Detection and reconstruction of complex structural cracking patterns with electrical imaging

Danny Smyl, Mohammad Pour-Ghaz, Aku Seppänen

Research output: Contribution to journalArticleScientificpeer-review

15 Citations (Scopus)


The ability to detect cracks in structural elements is an integral component in the assessment of structural heath and integrity. Recently, Electrical Resistance Tomography (ERT) -based sensing skins have been shown to reliably image progressive surface damage on structural members. However, so far the approach has only been tested in cases of relatively simple crack patterns. Because the spatial resolution of ERT is generally low, it is an open question whether the ERT-based sensing skins are able to image complex structural cracking patterns. In this paper, we test the accuracy of ERT for reconstructing cracking patterns experimentally and computationally. In the computational study, we use a set of numerical simulations that model progressive cracking in a rectangular beam geometry. We also investigate the effect of image reconstruction methods on the crack pattern estimates: In addition to the contemporary image reconstruction method used in the recent sensing skin studies, we test the feasibility of a novel approach where model-based structural prior information on the cracking probability is accounted for in the image reconstruction. The results of this study indicate that ERT-based sensing skins are able to detect and reconstruct complex structural cracking patterns, especially when structural prior information is utilized in the image reconstruction.
Original languageEnglish
Pages (from-to)123-133
Number of pages11
Publication statusPublished - Jul 2018
MoE publication typeA1 Journal article-refereed


  • Damage detection
  • Finite element analysis
  • Image analysis
  • Inverse problem

Fingerprint Dive into the research topics of 'Detection and reconstruction of complex structural cracking patterns with electrical imaging'. Together they form a unique fingerprint.

  • Cite this