| description abstract | Bridge networks are a pillar of any national economy, and regular maintenance should be a high priority. Many transportation agencies require the condition assessment of concrete bridge decks as part of maintenance programs. Conventional, nondestructive bridge deck inspection techniques include hammer sounding and chain dragging. Ground penetrating radar (GPR) has been extensively used in recent years and has become an invaluable technique due to its high level of accuracy, speed, and low cost. This study provides a method to locate the corroded areas of the bridge deck slab in addition to getting structural properties such as rebar diameter, number and spacing. A clear methodology for the data collection, processing, and analysis of collected raw data is presented. Data is collected on site using ground-coupled GPR and then numerical amplitude method is used in the analysis. The study also clearly shows how to determine (1) bridge deck slab thickness, (2) asphalt layer thickness, (3) arrangement of steel rebar of the slab, (4) plotting corrosion map, and (5) generation deterioration model. Gathered data are grouped into four clusters using three clustering techniques based on the RapidMiner platform. GPR data interpretation is performed using integration between the amplitude method and visual interpretation, with threshold calibration based on K-means clustering. Simultaneously, the generated corrosion map is validated using the Weibull-based deterioration curves adopted by the New York State Department of Transportation and used since 1981, and the condition ratings used by the Federal Highway Administration in the United States. The corrosion index calculated from the clustered data is so close to the condition index obtained by the deterioration curve. A finite-element analysis is conducted in parallel to compare the stresses on the slab deck with the developed corrosion map. A sensitivity analysis is performed to identify the formula governing the amplitudes with the rebar area based on the finite-element analysis and the validated corrosion map. | |
