• Journal of Broadcast Engineering
• /
• 제28권2호
• /
• pp.230-237
• /
• 2023

Self-driving is a key technology of the fourth industry and can be applied to various places such as cars, drones, cars, and robots. Among them, localiztion is one of the key technologies for implementing autonomous driving as a technology that identifies the location of objects or users using GPS, sensors, and maps. Locilization can be made using GPS or LIDAR, but it is very expensive and heavy equipment must be mounted, and precise location estimation is difficult for places with radio interference such as underground or tunnels. In this paper, to compensate for this, we proposes an image retrieval using attention module and image segmentation maps using color images acquired with low-cost vision cameras as an input.

• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.745-755
• /
• 2023

The wire rope is an indispensable production machinery in coal mines. It is the main force-bearing equipment of the underground traction system. Accurate detection of wire rope defects and positions exerts an exceedingly crucial role in safe production. The existing defect detection solutions exhibit some deficiencies pertaining to the flexibility, accuracy and real-time performance of wire rope defect detection. To solve the aforementioned problems, this study utilizes the camera to sample the wire rope before the well entry, and proposes an object based on YOLOv5. The surface small-defect detection model realizes the accurate detection of small defects outside the wire rope. The transfer learning method is also introduced to enhance the model accuracy of small sample training. Herein, the enhanced YOLOv5 algorithm effectively enhances the accuracy of target detection and solves the defect detection problem of wire rope utilized in mine, and somewhat avoids accidents occasioned by wire rope damage. After a large number of experiments, it is revealed that in the task of wire rope defect detection, the average correctness rate and the average accuracy rate of the model are significantly enhanced with those before the modification, and that the detection speed can be maintained at a real-time level.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• 제22권2호
• /
• pp.211-217
• /
• 2024

The growing significance of sustainable energy technologies underscores the need for safe and efficient management of spent nuclear fuels (SNFs), particularly via deep geological disposal (DGD). DGD involves the long-term isolation of SNFs from the biosphere to ensure public safety and environmental protection, necessitating materials with high corrosion resistance for DGD canisters. This study investigated the feasibility of a Cu-Ni film, fabricated via additive manufacturing (AM), as a corrosion-resistant layer for DGD canister applications. A wire-fed AM technique was used to deposit a millimeter-scale Cu-Ni film onto a carbon steel (CS) substrate. Electrochemical analyses were conducted using aerated groundwater from the KAERI underground research tunnel (KURT) as an electrolyte with an NaCl additive to characterize the oxic corrosion behavior of the Cu-Ni film. The results demonstrated that the AM-fabricated Cu-Ni film exhibited enhanced corrosion resistance (manifested as lower corrosion current density and formation of a dense passive layer) in an NaCl-supplemented groundwater solution. Extensive investigations are necessary to elucidate microstructural performance, mechanical properties, and corrosion resistance in the presence of various corroding agents to simplify the implementation of this technology for DGD canisters.

• IEMEK Journal of Embedded Systems and Applications
• /
• 제19권4호
• /
• pp.185-193
• /
• 2024

Visual-inertial odometry (VIO) is a method that leverages sensor data from a camera and an inertial measurement unit (IMU) for state estimation. Whereas conventional VIO has limited capability to estimate scale of translation, the performance of recent approaches has been improved by utilizing depth maps obtained from RGB-D camera, especially in indoor environments. However, the depth map obtained from the RGB-D camera tends to rapidly lose accuracy as the distance increases, and therefore, it is required to develop alternative method to improve the VIO performance in wide environments. In this paper, we argue that leveraging depth map estimated from a deep neural network has benefits to state estimation. To improve the reliability of depth information utilized in VIO algorithm, we propose a kernel-based sampling strategy to filter out depth values with low confidence. The proposed method aims to improve the robustness and accuracy of VIO algorithms by selectively utilizing reliable values of estimated depth maps. Experiments were conducted on real-world custom dataset acquired from underground parking lot environments. Experimental results demonstrate that the proposed method is effective to improve the performance of VIO, exhibiting potential for the use of depth estimation network for state estimation.

• Economic and Environmental Geology
• /
• 제52권2호
• /
• pp.189-197
• /
• 2019

HLW (High Level Radioactive Waste) is one of the problems that must be solved in the countries that implement nuclear power generation. Most countries that are concerned about HLW treatment are considering complete isolation from human society by disposing them deep underground. For perfect isolation, understanding the characteristics of underground rocks is very important. In particular, understanding the characteristics of discontinuity as a path way is one of the first things in order to predict the movement of exposed nuclear species to the surface. In this study, we used 500m underground core samples obtained from granite and gneiss area. The purpose of this study is to understand the characteristics of the discontinuities in each rock type and to analyze the properties of the joints in the underground relative to the surrounding environment. For this purpose, the types of discontinuities were classified and the distribution of each discontinuity were analyzed through visual analysis of the each sample obtained at 500m underground. This study can be used as a basic data for understanding the properties of discontinuities in the rock of the survey area and it can be also used as an important data for understanding the distribution characteristics of discontinuities according to the rock types.

