Numerical simulation of elastic waves for visualization of defects

D. G. Aggelis, N. Tsimpris, H. K. Chai, T. Shiotani, Y. Kobayashi

Research output: Contribution to journalArticlepeer-review

27 Citations (Scopus)


Due to aging of civil infrastructure, non-destructive testing (NDT) is of continuously increasing significance. The ultimate aim is the detection of defects in the interior of a structure with the highest possible accuracy, in order to proceed to specific repair action. Pulse velocity tomography is a technique to visualize the interior. Using a number of pulsers and receivers for introduction and recording of the transient elastic wave at different points on the surface of a structure, a potentially dense network of interconnecting wave paths are examined as to their wave transit time. Each "cell" of the cross section is assigned a velocity value and the interior is visualized. However, the confirmation of the results is not always possible except for cases where the cracks are visible from the surface or in case specific model defects are placed during casting of specimens. In the present paper numerical simulations of wave propagation were conducted for different geometries. Different types of inhomogeneity were applied, such as voids, deteriorated zones, as well as steel reinforcing bars. The transit time information calculated in these cases, led to the corresponding tomograms. Based on the comparison between the actual geometry of the void and the tomograms, the accuracy of the technique is studied, as well as parameters to improve the testing, like applied excitation frequency and number of sensors.

Original languageEnglish
Pages (from-to)1503-1512
Number of pages10
JournalConstruction and Building Materials
Issue number4
Publication statusPublished - 1 Apr 2011


  • Concrete
  • Elastic waves
  • Non-destructive testing
  • Simulation
  • Tomography
  • Voids


Dive into the research topics of 'Numerical simulation of elastic waves for visualization of defects'. Together they form a unique fingerprint.

Cite this