Accurate seismic ground response analysis is crucial for the design and safety of civil infrastructure and establishing effective mitigation measures against seismic risks and hazards. This is a complex process due to the nonlinear soil properties and complicated underground geometries. As a simplified approach, the one-dimensional wave propagation model, which assumes that seismic waves travel vertically through a horizontally layered medium, is widely adopted for its reasonable performance in many practical applications. This study explores the potential of sequence deep learning models, specifically 1D convolutional neural networks (1D-CNNs), long short-term memory (LSTM) networks, and transformers, as an alternative for seismic ground response modeling. Utilizing ground motion data from the Kiban Kyoshin Network (KiK-net), we train these models to predict ground surface acceleration response spectra based on bedrock motions. The performance of the data-driven models is compared with the conventional equivalent-linear analysis model, SHAKE2000. The results demonstrate that the deep learning models outperform the physics-based model across various sites, with the transformer model exhibiting the smallest average prediction error due to its ability to capture long-range dependencies. The 1D-CNN model also shows a promising performance, albeit with occasional higher errors than the other models. All the data-driven models exhibit efficient computation times of less than 0.4 s for estimation. These findings highlight the potential of sequence deep learning approaches for seismic ground response modeling.

In this study, coating equipment for the effective underwater repair of submerged structures was developed. The tensile bond characteristics of selected epoxy resin coatings were investigated by coating the surface of a specimen using each of the four types of equipment. Using the experimental results, the tensile bond strength and the coating thickness were analyzed according to the type of equipment, coating, and curing time. The results show that the type of coating equipment used had the greatest effect on the measured bond strength and coating thickness of the selected coatings. However, the effect of coating type and curing time on the bond strength and the thickness was observed to be insignificant. Compared with the developed equipment, the surface treatment of the coating was observed to be more effective when using the pre-existing equipment, and thus the bond performance of the coating was improved compared to using the pre-existing equipment. Based on the experimental results, improvements and needs involving the equipment for further research were discussed.

In this study, we estimate the potential efficiency of waste fishing net (WFN) fibers as concrete reinforcements. Three WFN fiber concentrations (1, 2, and 3% by volume) were mixed with concrete. Compressive strength, toughness, splitting tensile strength, and biaxial flexural tests were conducted. Compressive strength decreased but other properties increased as a function of fiber proportions. According to the mechanical strength observations and the ductility number, WFN fibers yielded benefits in crack arresting that improved the postcracking behavior and transformed concrete from a brittle into a quasi-brittle material. It is inferred that WFN fiber is a recycled and eco-friendly material that can be utilized as potential concrete reinforcement.

The efficient and accurate detection of cracks in concrete structures is critical for maintaining structural integrity and safety. This study compares two state-of-the-art convolutional neural network (CNN) models, Mask R-CNN and YOLOv8, for automated concrete crack detection, each model representing two mainstream approaches for object detection and instance segmentation: single-stage and two-stage approach. We evaluate both models on 1,203 concrete images with 7:2:1 training, testing, and validation split, and assess their accuracy and processing speed. Mask R-CNN achieves a mean Intersection over Union (IoU) of 96.5% with a minimum IoU of 77% and higher consistency, compared to YOLOv8’s 90.6%, which often shows complete failure with IoU of 0%. In terms of computation speed, YOLOv8 shows 0.3225 s of average processing time per image, slightly outperforming the speed of Mask R-CNN, 0.4867 s. Despite YOLOv8’s faster processing speed, considering the characteristics of concrete crack detection tasks where accuracy should be prioritized over speed, Mask R-CNN seems a more proper model for reliable crack detection. We also show the accuracy of Mask R-CNN for crack detection tasks can be further enhanced by employing the ResNeXt backbone.

The management of plastic waste is a massive challenge and the recycling of plastics for newer applications is a potential solution. This study investigates the feasibility of using polyethylene terephthalate (PET) powder in cementitious composites. The changes in the strength and microstructure of Portland cement incorporating PET powder with different replacement ratios were systematically analyzed through the measurements of compressive strength, isothermal calorimetry, X-ray diffraction, thermogravimetric analysis, and Raman spectroscopy. In addition, the possible chemical changes of cement paste samples were studied upon exposure to different conditions, including deionized water, seawater, and simulated pore solution. Based on the test results and analysis, no apparent chemical changes were observed in the cement paste samples, regardless of the exposure conditions. In contrast, the PET powder incorporated into concrete exhibited remarkable changes, which may have occurred during the mixing process. The results also suggested that the maximum replacement ratio of PET powder should be less than 10% of the binder (by mass) to minimize its influence on cement hydration, due to the interaction between water and PET. The PET-containing samples showed the presence of calcium aluminate hydrates which were absent in the neat paste sample.