• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.1-7
• /
• 2023

An artificial intelligence (AI) method based on image deep learning is proposed to predict the entire displacement shape of a structure using the feature of partial displacements. The performance of the method was investigated through a structural test of a steel frame. An image-to-image regression (I2IR) training method was developed based on the U-Net layer for image recognition. In the I2IR method, the U-Net is modified to generate images of entire displacement shapes when images of partial displacement shapes of structures are input to the AI network. Furthermore, the training of displacements combined with the location feature was developed so that nodal displacement values with corresponding nodal coordinates could be used in AI training. The proposed training methods can consider correlations between nodal displacements in 3D space, and the accuracy of displacement predictions is improved compared with artificial neural network training methods. Displacements of the steel frame were predicted during the structural tests using the proposed methods and compared with 3D scanning data of displacement shapes. The results show that the proposed AI prediction properly follows the measured displacements using 3D scanning.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.2
• /
• pp.102-111
• /
• 2023

Automatic Music Transcription (AMT) is a task that detects and recognizes musical note events from a given audio recording. In this paper, we focus on reducing the latency of real-time AMT systems on piano music. Although neural AMT models have been adapted for real-time piano transcription, they suffer from high latency, which hinders their usefulness in interactive scenarios. To tackle this issue, we explore several techniques for reducing the intrinsic latency of a neural network for piano transcription, including reducing window and hop sizes of Fast Fourier Transformation (FFT), modifying convolutional layer's kernel size, and shifting the label in the time-axis to train the model to predict onset earlier. Our experiments demonstrate that combining these approaches can lower latency while maintaining high transcription accuracy. Specifically, our modified model achieved note F1 scores of 92.67 % and 90.51 % with latencies of 96 ms and 64 ms, respectively, compared to the baseline model's note F1 score of 93.43 % with a latency of 160 ms. This methodology has potential for training AMT models for various interactive scenarios, including providing real-time feedback for piano education.

• Journal of the Korean Geotechnical Society
• /
• v.26 no.8
• /
• pp.27-37
• /
• 2010

Downdrag force develops when a pile is driven through a soil layer which will settle more than a pile. There is no obvious criterion for application of the current pile design method considering the negative skin friction. Therefore, in this study, numerical analyses were performed to investigate the behavior of a single pile subjected to negative skin friction and their results were used to determine the applicability of the current design method. Including three different sites in Song-do area and two different cases with friction pile and end bearing pile conditions, total six cases were considered. The load-settlement relationships and the neutral points were estimated for different end bearing conditions and the allowable bearing capacity of piles with negative skin friction was investigated through parametric studies. Based on the results showed that the negative skin friction made a major influence on the settlement of a pile and its stress. However the allowable bearing capacity may not be influenced by the negative skin friction. Compared with the allowable bearing capacity obtained from the ultimate bearing capacity with the safety factor of 3, the current design method with the safety factor of 3 underestimated the allowable bearing capacities regardless of the end bearing conditions. On the other hand, the current design method with the safety factor of 2 yielded reasonable results depending on the end bearing conditions.

• Journal of the Korean Geotechnical Society
• /
• v.24 no.2
• /
• pp.77-86
• /
• 2008

The disturbance zone and measured values are affected by the size of the penetrometer. The local value may be measured by the smaller penetrometer. An ultra small Micro-Cone penetrometer (5mm in outer diameter) is designed and manufactured to characterize soil properties with minimum disturbance during penetration tests. The tip resistance is measured by using stain gauges attached near the Micro-Cone. In addition, the friction sleeve is adopted to effectively remove the skin friction from the tip resistance. Design concern includes the installation of stain gauges, circuits, penetration systems, penetration rate, sampling rate, operating temperature, and calibration. Application tests show that the clay interface, and the soil layers consisting of clay and sand are clearly detected by the Micro-Cone. Furthermore, the cone tip resistances measured by the Micro-Cone and the miniature cone (16mm in outer diameter) are similar. Note the resolution is much higher in the Micro-Cone. This study shows that the Micro-Cone may effectively detect the soil interface with high resolution, and with minimum disturbance.

