An established damage detection technique in mechanical and aerospace engineering is to measure and analyze vibration signals in a machine component. Vibration signature analysis of steel bridge structures in a nonconventional way using artificial neural networks is presented. Multilayer perceptrons were adopted using the back-propagation algorithm for network training. The training patterns regarding vibration signature are derived analytically for a moving load traveling on a trussed bridge structure at a constant speed to simulate the inspection vehicle. Using the finite-element method, the moving forces are converted into stationary time-dependent force functions, generating vibration signals in the structure and the same is used to train the network. The performance of the trained networks is studied for their ability to detect damage from unknown signatures taken independently at one, three, and five nodes. Based on observation, the authors report that the prediction using the trained network with single-node signature measurement at a suitably chosen site is even better than that of three-node and five-node measurement data.


  • English

Media Info

  • Features: Appendices; Figures; References; Tables;
  • Pagination: p. 259-265
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00712689
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Oct 5 1995 12:00AM