• Journal of Digital Convergence
• /
• v.16 no.7
• /
• pp.223-229
• /
• 2018

Recently, the necessity of safety diagnosis of electrical devices has been increasing as the fire caused by electric devices has increased rapidly. This study is concerned with the safety diagnosis of electric equipment using intelligent Fuzzy technology. It is used as a diagnostic input for the multiple electrical safety factors such as the use current, cumulative use time, deterioration and arc characteristics inherent to the equipment. In order to extract these information in real time, a device composed of various sensor circuits, DSP signal processing, and communication circuit is implemented. The fuzzy logic algorithm using the Gaussian function for each information is designed and compiled to be implemented on a small DSP board. The fuzzy logic receives the four diagnostic information, deduces it by the fuzzy engine, and outputs the overall safety status of the device as a 100-step analog fuzzy value familiar to human sensibility. By experiments of a device that combines hardware and fuzzy algorithm implemented in this study, it is verified that it can be implemented in a small DSP board with human-friendly fuzzy value, diagnosing real-time safety conditions during operation of electric equipment. In the future, we expect to be able to study more intelligent diagnostic systems based on artificial intelligent with AI dedicated Micom.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.9
• /
• pp.4542-4551
• /
• 2013

Recently, much attention has focused on the study of eco-friendly concrete using recycled by-products for protecting marine ecosystem and durability of concrete exposed to marine condition. This study evaluated the durabilities of 4 different type of concrete mixtures(Control, Marine, Porous, New slag) with the seawater resistance by marine environment exposure experiment and freeze-thaw resistance, resistance to chloride ion penetration considering severe deterioration environment. In this study, we conducted seawater resistance using compressive strength according to the age(7/28/56 days) of specimen and curing conditions(standard(fresh water), tidal, immersion, artificial seawater). The results show that compressive strength of concrete exposed to marine environment exposure condition was lower than those of the standard curing condition. Also, compressive strength of New slag using eco-friendly materials for protecting marine ecosystem was lower than those of other concretes, there is need to improve the performance of New slag. The results for freeze-thaw resistance showed that all mixtures have excellent, but the Porous and New slag were lower than others. Also, the more improved resistance to chloride ion penetration than those of the Marine was measured in the New slag regardless of curing condition.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.2
• /
• pp.750-762
• /
• 2019

The importance of the call center as a contact point for the enterprise is growing. However, call centers have difficulty with their operating agents due to the agents' lack of knowledge and owing to frequent agent turnover due to downturns in the business, which causes deterioration in the quality of customer service. Therefore, through an N-bank call center case study, we developed a system to reduce the burden of keeping up business knowledge and to improve customer service quality. It is a "real-time agent advisor" system that provides agents with answers to customer questions in real time by combining AI technology for speech recognition, natural language processing, and questions & answers for existing call center information systems, such as a private branch exchange (PBX) and computer telephony integration (CTI). As a result of the case study, we confirmed that the speech recognition system for real-time call analysis and the corpus construction method improves the natural speech processing performance of the query response system. Especially with name entity recognition (NER), the accuracy of the corpus learning improved by 31%. Also, after applying the agent advisor system, the positive feedback rate of agents about the answers from the agent advisor was 93.1%, which proved the system is helpful to the agents.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.6 no.4
• /
• pp.267-274
• /
• 2018

Though the wet bottom ash has been used as a type of lightweight aggregate, dry bottom ash, new type bottom ash from coal combustion power plant, has scarcely researched. It is excellent lightweight aggregate in the view point of construction material. This study is performed to check the applicability of dry bottom ash as a fine aggregate in lightweight aggregate concrete, by analyzing various properties of fresh and hardened concrete. We get results that the slump of concrete is within the target range at less than 75% replacement rate of dry bottom ash, the air content is not affected by the replacement rate of dry bottom ash, the bleeding capacity is less than $0.025cm^3/cm^2$ at 75% under of the replacement rate of dry bottom ash, and the compressive strength of concrete show 90% or more comparing the base mix while initial strength development is a little low. Oven dry unit weight of concrete is reduced by 8.9% when replaced 100% dry bottom ash, and dry shrinkage tends to decrease depending on increase of replacement rate of dry bottom ash. Modulus of elasticity of concrete shows no decease at 50% over of the replacement rate of dry bottom ash, while modulus of elasticity of concrete decreases when the replacement rate increases further. The dry bottom ash, when used as a fine aggregate in lightweight concrete, can be used effectively without any deterioration in quality.

