Pipelines 2004 International Conference, August 1-4, 2004, San Diego, California

Subject/Index Terms

Filing Info

  • Accession Number: 01003279
  • Record Type: Publication
  • ISBN: 0784407452
  • Files: TRIS
  • Created Date: Aug 22 2005 4:12PM