Framework of Internal Damage Identification in Inhomogeneous Medium Interweaving Wave Scattering Model and Deep Learning

Various environmental conditions and loading forces may cause infrastructure material, concrete, and hot mix asphalt (HMA) to deteriorate. In particular, internal vertical cracks and internal reflective cracks (e.g., subsurface cracks) perpendicular to concrete surfaces are the most common, challenging, and critical types of infrastructure damage. Consequently, these extensive damages result in material property degradation, reinforcement corrosion, and even structural failure. Thus, effective detection of the cracks must be executed in a timely manner for better service life prediction and to monitor structural conditions at an early stage. There are recent advances in the study of surface-opening vertical crack detection (e.g., nonlinear diffuse ultrasonic waves, guided waves, and transmission energy). Despite these efforts, these studies for surface opening crack not internal damage, may present certain limitations and challenges for more in-depth understanding and monitoring of "internal" cracks. In particular, these internal reflective cracks commonly occur in many other infrastructures such as airport runway, pavement, and pipe, under the overlay caused by stress concentration at the bottom of the overlay. PI recently studied an analytical model to identify the internal reflective crack with various numerical integration methods to improve the analytical solution validated through finite element (FE) simulations and experimental study [8]. The advantage of this approach is that it provides an accurate depth-to-crack distance by using the relation between scattering energy, so-called wave response variation (WRV), and crack geometry. However, huge challenges in this effort of the analytical modeling for identifying are to reduce the gap between the nonlinear analytical and numerical WRV model and experimental WRV result; to identify the material inhomogeneity effect in the wave scattering model (WSM); to define the physics-based interpolations with machine learning (ML) technique. The followings are primary research gaps that need to be addressed in this project. Consequently, the project's overall goal is to advance understanding of a WSM of an internal vertical reflective crack in inhomogeneous material (IHM) leveraging deep learning. The testing data and its analysis of WRV by the crack and toward the establishment of a unique analytical model will be then integrated into IWSM with the physics-based ML interpolation for complex features and environments, potentially for large applications (e.g., buried concrete pipe in soil, one side accessible slab, reflective cracks from the concrete pavement joint). The project will also carry out the Trans-SET missions by performing research, technology transfer, education, workforce development, and outreach activities to solve transportation challenges in Region 6.

  • Supplemental Notes:
    • 22PUTA33


  • English


  • Status: Active
  • Funding: $122000
  • Contract Numbers:


  • Sponsor Organizations:

    Office of the Assistant Secretary for Research and Technology

    University Transportation Centers Program
    Department of Transportation
    Washington, DC  United States  20590
  • Managing Organizations:

    Transportation Consortium of South-Central States (Tran-SET)

    Louisiana State University
    Baton Rouge, LA  United States  70803
  • Project Managers:

    Mousa, Momen

  • Performing Organizations:

    University of Texas at Arlington

    Box 19308
    Arlington, TX  United States  76019-0308
  • Principal Investigators:

    Ham, Suyun

  • Start Date: 20220401
  • Expected Completion Date: 0
  • Actual Completion Date: 0
  • USDOT Program: University Transportation Centers Program

Subject/Index Terms

Filing Info

  • Accession Number: 01844949
  • Record Type: Research project
  • Source Agency: Transportation Consortium of South-Central States (Tran-SET)
  • Contract Numbers: 69A3551747106
  • Files: UTC, RIP
  • Created Date: May 9 2022 6:23AM