• Journal of the Korean Geotechnical Society
• /
• v.39 no.7
• /
• pp.31-37
• /
• 2023

Since tunnels are built underground, it is impossible to check visually the location and degree of deterioration of steel ribs. Therefore, in tunnel maintenance, GPR images are generally used to detect steel ribs. While research on GPR image analysis employing artificial neural networks has primarily focused on detecting underground pipes and road damage, there have been limited applications for analyzing tunnel GPR data, specifically for steel rib detection, both internationally and domestically. In this study, a one-step object detection algorithm called YOLO, based on a convolutional neural network, was utilized to automate the localization of steel ribs using GPR data. The performance of the algorithm is then analyzed. Two datasets were employed for the analysis. A dataset comprising 512 original images and another dataset consisting of 2,048 augmented images. The omission rate, which represents the ratio of undetected steel ribs to the total number of steel ribs, was 0.38% for the model using the augmented data, whereas the omission rate for the model using only the original data was 7.18%. Thus, from an automation standpoint, it is more practical to employ an augmented dataset.

• Journal of Wetlands Research
• /
• v.17 no.1
• /
• pp.62-74
• /
• 2015

Urban areas are variously under threat including deterioration of ecological functions. Many pond wetland types have been created as part of an effort to improve and restore this urban environment. This study was arranged to examine improvement plans of wetlands in urban areas by analyzing semi-natural wetlands in farm areas. As for environment for water quality, it suggested the inflow of natural water neighboring rivers or the direct inflow of rain as the improvement plans. The result which analyzed soil pH, OM, and T-N content of the soil environment mentioned that urban areas supplied artificial sluices, removed apoptotic bodies, and used artificial soil and waterproofing materials and use of natural materials in design and construction, the sluice state of the natural form, and negligence of autumn plants were suggested as the improvement plans. Florae appeared in the subject sites of the study have found that there are 35 families 69 species in urban areas and 53 families 142 species in rural areas. As the average has found that there are 18.5 families 29.3 species in 4 urban areas and 26.3 families 53.5 species in 4 rural areas, the big difference between them was analyzed. As the cause has found that there are differences in yearly plants in farming areas when compared to urban areas, creation of various basic environments including soil and water quality was suggested to make yearly plants settle down widely. Naturalized plants have found that there are no big differences between urban areas and rural areas. However, the average of the naturalized ratio in urban areas is 17.4% as the naturalized plants are about 1/4 of the appeared plants. As it was analyzed to be higher than 7.7%, the average of the naturalized ratio in farming areas as the big difference, creation of various inhabiting environments was suggested to make more yearly plants appear like the analyzed result of the life type. Consideration of placement, materials, and inhabiting environments was suggested to make creation of wetlands well appreciated to improve functions of wetlands in urban areas. It is expected that the above results of the study will be utilized in creation and improvement of the pond wetlands which can play a huge role in increase and improvement of biological diversity in urban areas.

• Journal of Intelligence and Information Systems
• /
• v.23 no.3
• /
• pp.139-153
• /
• 2017

Corporate bankruptcy can cause great losses not only to stakeholders but also to many related sectors in society. Through the economic crises, bankruptcy have increased and bankruptcy prediction models have become more and more important. Therefore, corporate bankruptcy has been regarded as one of the major topics of research in business management. Also, many studies in the industry are in progress and important. Previous studies attempted to utilize various methodologies to improve the bankruptcy prediction accuracy and to resolve the overfitting problem, such as Multivariate Discriminant Analysis (MDA), Generalized Linear Model (GLM). These methods are based on statistics. Recently, researchers have used machine learning methodologies such as Support Vector Machine (SVM), Artificial Neural Network (ANN). Furthermore, fuzzy theory and genetic algorithms were used. Because of this change, many of bankruptcy models are developed. Also, performance has been improved. In general, the company's financial and accounting information will change over time. Likewise, the market situation also changes, so there are many difficulties in predicting bankruptcy only with information at a certain point in time. However, even though traditional research has problems that don't take into account the time effect, dynamic model has not been studied much. When we ignore the time effect, we get the biased results. So the static model may not be suitable for predicting bankruptcy. Thus, using the dynamic model, there is a possibility that bankruptcy prediction model is improved. In this paper, we propose RNN (Recurrent Neural Network) which is one of the deep learning methodologies. The RNN learns time series data and the performance is known to be good. Prior to experiment, we selected non-financial firms listed on the KOSPI, KOSDAQ and KONEX markets from 2010 to 2016 for the estimation of the bankruptcy prediction model and the comparison of forecasting performance. In order to prevent a mistake of predicting bankruptcy by using the financial information already reflected in the deterioration of the financial condition of the company, the financial information was collected with a lag of two years, and the default period was defined from January to December of the year. Then we defined the bankruptcy. The bankruptcy we defined is the abolition of the listing due to sluggish earnings. We confirmed abolition of the list at KIND that is corporate stock information website. Then we selected variables at previous papers. The first set of variables are Z-score variables. These variables have become traditional variables in predicting bankruptcy. The second set of variables are dynamic variable set. Finally we selected 240 normal companies and 226 bankrupt companies at the first variable set. Likewise, we selected 229 normal companies and 226 bankrupt companies at the second variable set. We created a model that reflects dynamic changes in time-series financial data and by comparing the suggested model with the analysis of existing bankruptcy predictive models, we found that the suggested model could help to improve the accuracy of bankruptcy predictions. We used financial data in KIS Value (Financial database) and selected Multivariate Discriminant Analysis (MDA), Generalized Linear Model called logistic regression (GLM), Support Vector Machine (SVM), Artificial Neural Network (ANN) model as benchmark. The result of the experiment proved that RNN's performance was better than comparative model. The accuracy of RNN was high in both sets of variables and the Area Under the Curve (AUC) value was also high. Also when we saw the hit-ratio table, the ratio of RNNs that predicted a poor company to be bankrupt was higher than that of other comparative models. However the limitation of this paper is that an overfitting problem occurs during RNN learning. But we expect to be able to solve the overfitting problem by selecting more learning data and appropriate variables. From these result, it is expected that this research will contribute to the development of a bankruptcy prediction by proposing a new dynamic model.

