The aging of concrete structures is a threat to public safety; therefore, maintenance and repair of these structures have been highly emphasized. However, regular inspections to detect concrete cracks that rely on operators lack objectivity and consume a lot of time. To overcome this limitation, high-resolution image processing and deep learning have been adopted. Nevertheless, cracks on structure surfaces are still challenging to detect owing to the variety of shapes of cracks and the dependence of recognition performance on image conditions. Herein, we propose a new concrete crack detection method that applies the semantic segmentation technique using 1196 concrete crack images and labeled images produced in this study. A new segmentation algorithm is developed using a hierarchical convolutional neural network to improve speed, and a multi-loss update method is proposed to improve accuracy. The performance of the proposed network is evaluated in terms of accuracy and speed. The results show that the proposed network produces a 2.165% increase in the intersection over union of crack, 65.90% decrease in the average inference time, and 99.90% decrease in the number of parameters compared with the best accuracy results using existing segmentation networks. It is expected that the application of this improved crack detection method will result in faster and more accurate crack detection and, consequently, improved safety, thereby making it suitable for application in structure safety inspections.

This study investigates the geomechanical, hydraulic and thermal characteristics of natural sandy sediments collected during the Ulleung Basin gas hydrate expedition 2, East Sea, offshore Korea. The studied sediment formation is considered as a potential target reservoir for natural gas production. The sediments contained silt, clay and sand fractions of 21%, 1.3% and 77.7%, respectively, as well as diatomaceous minerals with internal pores. The peak friction angle and critical state (or residual state) friction angle under drained conditions were~26 • and~22 • , respectively. There was minimal or no apparent cohesion intercept. Stress-and strain-dependent elastic moduli, such as tangential modulus and secant modulus, were identified. The sediment stiffness increased with increasing confining stress, but degraded with increasing strain regime. Variations in water permeability with water saturation were obtained by fitting experimental matric suction-water saturation data to the Maulem-van Genuchen model. A significant reduction in thermal conductivity (from~1.4-1.6 to~0.5-0.7 W·m −1 ·K −1 ) was observed when water saturation decreased from 100% to~10%-20%. In addition, the electrical resistance increased quasi-linearly with decreasing water saturation. The geomechanical, hydraulic and thermal properties of the hydrate-free sediments reported herein can be used as the baseline when predicting properties and behavior of the sediments containing hydrates, and when the hydrates dissociate during gas production. The variations in thermal and hydraulic properties with changing water and gas saturation can be used to assess gas production rates from hydrate-bearing deposits. In addition, while depressurization of hydrate-bearing sediments inevitably causes deformation of sediments under drained conditions, the obtained strength and stiffness properties and stress-strain responses of the sedimentary formation under drained loading conditions can be effectively used to assess sediment responses to depressurization to ensure safe gas production operations in this potential target reservoir.

Abrasive waterjet cutting technology has come back into use in the field of rock excavation (such as for tunneling) due to the need for precision construction with low vibration. Because the abrasive particles play an important role in efficient erosion during the cutting process, the abrasive characteristics strongly affect the rock cutting performance. In this study, rock cutting tests were performed with five different coarse (40 mesh) garnets to explore the effect of the abrasive feed rate, physical properties, and particle size distribution on rock cutting performance. In addition, garnet particle disintegration was investigated with garnet characteristics for the abrasive waterjet. The test results indicate that the particle size distribution, garnet purity, specific gravity, and hardness are the most important parameters for rock cutting performance. This study offers better understanding of coarse garnet performance and efficiency according to the garnet characteristics. This should provide assistance in selection of the garnet needed to achieve the desired performance for hard rock cutting.

Abrasive waterjets are being increasingly used in civil engineering for rock and concrete cutting, particularly for the demolition or repair of old structures. The energy of an abrasive waterjet is primarily provided by the accelerated abrasive. The momentum transfer during mixing and acceleration determines the abrasive velocity, which affects the cutting performance. Meanwhile, the geometry of the focus at which mixing occurs influences the momentum transfer efficiency. In this study, the effects of the focus geometry on the optimum abrasive flow rate (AFR) and momentum transfer characteristics in hard rock cutting were investigated. Experiments were conducted using granite specimens to test the AFR under different focus geometry conditions such as diameter and length. The results show that the focus geometry significantly affects the maximum cutting depth and optimum AFR. The maximum cutting energy was analyzed based on the cutting efficiency of a single abrasive particle. In addition, the momentum transfer parameter (MTP) was evaluated from the empirical relationship between the maximum energy and the cutting depth for granitic rocks. Accordingly, a model for estimating the MTP based on the AFR was developed. It is expected that the results of this study can be employed for the optimization of waterjet rock cutting.

This study applies a comprehensive surrogate-based optimization techniques to optimize the performance of polymer electrolyte membrane fuel cells (PEMFCs). Parametric cases considering four variables are defined using latin hypercube sampling. Training and test data are generated using a multidimensional, two-phase PEMFC simulation model. Response surface approximation, radial basis neural network, and kriging surrogates are employed to construct objective functions for the PEMFC performance. There accuracies are tested and compared using root mean square error and adjusted R-square. Surrogates linked with optimization algorithms, i.e., genetic algorithm and particle swarm optimization are used to determine the optimal design points. Comparative study of these surrogates reveals that the kriging model outperforms the other models in terms of prediction capability. Furthermore, the PEMFC model simulations at the optimal design points demonstrate that performance improvements of around 56–69 mV at 2.0 A/cm2 are achieved with the optimal design compared to typical PEMFC design conditions.