Multiple sensor subsurface condition assessment of reinforced concrete bridge decks

This paper reports on a study aimed at accurately assessing the subsurface damage conditions of reinforced concrete bridge decks with multiple non-destructive evaluation (NDE) techniques. Information gained from such evaluations may significantly enhance the effectiveness of deck maintenance activities. The present state-of-the-art is that multiple NDE techniques are available, but that the accuracy and reliability of the methods are not guaranteed. NDE techniques generally employ electromagnetic, electrochemical or elastic wave principles. The sophistication ranges from the simple chain drag, to complicated multi-channel radar and ultrasound instruments. Recent field studies comparing NDE methods with each other and ground truth data have indicated a relatively high degree of variability and disagreement among the sensors. The foci of this study are threefold: 1. Compare the sensing data from multiple sensor methods applied to bridge deck specimens presently undergoing accelerated degradation in the laboratory. The sensors include multichannel ultrasound, ground penetrating radar (GPR), anode ladder, inductive rebar heating with infrared imaging and half-cell electrochemical potential. The accelerated degradation combines salt bath cycling with mechanical loading. 2. Attempt to understand the effect of different stages and type bridge degradation on the sensor signals. In particular, reinforcing corrosion, cracking around rebars and delamination are different, but related damage patterns. These different damage types can produce different effects on the sensors. For example, corrosion and water-chloride contamination may produce a strong absorption of radar waves, but produce relatively modest changes in ultrasound and chain drag tests. Conversely, air-filled delaminations are difficult to detect with standard GPR, but are readily detected with chain drag and impact-echo. 3. Explore methods of fusing the electrochemical, electromagnetic and elastic wave data based on degrees of belief in sensor information to produce an enhanced assessment of subsurface conditions. The ultimate goal is to produce an automated easy-to-use multisensor bridge deck assessment system.