• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.3
• /
• pp.419-432
• /
• 2019

Most of deep learning model training was proceeded by supervised learning, which is to train labeling data composed by inputs and corresponding outputs. Labeling data was directly generated manually, so labeling accuracy of data is relatively high. However, it requires heavy efforts in securing data because of cost and time. Additionally, the main goal of supervised learning is to improve detection performance for 'True Positive' data but not to reduce occurrence of 'False Positive' data. In this paper, the occurrence of unpredictable 'False Positive' appears by trained modes with labeling data and 'True Positive' data in monitoring of deep learning-based CCTV accident detection system, which is under operation at a tunnel monitoring center. Those types of 'False Positive' to 'fire' or 'person' objects were frequently taking place for lights of working vehicle, reflecting sunlight at tunnel entrance, long black feature which occurs to the part of lane or car, etc. To solve this problem, a deep learning model was developed by simultaneously training the 'False Positive' data generated in the field and the labeling data. As a result, in comparison with the model that was trained only by the existing labeling data, the re-inference performance with respect to the labeling data was improved. In addition, re-inference of the 'False Positive' data shows that the number of 'False Positive' for the persons were more reduced in case of training model including many 'False Positive' data. By training of the 'False Positive' data, the capability of field application of the deep learning model was improved automatically.

• Nuclear Engineering and Technology
• /
• v.54 no.12
• /
• pp.4474-4480
• /
• 2022

Deep geological disposal using a multibarrier system is a promising solution for treating high-level radioactive (HLRW) waste. The HLRW canister represents the first barrier for the migration of radionuclides into the biosphere, therefore, the corrosion behavior of canister materials is of significance. In this study, the electrochemical behaviors of SS316L, Ti-Gr.2, Alloy 22, and Cu in naturally aerated KAERI underground research tunnel (KURT) groundwater solutions were examined. The corrosion potential, current, and impedance spectra of the test materials were recorded using electrochemical methods. According to polarization and impedance measurements, Cu exhibits relatively higher corrosion rates and a lower corrosion resistance ability than those exhibited by the other materials in the given groundwater condition. In the anodic dissolution tests, SS316L exposed to the groundwater solution exhibited the most uniform corrosion, as indicated by its surface roughness. This phenomenon could be attributed to the extremely low concentration of chloride ions in KURT groundwater.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.23 no.6
• /
• pp.549-558
• /
• 2021

In order to keep infrastructures such as tunnels and underground facilities safe, cracks of concrete lining in tunnel should be detected by regular inspections. Since regular inspections are accomplished through manual efforts using maintenance lift vehicles, it brings about traffic jam, exposes works to dangerous circumstances, and deteriorates consistency of crack inspection data. This study aims to provide methodology to automatically extract cracks from tunnel concrete lining images generated by the existing tunnel image acquisition system. Specifically, we train a deep learning based semantic segmentation model with open dataset, and evaluate its performance with the dataset from the existing tunnel image acquisition system. In particular, we compare the model performance in case of using all of a public dataset, subset of the public dataset which are related to tunnel surfaces, and the tunnel-related subset with negative examples. As a result, the model trained using the tunnel-related subset with negative examples reached the best performance. In the future, we expect that this research can be used for planning efficient model training strategy for crack detection.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.25 no.4
• /
• pp.65-74
• /
• 2021

In order to efficiently inspect rapidly increasing old tunnels in many well-developed countries, many inspection methodologies have been proposed using imaging equipment and image processing. However, most of the existing methodologies evaluated their performance on a clean concrete surface with a limited area where other objects do not exist. Therefore, this paper proposes a 6-step framework for tunnel crack detection deep learning model development. The proposed method is mainly based on negative sample (non-crack object) training and Cascade Mask R-CNN. The proposed framework consists of six steps: searching for cracks in images captured from real tunnels, labeling cracks in pixel level, training a deep learning model, collecting non-crack objects, retraining the deep learning model with the collected non-crack objects, and constructing final training dataset. To implement the proposed framework, Cascade Mask R-CNN, an instance segmentation model, was trained with 1561 general crack images and 206 non-crack images. In order to examine the applicability of the trained model to the real-world tunnel crack detection, field testing is conducted on tunnel spans with a length of about 200m where electric wires and lights are prevalent. In the experimental result, the trained model showed 99% precision and 92% recall, which shows the excellent field applicability of the proposed framework.

• Tunnel and Underground Space
• /
• v.17 no.1 s.66
• /
• pp.26-31
• /
• 2007

The 11 nuclear power plants have been taking charge of more than 40% of the total electrical power development in Korea. In addition to the existing nuclear power plants at Gori, Wolsung, Youngkwang, etc., the 12 nuclear power plants are expected to be newly established until 2006. So, the 23 nuclear power plants will produce the electric power as much as more than 50% of the national gross production. However the nuclear power plants are inevitably generating the detrimental atomic wastes. Therefore the disposal techniques for the nuclear wastes should be ensured considering a very high safety factor. According to the basic researches in KAERI, the underground disposal repositories are reported to be most favorable for Korea. The KBS-3 disposal system has been strongly suggested by KAERI and this system has a deep tunnel with several disposal boreholes in tunnel floor. The nuclear wastes, which are sealed tightly in a canister, will be disposed in these boreholes. Considering the disposal tunnel in a great depth, the in-situ stress regimes will affect severely the tunnel stability. Consequently the effect of the in-situ stresses on the disposal tunnel and the role of the in-situ stresses in tunnel stability analysis are examined by the numerical studies.

