Tunnel and Underground Space ISSN:1225-1275(Print) 2287-1748(Online) 터널과 지하공간

The Incheon Metro Line 1 Geomdan Extension Phase 1 is the first project in South Korea where both a roadheader and TBM (Tunnel Boring Machine) are being used together. The shield TBM tunnel section is 1,057 m long, and is mostly composed of rock, including the section beneath the Airport Railroad and the Gyeongin Ara Waterway. A 7.8 m earth pressure balance shield TBM was used for tunnel excavation. The average monthly advance rate for both the North and South tracks is 239.1 m, with a maximum monthly advance rate of 334.5 m. This technical article comprehensively evaluates the productivity of the shield TBM, focusing on the TBM excavation performance. Above all, it aims to provide useful reference material for the successful execution of shield TBM tunnel construction.

In this study, a comprehensive management plan was proposed to ensure the stability of the existing subway when constructing ground structures adjacent to the subway. In the first step, the measurement inspection cycle is selected through proximity evaluation, in the second step, the stability of existing subway and station structures such as displacement and stress is reviewed through 3D numerical analysis considering the construction stage and groundwater influence, and in the third stage, the safety of train operation was reviewed by examining the track stability, and based on the numerical analysis results in the fourth stage, the displacement concentration section was selected as an intensive management section and it was proposed that intensive measurement management be performed.

There is so much research effort for developing and implementing deep learning-based surveillance systems to manage health and safety issues in construction sites. Especially, the development of deep learning-based object detection in various environmental changes has been progressing because those affect decreasing searching performance of the model. Among the various environmental variables, the accuracy of the object detection model is significantly dropped under low illuminance, and consistent object detection accuracy cannot be secured even the model is trained using low-light images. Accordingly, there is a need of low-light enhancement to keep the performance under low illuminance. Therefore, this paper conducts a comparative study of various deep learning-based low-light image enhancement models (GLADNet, KinD, LLFlow, Zero-DCE) using the acquired construction site image data. The low-light enhanced image was visually verified, and it was quantitatively analyzed by adopting image quality evaluation metrics such as PSNR, SSIM, Delta-E. As a result of the experiment, the low-light image enhancement performance of GLADNet showed excellent results in quantitative and qualitative evaluation, and it was analyzed to be suitable as a low-light image enhancement model. If the low-light image enhancement technique is applied as an image preprocessing to the deep learning-based object detection model in the future, it is expected to secure consistent object detection performance in a low-light environment.

Recently, many studies have been conducted for safety management in construction sites by incorporating computer vision. Anchor box parameters are used in state-of-the-art deep learning-based object detection and segmentation, and the optimized parameters are critical in the training process to ensure consistent accuracy. Those parameters are generally tuned by fixing the shape and size by the user's heuristic method, and a single parameter controls the training rate in the model. However, the anchor box parameters are sensitive depending on the type of object and the size of the object, and as the number of training data increases. There is a limit to reflecting all the characteristics of the training data with a single parameter. Therefore, this paper suggests a method of applying multiple parameters optimized through data split to solve the above-mentioned problem. Criteria for efficiently segmenting integrated training data according to object size, number of objects, and shape of objects were established, and the effectiveness of the proposed data split method was verified through a comparative study of conventional scheme and proposed methods.

In the field of high-level radioactive waste disposal targeting deep rock environments, hydraulic characteristic information serves as the most important key factor in selecting relevant disposal sites, detailed design of disposal facilities, derivation of optimal construction plans, and safety evaluation during operation. Since various rock types are mixed and distributed in a small area in Korea, it is important to conduct preliminary work to analyze the hydrogeological characteristics of rock aquifers for various rock types and compile the resulting data into a database. In this paper, we obtained hydraulic conductivity data, which is the most representative field hydraulic characteristic of a high-depth volcanic bedrock aquifer, and also analyzed and evaluated the field data. To acquire field data, we used a high- performance hydraulic testing system developed in-house and applied standardized test methods and investigation procedures. In the process of hydraulic characteristic data analysis, hydraulic conductivity values ​​were obtained for each depth, and the pattern of groundwater flow through permeable rock joints located in the test section was also evaluated. It is expected that the series of data acquisition methods, procedures, and analysis results proposed in this report can be used to build a database of hydraulic characteristics data for high-depth rock aquifers in Korea. In addition, it is expected that it will play a role in improving technical know-how to be applied to research on hydraulic characteristic according to various bedrock types in the future.