• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.101-125
• /
• 2022

Recently, with the development of computing technology and the improvement of the cloud environment, deep learning technology has developed, and attempts to apply deep learning to various fields are increasing. A typical example is anomaly detection, which is a technique for identifying values or patterns that deviate from normal data. Among the representative types of anomaly detection, it is very difficult to detect a contextual anomaly that requires understanding of the overall situation. In general, detection of anomalies in image data is performed using a pre-trained model trained on large data. However, since this pre-trained model was created by focusing on object classification of images, there is a limit to be applied to anomaly detection that needs to understand complex situations created by various objects. Therefore, in this study, we newly propose a two-step pre-trained model for detecting abnormal situation. Our methodology performs additional learning from image captioning to understand not only mere objects but also the complicated situation created by them. Specifically, the proposed methodology transfers knowledge of the pre-trained model that has learned object classification with ImageNet data to the image captioning model, and uses the caption that describes the situation represented by the image. Afterwards, the weight obtained by learning the situational characteristics through images and captions is extracted and fine-tuning is performed to generate an anomaly detection model. To evaluate the performance of the proposed methodology, an anomaly detection experiment was performed on 400 situational images and the experimental results showed that the proposed methodology was superior in terms of anomaly detection accuracy and F1-score compared to the existing traditional pre-trained model.

• The Journal of Engineering Geology
• /
• v.32 no.2
• /
• pp.221-239
• /
• 2022

Unknown geotechnical characteristics are key challenges in the design of piles for the plant, civil and building works. Although the N-values which were read through the standard penetration test are important, those N-values of the whole area are not likely acquired in common practice. In this study, the N-value is predicted by means of regression analysis with artificial intelligence (AI). Big data is important to improve learning performance of AI, so circular augmentation method is applied to build up the big data at the current study. The optimal model was chosen among applied AI algorithms, such as artificial neural network, decision tree and auto machine learning. To select optimal model among the above three AI algorithms is to minimize the margin of error. To evaluate the method, actual data and predicted data of six performed projects in Poland, Indonesia and Malaysia were compared. As a result of this study, the AI prediction of this method is proven to be reliable. Therefore, it is realized that the geotechnical characteristics of non-boring points were predictable and the optimal arrangement of structure could be achieved utilizing three dimensional N-value distribution map.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.26 no.2
• /
• pp.18-27
• /
• 2022

In this study, the time-dependent chloride ingress behavior in GGBFS concrete was evaluated considering marine exposure conditions and the properties of concrete mixtures. The concrete mixture for this study had 3 levels of water to binder ratio and the substitution rate of GGBFS, and outdoor exposure tests were performed considering submerged area, tidal area, and splash area. According to the evaluation results of diffusion coefficient considering properties of concrete mixtures, as the substitution rate of GGBFS increased, the decreasing rate of the diffusion coefficient decreased based on exposure periods of 730 days(2 years). As the evaluation result of the diffusion behavior according to the marine exposure conditions, the diffusion coefficient was evaluated in the order of submerged area, tidal area, and splash area. In tidal area, a relatively high diffusion coefficient was evaluated due to the repetition of wet and dry seawater. In this study, the effects of GGBFS substitution rate on the decreasing behavior of apparent chloride diffusion coefficient was analyzed in consideration of exposure conditions and periods. Linear regression analysis was performed with apparent chloride diffusion coefficient as output value and GGBFS substitution rate as input value. After 730 days of exposure, the effect of GGBFS on diffusion coefficient was significantly reduced. Even for OPC concrete, after 730 days, the diffusion coefficient was as low as that of GGBFS concrete, so the gradient of the regression equation decreased significantly. It is thought that improved durability performance for chloride ingress can be secured before 730 days through the use of GGBFS.

• Tunnel and Underground Space
• /
• v.32 no.1
• /
• pp.78-92
• /
• 2022

In this study, we proposed a diagnostics and assessment model for mining sites that can evaluate the smart mining technology level in a systematic and structured way. For this, the maturity of the smart mining was defined, and detailed assessment items of the diagnostics and assessment model for smart mining were derived by considering the smart factory diagnostics and assessment model (KS X 9001-3) used in the manufacturing industry. While maintaining the existing system, the existing 46 detailed assessment items were modified to be suitable for mining. As a result, a total of 29 detailed assessment items were derived in the areas of promotion strategy, process, information system and automation, and performance. Based on this, a questionnaire was designed to diagnose the level of smart mining technology, and assessment was performed by applying it to domestic iron mines. The level of smart mining technology in the study area was found to be level 2, and it could be inferred that it was about 40% lower than the average smart level of the general manufacturing industry. In addition, by using the developed model, it was possible to recognize the weak points of the mine at each stage of the introduction, operation, and advancement of smart mining, and to suggest investment and improvement directions.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.281-289
• /
• 2022

Plasma ozone is utilized in a variety of applications in the field of sterilization due to its high sterilization performance. Dielectric materials used in DBD(dielectric barrier discharges) are mainly polymer, quartz and ceramics. These dielectric layers have the advantage of limiting the amount of supplied electron charge and allowing plasma to occur evenly on the surface of dielectric. Actually, the target or environment for sterilization is often a complex structure, so research and academic study are needed by utilizing the concept of space sterilization. In this study, the device is applied to generate DBD plasma at atmospheric pressure for disinfection due to the effectiveness in producing radicals and ozone. The generator of plasma ozone is a basic structure of dielectric barrier discharge by placing ceramic tube dielectrics and stainless steel electrical conductors at regular intervals. Various applications can be developed based on the proposed design method. Plasma ozone generation for space sterilization device is recognized as an excellent sterilization device. Through the design and verification of the device, we intend to establish an optimal design of the spatial sterilization device and provide the basis data for sterilization applications.

