Multi-sensor data collection and fusion using autoencoders in condition evaluation of concrete bridge decks

Abstract

This paper presents a multi-sensor data collection and data fusion procedure for nondestructive evaluation/testing (NDE/NDT) of a concrete bridge deck. Three NDE technologies, vertical electrical impedance (VEI), ground-penetrating radar (GPR), and high-definition imaging (HDI) for surface crack detection, were deployed on the bridge deck. A neural network autoencoder was trained to quantify the relationship between VEI and GPR results using the data collected at common positions. This relationship was then used for fusion of VEI and GPR data to increase the reliability and spatial resolution of the NDE measurements and to generate a data-fused condition map that showed novel characteristics. Threshold values for VEI and GPR tests were obtained and used to determine the color scale in the fused map. Surface cracks identified from HDI show reasonable agreement with the deterioration areas on the data-fused condition map. Chloride concentration measurements on sound and deteriorated areas of the deck were consistent with the NDE results.

Real Depth-Correction in Ground Penetrating RADAR Data Analysis for Bridge Deck Evaluation

“…VEI is an NDE technique that uses a large-area probe which measures the impedance of the concrete cover when an alternate current is applied to the deck. Detailed data analysis for multiple NDE tests on this bridge has been reported in another paper [ 17 ]. Depth-correction of GPR amplitudes with migrated-depth method showed a good agreement between VEI and GPR [ 17 ].…”

“…Detailed data analysis for multiple NDE tests on this bridge has been reported in another paper [ 17 ]. Depth-correction of GPR amplitudes with migrated-depth method showed a good agreement between VEI and GPR [ 17 ]. In addition to VEI, at two locations on the bridge, chloride concentration testing was performed during the field testing [ 17 ].…”

“…Depth-correction of GPR amplitudes with migrated-depth method showed a good agreement between VEI and GPR [ 17 ]. In addition to VEI, at two locations on the bridge, chloride concentration testing was performed during the field testing [ 17 ]. The locations of two sampling points are shown in Table 2 and Figure 10 .…”

“…Region 1 is highlighted in all maps. For example, the amplitude range inside Region 1 is [0, −18] dB in Figure 10 a, while the same region shows a limited range of [−5, −13] dB in Figure 10 c, [−10, −18] dB in Figure 10 e. The attenuation rate range for the same region is [−2.2, −0.86] dB/cm in Figure 10 g. Compared to the depth- based correction methods and VEI map [ 17 ], the TWTT-based methods over correct the amplitudes and underestimate the deterioration in this region.…”

“…In the simulation data, the EM wave velocity is calculated based on Equation ( 1 ), and then migration is performed using the true velocity for each case. In practice, the actual wave velocity over each rebar can be obtained through an automated migration process [ 6 , 16 , 17 ].…”