Clustering-Based Threshold Model for Condition Assessment of Concrete Bridge Decks with Ground-Penetrating Radar

Abstract

Ground-penetrating radar (GPR) has been extensively studied for condition assessment of concrete bridge decks in North America. Although several methods for analyzing GPR data have been proposed, the commonly accepted method evaluates the condition of concrete bridge decks on the basis of the difference between reflection amplitudes of the top rebar layer. It is assumed in the method that strong reflection indicates sound concrete, whereas the area with high-amplitude attenuation is associated with concrete corrosion. The final result is a contour map of reflection amplitude in decibel scale with the thresholds selected arbitrarily to define the severity of concrete deterioration. Because subjective determination of threshold values may lead to inconsistency in the result obtained, this paper proposes a robust method for resolving that issue. Specifically, after depth correction was performed for top rebar amplitudes, on the basis of K-means clustering technique these amplitude data were grouped into a number of condition categories. Through two case studies in North America, the methodology was implemented and compared with the results provided by other technologies, namely, concrete resistivity, half-cell potential, and laboratory chloride content analysis. The implementation showed that while the proposed method was simple to employ, it still provided reasonable results that were in line with the outputs provided by the other techniques.

Non-Destructive Corrosion Inspection of Reinforced Concrete Using Ground-Penetrating Radar: A Review

“…The relationship can be used to predict the corroded area based on the analysis of the statistical parameters of the amplitudes. In some studies [ 77 , 78 ], the k-means clustering method was used to determine thresholds values.…”

“…In addition to the numerical approach, a visual or combined visual and numerical approach has been supported by a number of authors [ 22 , 77 , 85 ]. The visual approach implies the visual analysis of B-scans.…”

“…A set of if/then rules was created to locate anomalies that alter the signal but do not indicate deterioration. Dinh et al [ 77 ] used visual analysis of ground-penetrating radar profiles as a tool to determine the number of condition categories as input to the k-means clustering method. It is a combined method: after determining the number of condition categories, the amplitudes of the signal are grouped and thresholds between the groups are determined.…”

Ground penetrating radar-based deterioration assessment of RC bridge decks

“…Another method was developed by Dinh et al (2015) where this method combines numerical amplitude and image-based analyses in a model named clustering-based threshold calibration. In this model, the analyst decides on the number of condition categories that the bridge deck should have based on analyzing the profiles visually and then, by using k-means clustering technique, the amplitudes of the top reinforcing bars will be divided into the same number of categories determined by the analyst.…”

Attenuation-Based Methodology for Condition Assessment of Concrete Bridge Decks Using GPR

“…Concerning the latter method, although a threshold of -6 to -8 dB was written in the standard, these values are yet a research topic. Most recently, Dinh et al [9] developed a model to determine flexible amplitude thresholds, based on K-means clustering technique. In another research, Martino et al [11] tried to develop a threshold model based on the distribution features of depth-corrected amplitude.…”