• Journal of the Korean Geotechnical Society
• /
• v.35 no.9
• /
• pp.29-36
• /
• 2019

Axial compressive failure loads ($P_n$) of diameter 500 mm and diameter 600 mm A type PHC pile were calculated as 7.7 MN and 10.6 MN, respectively. In the static pile load tests, the maximum axial compressive loads of the above 2 kinds of A type pile were measured as 6.9 MN and 8.8 MN respectively, therefore these measured maximum loads were at the level of 90% and 83% of $P_n$ respectively. Long-term allowable axial compressive loads ($P_a$) of the above 2 kinds of A type pile were 1.7 MN and 2.3 MN respectively. From the bi-directional pile load test data on the prebored PHC piles, it was confirmed that the allowable axial compressive bearing resistance was estimated as 131% of the long-term allowable compressive load of the PHC pile and showed higher than the allowable bearing capacity calculated by the current design method. Therefore, it has been verified that the PHC pile can be used up to the maximum long-term allowable compressive load, and it is suggested that the ultimate pile capacity formula used in the current design for prebored PHC piles should be improved to accommodate the actual capacity.

• Journal of the Korean Geotechnical Society
• /
• v.35 no.9
• /
• pp.15-28
• /
• 2019

Long-term allowable compressive Loads of PHC piles were analyzed based on qualification tests results by 17 small and medium PHC pile producing companies and product specifications by 6 major and 17 small and medium PHC pile producing companies. At the present stage, an average long-term allowable compressive load of PHC pile was designed at 70% level from current design data, and safety factor of 4.0 was applied to long-term allowable compressive loads of PHC pile despite of its excellent quality. Most quality standards of PHC pile are specified at KS F 4306. But compressive strength test method of spun concrete is specified at KS F 2454. As a result of analyzing quality test data supplied by each manufacturer, all quality test results showed higher performances than standard values. Therefore, it was considered that the capacity of PHC pile can be used up to the maximum allowable compressive load of PHC pile when PHC pile is designed.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.45 no.1
• /
• pp.95-103
• /
• 2019

Lips have a defect in maintenance of moisture due to their thin layer. As aging progresses, lips lose volume and redness, and become wrinkled. Fat grafting and filler surgery have been used to achieve attractive lips, but little research has been reported to develop better materials to replace the present methods. Recently, a study suggests that the increase of adipocyte number can be enhancing the expansion endogenous fat. In previous study, we identified that the efficacy of Magnolia officinalis bark extract (MOBE) was effective on the induction of adipogenic differentiation. In this study, we confirmed that MOBE enhanced the differentiation of human adipose-derived stem cells on the fat mimic 3D structure built by 3D bioprinting method From further experiments in human, we established a method to quantify the severity of lip wrinkle by measurement of standard deviation of gray value using Image J software. Finally, we found that topical treatment with 1% MOBE formulated lip balm significantly improved the lip wrinkle after using for 12 weeks. In conclusion, these findings suggest that MOBE has great potential, as a cosmetic ingredient, to reduce the lip wrinkle through the effect of promoting adipogenic differentiation.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.3
• /
• pp.323-345
• /
• 2019

It is essential to predict ground conditions ahead of a tunnel face in order to successfully excavate tunnels using a shield TBM. This study proposes a forward prediction method for a mixed soil ground and/or a ground containing core stones by using electrical resistivity and induced polarization exploration. Soil conditioning in EPB shield TBM is dependent upon the composition of mixed soils; a special care need to be taken when excavating the core-stoned soil ground using TBM. The resistivity and chargeability are assumed to be measured with four electrodes at the tunnel face, whenever the excavation is stopped to assemble one ring of a segment lining. Firstly, the mixed ground consisting of weathered granite soil, sand, and clay was modeled in laboratory-scale experiments. Experimental results show that the measured electrical resistivity considerably coincides with the analytical solution. On the other hand, the induced polarization has either same or opposite trend with the measured resistivity depending on the mixed ground conditions. Based on these experimental results, a method to predict the mixed soil ground that can be used during TBM tunnel driving is suggested. Secondly, tunnel excavation from a homogeneous ground to a ground containing core stones was modeled in laboratory scale; the irregularity of the core stones contained in the soil layer was modeled through random number generation scheme. Experimental results show that as the TBM approaches the ground that contains core stones, the electrical resistivity increases and the induced polarization fluctuates.

• Composites Research
• /
• v.33 no.6
• /
• pp.371-376
• /
• 2020

The adhesive strength can be improved through surface treatment. The most common method is to improve physical bonding by varying the surface conditions. This study presents the effect of laser surface treatment on the adhesive strength of CFRP. The surface roughness was patterned using a 1064 nm laser. The effects of the number of laser shots and the direction and length of the pattern on the adhesion of the CFRP/CFRP single joint were investigated through tensile tests. Tests according to ASTM D5868 were performed, and the bonding mechanism was determined by analyzing the damaged surface after a fracture. The optimized number of the laser shots and the optimized depth of the roughness should be required to increase the bonding strength on the CFRP surface. When considering the shear stress in the tensile direction, the roughness pattern in the direction of 45° that increases the length of the fracture path in the adhesive layer resulted in an increase of the adhesive strength. The surface treatment of the bonding surface using a laser is a suitable method to acquire a mechanical bonding mechanism and improve the bonding strength of the CFRP bonding joint. The study on the optimized laser process parameters is required for utilizing the benefits of laser surface processing.