• Journal of the Korea institute for structural maintenance and inspection
• /
• 제23권7호
• /
• pp.49-57
• /
• 2019

Recently, the use of deep deck plate has been increased in various structures, such as underground parking lots, logistics warehouses, because it can reduce construction periods and labor costs. In this study, a newly developed Double Deck (D-deck) plate which can leads to save story heights has been introduced, and experimental tests on a total of five D-deck plates under construction loads have been carried out to investigate their structural performance at construction stage. The loads were applied by sands and concrete to simulate the actual distributed loading conditions, and the vertical deflection of D-Deck and the horizontal deformation of web were measured and analyzed in detail. As a result, it was confirmed that all the D-decks showed very small vertical deflection of less than 5.34 mm under construction loads, which satisfies the maximum deflection limit of L / 180. In addition, the D-Deck plate was found to have a sufficient rigidity to resist construction loads in a stable manner.

• Tunnel and Underground Space
• /
• 제29권6호
• /
• pp.508-522
• /
• 2019

Because of the slope failure, not only property damage but also human damage can occur, slope stability analysis should be conducted to predict and reinforce of the slope. This paper, defines the ground areas that can be characterized in terms of slope failure such as Rockmass jointset, Rockmass fault, Soil, Leakage water and Crush zone in sloped images. As a result, it was shown that the deep learning instance segmentation network can be used to recognize and automatically segment the precise shape of the ground region with different characteristics shown in the image. It showed the possibility of supporting the slope mapping work and automatically calculating the ground characteristics information of slopes necessary for decision making such as slope reinforcement.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 한국정밀공학회 2005년도 추계학술대회 논문집
• /
• pp.470-473
• /
• 2005

Spent nuclear fuels are regarded as a high level radioactive waste and they will be disposed in a deep geological repository. To maintain the safety of the repository for hundreds of thousands of years, the spent fuels are encapsulated in a disposal canister and the canister containing spent fuels should have the structural integrity and the corrosion resistance below the several hundreds meters from the ground surface. In this study, the concept of the spent fuel encapsulation process and the process equipment fur deep geological disposal were established. To do this, the design requirements, such as the functions and the spent fuel accumulations, were reviewed. Also, the design principles and the bases were established. Based on the requirements and the bases, the encapsulation process and the equipment from spent fuel receiving process to transferring canister into the underground repository including hot cell processes was established. The established concept of the spent fuel encapsulation process and the process equipment will be improved continuously with the future studies. And this concept can be effectively used in implementing the reference repository system of our own case.

• Geomechanics and Engineering
• /
• 제19권4호
• /
• pp.315-327
• /
• 2019

Deep excavations for development of subway systems in metropolitan regions surrounded by adjacent buildings is an important geotechnical problem, especialy in Tabriz city, where is mostly composed of young alluvial soils and weak marly layers. This study analyzes the wall displacement and ground surface settlement due to deep excavation in the Tabriz marls using two dimensional finite element method. The excavation of the station L2-S17 was selected as a case study for the modelling. The excavation is supported by the concrete diaphragm wall and one row of steel struts. The analyses investigate the effects of wall stiffness and excavation width on the excavation-induced deformations. The geotechnical parameters were selected based on the results of field and laboratory tests. The results indicate that the wall deflection and ground surface settlement increase with increasing excavation depth and width. The change in maximum wall deflection and ground settlement with considerable increase in wall stiffness is marginal, however the lower wall stiffness produces the larger wall and ground displacements. The maximum wall deflections induced by the excavation with a width of 8.2 m are 102.3, 69.4 and 44.3 mm, respectively for flexible, medium and stiff walls. The ratio of maximum ground settlement to maximum lateral wall deflection approaches to 1 with increasing wall stiffness. It was found that the wall stiffness affects the settlement influence zone. An increase in the wall stiffness results in a decrease in the settlements, an extension in the settlement influence zones and occurrence of the maximum settlements at a larger distance from the wall. The maximum of settlement for the excavation with a width of 14.7 m occurred at 6.1, 9.1 and 24.2 m away from the wall, respectively, for flexible, medium and stiff walls.

• Journal of the Korean Geotechnical Society
• /
• 제35권9호
• /
• pp.37-45
• /
• 2019

Quick identification of the condition of tunnel face and optimized determination of support patterns during tunnel excavation in underground construction projects help engineers prevent tunnel collapse and safely excavate tunnels. This study investigates a CNN technique for quick determination of rock quality classification depending on the condition of tunnel face, and presents the procedure for rock quality classification using a deep learning technique and the improved method for accurate prediction. The VGG16 model developed by tens of thousands prestudied images was used for deep learning, and 1,469 tunnel face images were used to classify the five types of rock quality condition. In this study, the prediction accuracy using this technique was up to 83.9%. It is expected that this technique can be used for an error-minimizing rock quality classification system not depending on experienced professionals in rock quality rating.