Detectability of delamination regions using infrared thermography in concrete members strengthened by CFRP jacketing

Jian Cheng Gu, Shigeki Unjoh, Hideki Naito

Research output: Contribution to journalArticlepeer-review

20 Citations (Scopus)

Abstract

In this study, we analyzed the accuracy of a nondestructive testing (NDT) method using infrared thermography for detecting delamination in CFRP-jacketed concrete members. Based on an existing bridge strengthened by CFRP jacketing, four specimens were designed and evaluated. Sixteen artificial delaminations with different parameters were created in these specimens. The conditions of the artificial delaminations, including size and depth, presence of surface cover mortar, and water content in the delamination void, were considered as test parameters. The infrared thermography was applied to the specimens with active heating. Changes in the surface temperature were captured with time, and the time history change of the surface temperature in each pixel was analyzed. A method to identify the boundaries and regions of delamination was proposed and studied. The results showed that the detectability was positively correlated with the size and depth of the delamination, while the presence of surface cover mortar and water decreased the accuracy of the detection. The proposed method could successfully identify delamination regions that were at least 5 cm × 5 cm in size. However, it failed to identify the regions with just the delamination but almost no measurable depth, because these regions had less heat-insulated air in voids.

Original languageEnglish
Article number112328
JournalComposite Structures
Volume245
DOIs
Publication statusPublished - 2020 Aug 1

Keywords

  • Active heating test
  • Boundary detection
  • CFRP jacketing
  • Delamination
  • Infrared thermography
  • NDT

Fingerprint

Dive into the research topics of 'Detectability of delamination regions using infrared thermography in concrete members strengthened by CFRP jacketing'. Together they form a unique fingerprint.

Cite this