• Journal of the Korean GEO-environmental Society
• /
• v.12 no.10
• /
• pp.39-49
• /
• 2011

Recently PBD method, one of acceleration of consolidation methods is used in the soft ground to shorten consolidation time for fast settlement during construction. It is economical and easy to work. Discharge capacity of PBD is sensitive in proportion to thickness of soft ground layer, and drainage of PBD declines due to disturbance effect in surrounding ground by mandrel used for vertical drainage setting and setting machines and type. Also, deviation of discharge capacity is large according to ground condition, construction condition and soil properties. In addition, when embankment loading is not conducted instantly after PBD setting due to rain or lack of embankment material supply, it causes leaving period problems. But cause and analysis of those problems for discharge capacity is lack. So, in this test, ground improvement and discharge capacity is investigated by implementing composite discharge capacity test for analysis of an effect factor of PBD discharge capacity with leaving period. After fixing the vertical drain on a cylindrical cylinder, put churned sample into the cylinder. Then leave 0day, 30day, 60day and 90day. And then, load following the loading step of 30, 70 and 120kPa using a pressure device. As a result, the longer leaving period, discharge capacity is reduced. It is caused by a decrease of discharge area caused by creep transformation moisture absorption of PBD filter after long leaving period.

• Knowledge Management Research
• /
• v.22 no.4
• /
• pp.27-40
• /
• 2021

Click-Through Rate (CTR) prediction is a key function that determines the ranking of candidate items in the recommendation system and recommends high-ranking items to reduce customer information overload and achieve profit maximization through sales promotion. The fields of natural language processing and image classification are achieving remarkable growth through the use of deep neural networks. Recently, a transformer model based on an attention mechanism, differentiated from the mainstream models in the fields of natural language processing and image classification, has been proposed to achieve state-of-the-art in this field. In this study, we present a method for improving the performance of a transformer model for CTR prediction. In order to analyze the effect of discrete and categorical CTR data characteristics different from natural language and image data on performance, experiments on embedding regularization and transformer normalization are performed. According to the experimental results, it was confirmed that the prediction performance of the transformer was significantly improved when the L2 generalization was applied in the embedding process for CTR data input processing and when batch normalization was applied instead of layer normalization, which is the default regularization method, to the transformer model.

• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.429-440
• /
• 2021

With the recent development of deep convolutional neural network learning, deep learning techniques applied to single image super-resolution are showing good results. One of the existing deep learning-based super-resolution techniques is RDN(Residual Dense Network), in which the initial feature information is transmitted to the last layer using residual dense blocks, and subsequent layers are restored using input information of previous layers. However, if all hierarchical features are connected and learned and a large number of residual dense blocks are stacked, despite good performance, a large number of parameters and huge computational load are needed, so it takes a lot of time to learn a network and a slow processing speed, and it is not applicable to a mobile system. In this paper, we use the residual dense structure, which is a continuous memory structure that reuses previous information, and the residual dense channel attention block using the channel attention method that determines the importance according to the feature map of the image. We propose a method that can increase the depth to obtain a large receptive field and maintain a concise model at the same time. As a result of the experiment, the proposed network obtained PSNR as low as 0.205dB on average at 4× magnification compared to RDN, but about 1.8 times faster processing speed, about 10 times less number of parameters and about 1.74 times less computation.

• The Journal of Engineering Geology
• /
• v.31 no.4
• /
• pp.589-601
• /
• 2021

Artificial recharge technology is a method for solving problems such as groundwater level drop and ground subsidence caused by groundwater withdrawal. This study investigated the applicability of using the hydraulic conductivity of an aquifer to predict injection test results for aquifer restoration. Pumping and injection tests were performed under the same conditions as those for the artificial injection facility located in Icheon, Gyeonggi-do. The hydraulic conductivity of the aquifer, which plays a decisive role in restoring the groundwater level, was derived from the pumping test. A numerical model of a simplified on-site aquifer was constructed, and a transient analysis was applied with the same conditions as the pumping test. The correlation between the measured and the resulting model values is strong (R² = 0.78). The injection test was performed in a sedimentary layer composed of silt sand and clay sand. From the results of the injection test, an empirical formula was derived using Theim's formula, which is a common well analysis solution to determine the parameters of the aquifer from time-level data. The model values from the empirical formula have a high degree of correlation (R² = 0.99) with measured values. Under specific conditions, for areas where it is difficult to conduct an injection test, the formula from this study, which relies on the hydraulic conductivity of the aquifer determined through the pumping test, may be used to predict reliable injection rates for groundwater restoration.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.33-42
• /
• 2022

In this paper, we proposed a deep learning model based on multi-scale dilated convolution feature fusion for the segmentation of black ice region in road image to send black ice warning to drivers in real time. In the proposed multi-scale dilated convolution feature fusion network, different dilated ratio convolutions are connected in parallel in the encoder blocks, and different dilated ratios are used in different resolution feature maps, and multi-layer feature information are fused together. The multi-scale dilated convolution feature fusion improves the performance by diversifying and expending the receptive field of the network and by preserving detailed space information and enhancing the effectiveness of diated convolutions. The performance of the proposed network model was gradually improved with the increase of the number of dilated convolution branch. The mIoU value of the proposed method is 96.46%, which was higher than the existing networks such as U-Net, FCN, PSPNet, ENet, LinkNet. The parameter was 1,858K, which was 6 times smaller than the existing LinkNet model. From the experimental results of Jetson Nano, the FPS of the proposed method was 3.63, which can realize segmentation of black ice field in real time.