Graduation Year

2026

Document Type

Thesis

Degree

M.S.C.E.

Degree Name

MS in Civil Engineering (M.S.C.E.)

Degree Granting Department

Engineering

Major Professor

Ayumi Manawadu, Ph.D.

Committee Member

Gray Mullins, Ph.D.

Committee Member

Tristen Mee, Ph.D.

Keywords

Infrared Thermography, Machine Learning, Object Detection, Tap Test, Visual Inspection

Abstract

Externally bonded fiber-reinforced polymer (FRP) composites are widely used to strengthen reinforced concrete (RC) structures. However, the long-term performance of these systems depends on the integrity of the FRP–concrete bond. Interfacial defects such as debonding, delamination, cracking, and voids may reduce structural capacity and are often difficult to detect using conventional inspection techniques. This study investigates the effectiveness of nondestructive testing (NDT) methods for detecting interfacial defects in FRP-strengthened RC members. It explores the potential for automated defect identification using thermal imaging and machine learning.

Laboratory specimens with predefined defects were fabricated and tested using visual inspection (VI), tap testing (TT), and infrared thermography (IRT). The same inspection methods were also applied to field specimens consisting of existing prestressed concrete piles rehabilitated with carbon FRP. The results from the different NDT methods were compared in terms of defect detectability, reliability, and field applicability.

Thermal images obtained from laboratory and field testing were post-processed and used to develop an object-detection model based on the YOLO (You Only Look Once) algorithm. The model was trained to identify multiple defect types, including cracking, debonding, delamination, and spalling. The automated detection approach demonstrated the ability to identify defects from thermal images with varying levels of accuracy, providing a quantitative alternative to traditional qualitative interpretation.

The results indicate that visual inspection is effective for identifying surface-level defects, while tap testing and infrared thermography can detect subsurface damage. Infrared thermography, combined with machine learning, showed strong potential for rapid, automated defect detection in FRP-strengthened RC members. The findings of this study contribute toward the development of more reliable and quantitative inspection methods for FRP–concrete bond evaluation and support the implementation of data-driven approaches for infrastructure assessment.

Download

Share

COinS