• Korean Journal of Heritage: History & Science
• /
• v.43 no.1
• /
• pp.100-117
• /
• 2010

This study analyzed vegetation structure of natural monuments Koelreuteria paniculata community in search of a conservation and management plan. Plant sociological analysis of Koelreuteria paniculata community indicates that it can be classified into Achyranthes japonica subcommunity and Rhodotypos scandens subcommunity and Trachelospermum asiaticum var. intermedium subcommunity. While Koelreuteria paniculata community of Ahnmyeondo is composed of sub tree layer and herb layer, those of Pohang and Wando are composed of tree layer, Sub tree layer, shrub layer, herb layer. The results of tree vitality analysis showed that those in Ahnmyeondo appeared to be relatively low when compared to those in Pohang and Wando-gun. This can be understood in two different aspects: disease and insects vulnerability due to a relatively simple structure and lack of competitive species, and decreased vitality / natural branch losses due to crown competition arising from high density. The result of soil characteristics analysis showed that soil texture, soil pH, organic matter, $p_2O_5$, exchange positive ion were sufficient for tree growth while total nitrogen was not, so that discretion would be needed for fertilizer application. As there were damages of disease and inscet, but only for 10~15% of the entire area; it still requires consistent preconsideration. The study suggests the management methods for preservation of Koelreuteria paniculata community. First, securing designated areas is necessary in order to minimize environment deterioration due to surrounding development. Especially, for sections with decreased areas, expansion of designated areas through land purchase should also be considered. Second, artificial interference may affect the livestock. Therefore, monitoring of artificial interference is necessary, based on which protection projects must be conducted. Third, from analysis of young plants which influence the maintenance mechanisms of Koelreuteria paniculata community, a decrease compared to the prior year was observed; investigation is needed. Therefore, an active management policy through status examination of livestock such as germination and young plants is necessary.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.22 no.3
• /
• pp.82-98
• /
• 2019

The importance of spatial information is rapidly rising. In particular, 3D spatial information construction and modeling for Real World Objects, such as smart cities and digital twins, has become an important core technology. The constructed 3D spatial information is used in various fields such as land management, landscape analysis, environment and welfare service. Three-dimensional modeling with image has the hig visibility and reality of objects by generating texturing. However, some texturing might have occlusion area inevitably generated due to physical deposits such as roadside trees, adjacent objects, vehicles, banners, etc. at the time of acquiring image Such occlusion area is a major cause of the deterioration of reality and accuracy of the constructed 3D modeling. Various studies have been conducted to solve the occlusion area. Recently the researches of deep learning algorithm have been conducted for detecting and resolving the occlusion area. For deep learning algorithm, sufficient training data is required, and the collected training data quality directly affects the performance and the result of the deep learning. Therefore, this study analyzed the ability of detecting the occlusion area of the image using various image quality to verify the performance and the result of deep learning according to the quality of the learning data. An image containing an object that causes occlusion is generated for each artificial and quantified image quality and applied to the implemented deep learning algorithm. The study found that the image quality for adjusting brightness was lower at 0.56 detection ratio for brighter images and that the image quality for pixel size and artificial noise control decreased rapidly from images adjusted from the main image to the middle level. In the F-measure performance evaluation method, the change in noise-controlled image resolution was the highest at 0.53 points. The ability to detect occlusion zones by image quality will be used as a valuable criterion for actual application of deep learning in the future. In the acquiring image, it is expected to contribute a lot to the practical application of deep learning by providing a certain level of image acquisition.