• Tunnel and Underground Space
• /
• v.31 no.6
• /
• pp.469-479
• /
• 2021

As a material for preventing spalling of concrete, the effectiveness of PP fiber has already been confirmed. However, it is necessary to consider the maximum temperature that occurs during a fire, and to solve the mixing problem and the strength reduction problem that occur depending on the mixing amount. In this study, the fire resistance performance of tunnel segment linings according to the PP fiber content and air volume under the RABT fire scenario was investigated. As a result, no spalling or cross-sectional loss occurred in all test specimens, and when the PP fiber content was small, the maximum temperature was relatively high and the maximum temperature arrival time was also fast. On the other hand, no trend was found for the maximum temperature and arrival time according to the difference in air volume. In the internal temperature distribution results for the PP fiber mixing amount of 0.75, 1.0, 1.5, and 2.0 kg/m³, the results of 0.75 and 1.0 kg/m³ showed similar temperature distribution, and the results of 1.5 and 2.0 kg/m³ were similar. It was confirmed that the internal temperature distribution tends to decrease at the same depth when the amount of PP fiber mixed is large, and it was confirmed that a remarkable difference occurred from the results of 1.0 kg/m³ and 1.5 kg/m³ of PP fiber mixed amounts.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.26 no.2
• /
• pp.59-67
• /
• 2022

In this study, the monthly airborne chlorides flying into the offshore bridges were investigated depending on the sea level. The target structures were 9 bridges located on the western and southern coast of South Korea. The airborne chlorides were measured at different sea levels on each bridge every month during 1 year. The results showed that the strongest seasonal wind from the northwest in winter expecially have led increase of the airborne chlorides, and its effect was more significant in the western coast than the southern coast. It was also found that the airborne chlorides declined with the increase of sea level. Three types of curves were suggested for analyzing the decrease trend with the sea level, based on the airborne chlorides at the lowest measurement height of main tower. The trend was varied depending on the sea area, and even in the same sea area, the local topographic condition affected the airborne chlorides. It means that the location and local topography should be considered simultaneously for durability management in the framework of the chloride source, and then the influence of the chloride source should be classified, e.g. safe and dangerous. From these results, it is expected that it could be used as baseline data for the evaluation of the deterioration environment in the Detailed guidelines for safety and maintenance of facilities [Performance evaluation]_Bridge.

• Smart Media Journal
• /
• v.11 no.5
• /
• pp.75-81
• /
• 2022

In order to respond to carbon neutrality and climate change in agriculture, agricultural machinery, which has been developed centered on internal combustion engines, needs to be converted to an electric-based technology that does not emit greenhouse gases. In this study, simulations for electric UTV suspension design were performed to reduce vibration and shock of electric UTV for agricultural use and to improve driving stability and control performance of the vehicle. The simulation was performed by dividing the tolerance load of the vehicle body and the loaded load state. The range of motion of the suspension spring of UTV is within 30% of the range of motion under condition B under tolerance, the displacement of the UTV suspension with full load is reduced from 264mm to 121mm, and the damping speed is 260mm/s to 300mm/s that it can be seen that the range of motion is within 60%. Suspension design of electric UTV for multi-purpose agricultural work is a very important factor for maintaining agricultural work ability in towing work such as tillage as well as driving and terrain adaptation. The results of this study can be usefully used to determine the spring parameters with the appropriate damping range so that the electric UTV can be used for various agricultural tasks.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.4
• /
• pp.435-444
• /
• 2022

There are three focuses in the paper. Firstly, the study identified what channels were most viewed by YouTube users to watch the 'Daejang-dong scandal,' which was the most powerful agenda to influence the candidate preference among voters during the 20th presidential election. Secondly, the study analyzed whether the political tone of the first videos was in line with that of the subsequent videos. Finally, we compared the sentiment of comments on the first and subsequent videos. The results showed that TBS 'News Factory' and 'TV Chosun News' represented liberal and conservative factions, respectively. Secondly, the political tone of channels that were viewed subsequently was neutral, but the conservative channel users left more negative comments and that was significant statistically. In addition, about 80% of the conservative and liberal channel users shared the same political tendency with the channel they watched first, and more than 90% of the comments left at the subsequent videos in line with that of at the first news. Based on these results, the study concluded that the voters tended to seek political news that was similar with their political ideology, and it was considered a sort of echo chamber phenomenon on the YouTube. The study suggests that the performance of high-quality journalism by traditional news outlet might contribute to decrease the negative influence of political contents on YouTube users.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.7
• /
• pp.299-306
• /
• 2022

In recent, sensors embedded in robots, equipment, and circuits have become common, and research for diagnosing device failures by learning measured sensor data is being actively conducted. This failure diagnosis study is divided into a classification model for predicting failure situations or types and a regression model for numerically predicting failure conditions. In the case of a classification model, it simply checks the presence or absence of a failure or defect (Class), whereas a regression model has a higher learning difficulty because it has to predict one value among countless numbers. So, the reason that regression modeling is more difficult is that there are many irregular situations in which it is difficult to determine one output from a similar input when predicting by matching input and output. Therefore, in this paper, we focus on input and output data with periodicity, analyze the input/output relationship, and secure regularity between input and output data by performing sliding window-based input data patterning. In order to apply the proposed method, in this study, current and temperature data with periodicity were collected from MMC(Modular Multilevel Converter) circuit system and learning was carried out using ANN. As a result of the experiment, it was confirmed that when a window of 2% or more of one cycle was applied, performance of 97% or more of fit could be secured.