• Journal of the Korean Geotechnical Society
• /
• v.25 no.8
• /
• pp.33-45
• /
• 2009

The conventional single-core PBDs have been widely used in order to accelerate consolidation settlement of soft grounds. When using the single-core PBD in a thick clay deposit, a delay of consolidation may occur due to high confining pressure in the thick deposit and necking of drains. This study is to compare the performances of soil improvement by the single-core and double-core PBD installed at a site in Busan New Port which exhibits approximately a 40m-thick clay layer. An in-situ test program was performed at the test site where a set of the double-core PBDs and single-core PBDs were installed to compare the efficiency of each drain. In addition, the discharge capacity of each PBD has been measured using the modified Delft Test. A series of laboratory tests for estimating in-situ soil properties have also been performed in order to obtain input parameters for a numerical program ILLICON. The discharge capacity of the double-core PBD is higher than that of the single-core PBD in the modified Delft Test. However it is observed from the comparative in-situ test and numerical analysis that there is no difference in the performance of ground improvement between the two drain systems. This discrepancy comes from the fact that the amount of water released during consolidation in most common field conditions is much smaller than the capacity of even the single core PBD. And thus, considering actual field conditions, it can be concluded that the single-core PBD has enough discharge capacity even in the thick clay deposit such as this test site.

• Journal of the Korean Geotechnical Society
• /
• v.24 no.5
• /
• pp.107-115
• /
• 2008

This paper is performed to examine the effect of creep during the primary consolidation and the applicability of the Yin's EVP (Elasto-Visco-Plastic) model. In ordinary consolidation theories using the elastic model, the primary consolidation process can be expressed but the secondary consolidation process cannot. It is due to the viscosity, which can express the secondary consolidation, and is sometimes related to the scale effect (difference of the thickness of clay layer between laboratory sample and field condition) such as hypotheses Type A and Type B shown by Ladd et al. (1977). Usually, the existence of the creep during the primary consolidation has been conformed and the Type B is well acceped. On the other hand, from the large-scaled consolidation tests the intermediate characteristic between Type A and Type B was proposed as Type C by Aboshi (1973). In this study, to clarify the effect of creep on the settlement-time relation during the primary consolidation in detail, Type B consolidation tests were performed using the separate-type consolidation test apparatus for a peat and clay. Then the test results were analyzed by using Yin's EVP Model (Yin and Graham, 1994). In conclusion, followings were obtained. At the end of primary consolidation, the compression for the subspecimens should not be the same because of the difference of the excess pore water pressure dissipation rate. And the average settlement measured by the separate-type consolidometer coincides with the analyzed one using the Yin's EVP model. As for the dissipation of the excess pore water pressure, however, the measured excess pore water pressure dissipates faster compared with the Yin's model.

• Geophysics and Geophysical Exploration
• /
• v.25 no.4
• /
• pp.214-216
• /
• 2022

Distributed acoustic sensing (DAS), an increasingly growing acquisition technique in the oil and gas exploration and seismology fields, has been used to record seismic signals using optical cables as receivers. With the development of imaging methods for DAS data, full waveform inversion (FWI) is been applied to DAS data to obtain high-resolution property models such as P- and S-velocity. However, because the DAS systems measure strain from the phase distortion between two points along optical cables, DAS data must be transformed from strain to particle velocity for FWI algorithms. In this study, a plane-wave FWI algorithm based on the relationship between strain and horizontal particle velocity in the plane-wave assumption is proposed to apply FWI to DAS data. Under the plane-wave assumption, strain equals the horizontal particle velocity, which is scaled by the velocity at the receiver position. This relationship was confirmed using a numerical experiment. Furthermore, 4-layer and modified Marmousi-2 velocity models were used to verify the applicability of the proposed FWI algorithm in various survey environments. The proposed FWI was implemented in land and marine survey environments and provided high-resolution P- and S-velocity models.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.2
• /
• pp.343-350
• /
• 2022