• Journal of Biosystems Engineering
• /
• v.3 no.1
• /
• pp.47-63
• /
• 1978

This study was conducted to obtain a basic information necessary to assess present rice milling technology in Korea The profiles for milling plants was analyzed by survey work.For the private custom-work mills, which process about 80 percent of domestic rice consumption ,their actual milling test for the identical samples as used for filed mills was conducted. Two rice varieties Japonica and Tongil-type were associated with the experiments. The results are summarized as follows: 1. Analyses for private custom-work mills showed their general aspects as; about 91 percent of the mills belonged to an individual owner ship ; more than 93 percent of the mills was established earlier than 1950 ; about 80 percent of the mills was powered with electric motor; mills having less than two employees were about 75 percent; about 45 percent of the mills provided for warehouse in storing customers cereal grains. 2. The polishers installed in 1,255mills within the surveyed area (7 counties) have been supplied by 44 different domestic manufacturers ;in but about 60 percent of which was supplied by 6major manufacturers. The polishers could be classified into two categories in terms of principles of their polishing actions ;jet-pearler and friction types. About 51 percent of the mills was equipped with the former which has been recognized as giving greater milling recovery than the friction types. 3. Reason for owners of private mills to supplement new machines was due mainly to pgrading their mills to meet the requirements that established by the Government. However, about 60 percent of the mill owners intended to replace with new pearler by their own needs to meet with new high yielding varieties. 4. Processing systems of each previate rice mills surveyed could be classified into three categories, depending upon whether the systems posessed such components as precleaner and paddy separator or not. Only 36.7 percent of mills was installed with both precleanr ad paddy seperrator, 5.0 percent of mills did have neither percleaner nor paddy seperator, and rest of them equipped only one of the two. Hence,it is needed for about 63% of rice miils to be supplemented with these basic facilities to meet with the requirements for the standaized system. 5. Actual milling capacity measured at each field rice mills was shown a wide variation, having range from about 190 to 1,210 kg/hr. The percentages of mills classified according to daily milling capacity based on this hourly capacity were 24.3% for the capacity less than 3 M/T a day; 20.0% for 3-4 M/T; 15.6% for 4-5 M/T; 6.7% for 5-6 M/T; 22.3% for 6-7 M/T; and 11.0% for more than 7 M/T a day. 6. Actual amount of rice processed was about 310 M/T a year in average. About 42% of total milled rice was processed during October to Decembear, which formed a peak demand period for rice mills. The amount of rice milled during January to May was relatively small, but it had still a large amount compared to that during June to September. 7. Utilization rate of milling facility, i. e., percentage of the actual amount of milled rice to the capacity of rice mills, was about 18% on the year round average, about 41% in the peak demand season, and about 10% during June to September. Average number of operating days for mills surveyed was about 250 days a year, and about 21 days a month. 8. Moisture contents of paddy at the time of field mill tests were ranged 14.5% to 19.5% for both Japonica and Tong-i] varieties, majority of paddy grains having moisture level much higher than 1530. To aviod potential reduction of milling recovery while milling and deterioration of milled rice while storage due to these high grain mJisture contents, it may be very important for farmers holding rice to dry by an artificial drying method. 9. Milling recovery of JapJnica varieties in rice mills was 75.0% in average and it was widely ranged from 69.0% to 78.0 % according to mills. Potential increase in milJing recovery of Japonica variety with improvement of mill facilities was estimated to about 1.9%. On the other hand, milling recovery of Tong-il varieties in the field mill tests was 69.8% in average and it ranged from 62% to 77 %, which is much wider than that of Japonica varieties. It is noticed that the average milling recovery of Tong-il variety of 69.8% was much less than that of the Japonica-type. It was estimated th3.t up to about 5.0% of milling recovery for Tong-il variety could be improved by improving the present lo'.ver graded milling technology. 10. Head rice recoveries, as a factor of representing the quality of commercial goods, of Japonica and Tong-il varieties were 65.9% and 53.8% in average, and they were widely ranged from 52% to 73% and from 44% to 65% , respectively. It was assessed that head rice recovery of Japonica varieties can be improved up 3.3% and that of Tong-il varieties by 7.0% by improving mill components and systems.