• Tunnel and Underground Space
• /
• v.12 no.3
• /
• pp.198-209
• /
• 2002

To assess deep geological environment for the research and development of hish-level radioactive waste disposal, six boreholes of 3" in diameter were installed in two granitic areas. An areal extent of the rock block scale in the study sites was estimated by the lineament analysis from satellite images and shaded relief maps. The characterization of fracture system developed in rock block scale was carried out based on the acoustic televiewer logging in deep boreholes. In the Yuseong site, the granite rock mass was divided into the upper and lower zones at around -160m based on the probabilistic distribution characteristics of the geometric parameters such as orientation, fracture frequency, spacing and aperture size. Since the groundwater flow is dependent on the fracture system in a fractured rock mass, the correlation of the fracture frequency and cumulative aperture size to the hydraulic conductivity was also discussed.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.25 no.6
• /
• pp.555-567
• /
• 2023

The cracks in the tunnel are currently determined through visual inspections conducted by inspectors based on images acquired using tunnel imaging acquisition systems. This labor-intensive approach, relying on inspectors, has inherent limitations as it is subject to their subjective judgments. Recently research efforts have actively explored the use of deep learning to automatically detect tunnel cracks. However, most studies utilize public datasets or lack sufficient objectivity in the analysis process, making it challenging to apply them effectively in practical operations. In this study, we selected test datasets consisting of images in the same format as those obtained from the actual inspection system to perform an objective evaluation of deep learning models. Additionally, we introduced ensemble techniques to complement the strengths and weaknesses of the deep learning models, thereby improving the accuracy of crack detection. As a result, we achieved high recall rates of 80%, 88%, and 89% for cracks with sizes of 0.2 mm, 0.3 mm, and 0.5 mm, respectively, in the test images. In addition, the crack detection result of deep learning included numerous cracks that the inspector could not find. if cracks are detected with sufficient accuracy in a more objective evaluation by selecting images from other tunnels that were not used in this study, it is judged that deep learning will be able to be introduced to facility safety inspection.

• Tunnel and Underground Space
• /
• v.28 no.6
• /
• pp.620-634
• /
• 2018

The NTR(New Tubular Roof) method was used to secure the stability of the tunnel and minimize the subsidence of the road. The tunnel was constructed at about 7.5 meters deep below the highway. with a width of about 21 meters. Following the NTR method, 13 steel pipes with a diameter of 2.3 meters were digged and pushed in longitudinally along the tunnel profile and cut out sides of pipes to connect to adjacent pipes, then filled the inside of pipes and the connected space between pipes with concrete to complete the lining of the tunnel to be excavated. As the steel pipes were digged in sequentially, the area of relaxation was connected to each other and behaves like a gradually widening tunnel. When the steel pipes were digged in to the widest points of the tunnel, the settlement rate of the road surface was increasing to the maximum as 2.2 mm and the total settlement until the lining construction was approximately 7.7 mm. After that, by excavating a tunnel inside the pre-installed lining, an additional settlement of about 4.3 mm was occurred, resulting in the total settlement of about 11.8 mm after completing of tunnel construction.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.13 no.4
• /
• pp.291-304
• /
• 2011

Recently concern for subsea tunnels is increasing, The effect of high water pressure can not be ignored in the case of a deep subsea tunnel. Reinforcement like grouting is necessary for the stability of such a subsea tunnel. In this study, therefore, it was investigated how the water barrier and shear strength increment resulted from grouting had an effect on the stability of a subsea tunnel. To this end, two-dimensional hydromechanical coupled analyses were performed for a sensitivity analysis in terms of different range, permeability coefficient, and cohesion of grouting reinforcement for the rock classes I, III, and V with respect to RMR system. The mutual relationship between strength increment and water pressure increased by barrier effect due to grouting was investigated by analyzing the numerical results.

• Journal of the Korean Geotechnical Society
• /
• v.19 no.6
• /
• pp.39-47
• /
• 2003

Rockbursts are mainly caused by a sudden release or the stored strain energy in the rock mass. They have been the major hazard in deep hard rock mines but rarely occur in tunnels. Due to the short history and limited information on rockbursts, the topic has rarely been studied in Korea. Some cases of rockbursts, however, have been reported during construction of a mountain tunnel for waterway. This study focuses on analyzing data on rockbursts obtained from a TBM (Tunnel Boring Machine) tunnel and suggests methods for a comprehensive understanding on rockbursts. From the analysis of the field data of rockbursts, it was found that most rockbursts mainly occurred at the section between the tunnel face and the TBM operating room, and the rock bursting phenomena lasted up to 20 days after excavation in certain areas. The data also show that the bursting spots are located all around the tunnel surface including the face, the wall, and the roof, The maximum size of bursting spots is usually less than 100cm. This study also suggests new scale systems of brittleness and uniaxial compressive strength to evaluate the possible tendency for a rockburst. These systems are scaled based on the scale system of strain energy density. In addition, with these scale systems, this research shows that there are potentially higher tendencies for rockbursts in this specific tunnel. Moreover this research suggests that properties of rock and rock mass, RMR (Rock Mass Rating) value, tunneling method, excavating speed, and depth of tunnel have a strong correlation with rockbursts.