Rebar detection — POD approach to determine the reliability of GPR systems and to quantify the influence of different material parameters

Abstract

GPR has emerged as one of the most relevant NDT technologies, which can be employed in various civil engineering (CE) applications. While the general applicability of GPR has been demonstrated, an objective evaluation of the limitations of GPR for qualitative inspection tasks is still missing. Therefore, the POD method, based on the “â vs. a” approach, was transferred to the application of GPR for the detection of rebars at different depths. The main goal of the presented research was to establish a universal procedure, which is based on the POD approach to determine the detection limits of a GPR system and excludes the human factor. Important parameters are the a90/95 value, the slope of the POD curve, and the decision threshold âdec. With this information, different GPR systems can be compared objectively to choose the best system for each individual inspection task. Furthermore, a program of measurements on four pairs of specimens was conducted to determine the influence of different material parameters. This experiment was based on the factorial design to separate each investigated material parameter. In summary, the maximum grain size has no systematic influence on the reliable detection depth a90/95 of metallic rebars. As a further important result of the POD analysis, a rebar with a small diameter of 12 mm is detectable at larger depths than rebars with a diameter of 28 mm. This has to be taken into consideration, when tendons with larger diameters than rebars have to be detected.