Monitoring damage in concrete structures is crucial for maintaining the health of structural systems. The implementation of computer vision has been the key for providing accurate and quantitative monitoring. Recent development uses the robustness of deep-learning-aided computer vision, especially the convolutional neural network model. The convolutional neural network is not only accurate but also flexible in various scenarios. The convolutional neural network has been constructed to classify image in terms of individual pixel, namely pixel-level detection, which is especially useful in detecting and classifying damage in fine-grained detail. Moreover, in the real-world scenario, the scenes are mostly very complex with varying foreign objects other than concrete. Therefore, this study will focus on implementing a pixel-level convolutional neural network for concrete surface damage detection with complicated surrounding image settings. Since there are multiple types of damage on concrete surfaces, the convolutional neural network model will be trained to detect three types of damages, namely cracks, spallings, and voids. The training architecture will adopt U-Net and DeepLabV3+. Both models are compared using the evaluation metrics and the predicted results. The dataset used for the neural network training is self-built and contains multiple concrete damages and complex foregrounds on every image. To deal with overfitting, the dataset is augmented, and the models are regularized using L1 and Spatial dropout. U-Net slightly outperforms DeepLabV3+ with U-Net scores 0.7199 and 0.5993 on F1 and mIoU, respectively, while DeepLabV3+ scores 0.6478 and 0.5174 on F1 and mIoU, respectively. Given the complexity of the dataset and extensive image labeling, the neural network models achieved satisfactory results.

Corrosion of the steel reinforcing bars in concrete structures is one of the major maintenance problems. Corrosion results in expansive pressure on the surrounding concrete, which causes internal damage that may become visible as surface cracking. Such damage may degrade structural safety and serviceability. Effective maintenance requires the evaluation of residual performance based on estimates of spatially nonuniform levels of corrosion, which are typically obtained through surface measurements only. In this study, the authors have developed a simulation system for estimating the levels of internal corrosion along the reinforcing bar length from surface crack information. This innovative system is produced by integrating the technique of Model Predictive Control (MPC) with Rigid‐Body‐Spring Models (RBSM) of corrosion‐induced cracking at the concrete mesoscale. In this study, MPC controls the simulated surface cracks such that they match the observed cracks by optimizing the internal expansions of springs representing the steel‐concrete interface within the RBSM. The applicability of the system is verified using both synthetic crack width data and crack data collected from in‐house laboratory testing. In the laboratory testing, corrosion levels were quantified by 3D scanning of the extracted reinforcing bars. The simulation results agree with the corrosion measurements, demonstrating the potential of the MPC‐RBSM system for predicting the corrosion distribution along reinforcing bars using surface crack data.

Bond strength assessment is important for reinforced concrete structures with rebar corrosion since the bond degradation can threaten the structural safety. In this study, to assess the bond strength in concrete-corroded rebar interface, one of the machine learning techniques, artificial neutral network (ANN), was utilized for the application. From existing literature, data related to the bond strength of concrete and corroded rebar were collected. The ANN model was applied to understand the factors on bond property degradation. For the input in the ANN model, the following factors were considered the relative bond strength: (1) corrosion level; (2) crack width; (3) cover-to-diameter ratio; and (4) concrete strength. For the cases with confinement (stirrups), (5) the diameter/stirrups spacing ratio was also considered. The assessment was conducted from input with single parameter to multiple parameters. The scaled feed-forward multi-layer perception ANN model with the error back-propagation algorithm of gradient descent and momentum was found to match the experimental and computed results. The correlation of each parameter to the bond strength degradation was clarified. In cases without confinement, the relative importance was (1) > (2) > (4) > (3), while it was (2) > (1) > (3) > (5) > (4) for the cases with confinement.

Nowadays, seismic design code in Japan is becoming more stringent. To satisfy the strict requirement, larger numbers of reinforcements must be placed, resulting in increased reinforcement congestion. To reduce the reinforcement congestion, mechanical anchorage is becoming popular in use instead of conventional hook rebar. However the behavior of mechanical anchorage placed in thin cover zone is not well understood, and the use of this is still limited. In this study, the discrete element method was used to investigate the performance of mechanical anchorage placed in thin cover depth zone, and the pull-out test in previous research was simulated. To assume that anchor plates welded to bar end are difficult to place side by side due to closely-spaced reinforcements, the position of anchor plate was taken as main parameter. The result shows that the anchorage performance changed according to the transverse bars effect and the shift of the mechanical anchorage. The simulations show good agreement with experiment data in terms of the anchorage capacity, crack pattern and failure mode.

The application of reinforced concrete for permanent and temporary deep ocean structures has recently become more prevalent; however, the static and dynamic effects of high water pressure on concrete remain unexplored. This paper investigates the influence of high water pressure (60 MPa) on four series of concrete cylinders with and without an embedded steel bar under sustained and cyclic loading conditions. The residual compressive strength, bond strength, and associated evolution of surface and internal damage are evaluated after exposing concrete cylinders to a water pressure of 60 MPa. The first series is exposed to sustained water pressure for 7 and 60 days, while the other series is tested under repeated water pressure for 10, 20, 30, 60, and 150 cycles. The results reveal that residual compressive strength falls immediately by 16% within 7 days of sustained high water pressure, but the strength then remains stable up to 60 days. Under repeated high water pressure, residual compressive strength gradually reduces by up to 40% until 60 cycles, after which it remains reasonably stable until 150 cycles as crack propagation is arrested at a certain depth within the concrete cylinders. The bond strength between the steel bar and matrix is observed to decrease considerably under repeated cycles of 60 MPa water pressure up to 26%. The damage gradually propagates at the matrix/steel bar interface under the repeated water pressure, resulting in a reduction in residual pullout capacity.