It is worth verifying the effectiveness of data integration between data with different features. This study investigated whether the data integration affects the accuracy of deep neural network (DNN), and which integration method shows the best improvement. This study used two different public datasets. One public dataset was taken in an actual farm in India. And another was taken in a laboratory environment in Korea. Leaf images were selected from two different public datasets to have five classes which includes normal and four different types of plant diseases. DNN used pre-trained VGG16 as a feature extractor and multi-layer perceptron as a classifier. Data were integrated into three different ways to be used for the training process. DNN was trained in a supervised manner via the integrated data. The trained DNN was evaluated by using a test dataset taken in an actual farm. DNN shows the best accuracy for the test dataset when DNN was first trained by images taken in the laboratory environment and then trained by images taken in the actual farm. The results show that data integration between plant images taken in a different environment helps improve the performance of deep neural networks. And the results also confirmed that independent use of plant images taken in different environments during the training process is more effective in improving the performance of DNN.

• Korean Chemical Engineering Research
• /
• v.61 no.3
• /
• pp.341-347
• /
• 2023

In PEMFC, PtCo/C alloy catalysts are widely used because of good performance and durability. However, few studies have been reported on the durability of carbon supports of PtCo/C evaluated at high voltages (1.0~1.5 V). In this study, the durability of PtCo/C catalysts and Pt/C catalysts were compared after applying the accelerated degradation protocol of catalyst support. After repeating the 1.0↔1.5V voltage change cycles, the mass activity, electrochemical surface area (ECSA), electric double layer capacitance (DLC), Pt dissolution and the particle growth were analyzed. After 2,000 cycles of voltage change, the current density per catalyst mass at 0.9V decreased by more than 1.5 times compared to the Pt/C catalyst. This result was because the degradation rate of the carbon support of the PtCo/C catalyst was higher than that of the Pt/C catalyst. The Pt/C catalyst showed more than 1.5 times higher ECSA reduction than the PtCo/C catalyst, but the corrosion of the carbon support of the Pt/C catalyst was small, resulting in a small decrease in I-V performance. In order to improve the high voltage durability of the PtCo/C catalyst, it was shown that improving the durability of the carbon support is essential.

• Journal of the Korea Safety Management & Science
• /
• v.25 no.2
• /
• pp.17-28
• /
• 2023

This study aimed to verify the validity of the evaluation items and weight determination of the indoor environmental safety area, which has the most frequent accidents, among the safety certification evaluations of educational facilities by the Ministry of Education of the Republic of Korea, which has been conducted since May 2021. As a preceding study, the evaluation items of the school safety evaluation checklist being implemented in the US state of Vermont were compared, and the causes of accidents judged by teachers in the accident experiences written by 200 Korean teachers were compared with the safety certification evaluation items belonged to the Ministry of Education. In addition, research literature using the AHP analysis technique on safety risks of elementary and secondary schools in China and safety evaluation index study cases of 539 elementary school children in Indonesia were analyzed. Through these preceding studies, measures to add and adjust evaluation items were derived and the validity and importance rankings of evaluation items were calculated through AHP questionnaires to teachers and safety experts. In addition, a survey was conducted on 104 ordinary people to verify the results of expert analysis. As a result of expert AHP analysis, 'safety education and disaster response training (.396)' was the highest priority for the relative importance of the first layer, followed by 'safety measures (.387)' and 'building materials'. Safety (.216)' was found to be the highest priority. In the overall importance ranking of the 13 second-tier screening items, safety accident prevention education had the highest priority and disaster preparedness training ranked second, proving that the Ministry of Education's review weight was underestimated. In addition, slip and collision accident countermeasures, which were not in the existing Ministry of Education review items, ranked 4th, laboratory practice room safety measures ranked 6th, and sanitation, cleanliness, hazardous substance management, and cafeteria/cooking room safety measures ranked 9th, indicating a significant level of importance. Referring to the importance ranking, which is the result of this study, it is suggested that it is necessary to review the weight of each review item again.