Rational condition assessment of RC bridge decks subjected to corrosion-induced delamination

Precise evaluation of the condition of RC bridge decks, particularly for corrosion-induced delaminations, is decisive for ensuring bridge performance and safety. Although infrared thermography (IRT) and ground penetrating radar (GPR) have considerably aided this task, cogent procedures for deciphering their data have yet to be developed. Postprocessing of IRT and GPR data still relies on user interpretation to construe quantitative appraisal of subsurface anomalies, where predefined threshold values are arbitrarily selected. Such a subjective analysis can produce inconsistent results. The present study proposes a robust procedure for analysing IRT and GPR test data and uniting their results. A mosaicked thermogram of the entire bridge deck from thermal images was created using custom-developed codes, whereas the numerical amplitude method was used to analyze GPR data. To identify objective thresholds, the k-means clustering function was used. Accordingly, condition maps of bridge decks delineating different severity levels of delamination and potential active corrosion could be generated. The analyzed IRT and GPR data were then integrated to develop an overall bridge-deck condition map. The proposed procedure was instigated to analyze thermal and radar data acquired from two full-scale RC bridge decks. It is shown that rationally integrating IRT and GPR results offers a superior tool for detecting subsurface anomalies in bridge decks. Thus, it could enhance bridge-deck inspection programs, allowing stakeholders to optimize budgets and prioritize maintenance, repair, and rehabilitation efforts.