• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.32 no.6
• /
• pp.349-357
• /
• 2019

Finite element method (FEM) is utilized in the development of products to realistically analyze and predict the mechanical behavior of materials in various fields. However, the approach based on the numerical analysis of glass fiber reinforced plastic (GFRP) composites, for which the fiber orientation and strain rate affect the mechanical properties, has proven to be challenging. The purpose of this study is to define and evaluate the mechanical properties of glass fiber reinforced plastic composites using the numerical analysis models of Digimat, a linear, nonlinear multi-scale modeling program for various composite materials such as polymers, rubber, metal, etc. In addition, the aim is to predict the behavior of realistic polymeric composites. In this regard, the tensile properties according to the fiber orientation and strain rate of polybutylene terephthalate (PBT) with short fiber weight fractions of 30wt% among various polymers were investigated using references. Information on the fiber orientation was calculated based on injection analysis using Moldflow software, and was utilized in the finite element model for tensile specimens via a mapping process. LS-Dyna, an explicit commercial finite element code, was used for coupled analysis using Digimat to study the tensile properties of composites according to the fiber orientation and strain rate of glass fibers. In addition, the drawbacks and advantages of LS-DYNA's various anisotropic material models were compared and evaluated for the analysis of glass fiber reinforced plastic composites.

• Journal of Korea Multimedia Society
• /
• v.9 no.1
• /
• pp.23-32
• /
• 2006

This paper describes the optimization of a language model and an acoustic model to improve speech recognition using Korean unit digits. Since the model is composed of a finite state network (FSN) with a disyllable, recognition errors of the language model were reduced by analyzing the grammatical features of Korean unit digits. Acoustic models utilize a demisyllable pair to decrease recognition errors caused by inaccurate division of a phone or monosyllable due to short pronunciation time and articulation. We have used the K-means clustering algorithm with the transformed successive state splitting in the feature level for the efficient modelling of feature of the recognition unit. As a result of experiments, 10.5% recognition rate is raised in the case of the proposed language model. The demi-syllable fair with an acoustic model increased 12.5% recognition rate and 1.5% recognition rate is improved in transformed successive state splitting.

• Journal of the Korean Institute of Gas
• /
• v.23 no.5
• /
• pp.52-58
• /
• 2019

In this study, we have developed a forecasting model for city- gas acceptance. City-gas corporations have to report about city-gas sale volume next year to KOGAS. So it is a important thing to them. Factors influenced city-gas have differences corresponding to usage classification, however, in city-gas acceptence, it is hard to classificate. So we have considered tha outside temperature as factor that influence regardless of usage classification and the model development was carried out. ARIMA, one of the traditional time series analysis, and LSTM, a deep running technique, were used to construct forecasting models, and various Ensemble techniques were used to minimize the disadvantages of these two methods.Experiments and validation were conducted using data from JB Corp. from 2008 to 2018 for 11 years.The average of the error rate of the daily forecast was 0.48% for Ensemble LSTM, the average of the error rate of the monthly forecast was 2.46% for Ensemble LSTM, And the absolute value of the error rate is 5.24% for Ensemble LSTM.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.1
• /
• pp.179-184
• /
• 2006

Rapid and high-precision positioning with a Global Navigation Satellite System (GNSS) is feasible only when very precise carrier-phase observations can be used. There are two kinds of mathematical models for ambiguity resolution. The first one is based on both pseudorange and carrier phase measurements, and the observation equations are of full rank. The second one is only based on carrier phase measurement, which is a rank-defect model. Though the former is more commonly used, the latter has its own advantage, that is, ambiguity resolution will be freed from the effects of pseudorange multipath. Galileo will be operational. One of the important differences between Galileo and current GPS is that Galileo will provide signals in four frequency bands. With more carrier-phase data available, frequency combinations with long equivalent wavelength can be formed, so Galileo will provide more opportunities for fast and reliable ambiguity resolution than current GPS. This paper tries to investigate phase only fast ambiguity resolution performance with four Galileo frequencies for short baseline. Cascading Ambiguity Resolution (CAR) method with selected optimal frequency combinations and LAMBDA method are used and compared. To validate the resolution, two tests are used and compared. The first one is a ratio test. The second one is lower bound success-rate test. The simulation test results show that, with LAMBDA method, whether with ratio test or lower bound success rate validation criteria, ambiguity can be fixed in several seconds, 8 seconds at most even when 1 sigma of carrier phase noise is 12 mm. While with CAR method, at least about half minute is required even when 1 sigma of carrier phase noise is 3 mm. It shows that LAMBDA method performs obviously better than CAR method.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.4
• /
• pp.19-27
• /
• 2020

In this paper, we propose a machine learning method for diagnosing the failure of a gas pressure regulator. Originally, when implementing a machine learning model for detecting abnormal operation of a facility, it is common to install sensors to collect data. However, failure of a gas pressure regulator can lead to fatal safety problems, so that installing an additional sensor on a gas pressure regulator is not simple. In this paper, we propose various machine learning approach for diagnosing the abnormal operation of a gas pressure regulator with only the flow rate and gas pressure data collected from a gas pressure regulator itself. Since the fault data of a gas pressure regulator is not enough, the model is trained in all classes by applying the over-sampling method. The classification model was implemented using Gradient boosting, 1D Convolutional Neural Networks, and LSTM algorithm, and gradient boosting model showed the best performance among classification models with 99.975% accuracy.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.1
• /
• pp.19-28
• /
• 2022

Recently, cryptocurrency has attracted much attention, and price prediction studies of cryptocurrency have been actively conducted. Especially, efforts to improve the prediction performance by applying the deep learning model are continuing. LSTM (Long Short-Term Memory) model, which shows high performance in time series data among deep learning models, is applied in various views. However, it shows low performance in cryptocurrency price data with high volatility. Although, to solve this problem, new input features were found and study was conducted using them, there is a lack of study on input features that drop predictive performance. Thus, in this paper, we collect the recent trends of six cryptocurrencies including Bitcoin and Ethereum and analyze effects of input features on the cryptocurrency price predictive performance through statistics and deep learning. The results of the experiment showed that cryptocurrency price predictive performance the best when open price, high price, low price, volume and price were combined except for rate of closing price fluctuation.

• Korean Journal of Social Welfare Studies
• /
• v.42 no.4
• /
• pp.365-388
• /
• 2011

By analyzing 1998~2008 Korean Labor and Income Panel Study(KLIPS), this study examines socio-economic characteristics of people who become poor. The study also explores the reason why they are in the state of poverty. To find determinants affecting poverty entrance, discrete-time hazard models are applied. Major findings are as follows. The socio-economic characteristics driving people into poverty are in the middle way of the long-term poor and the non-poor, combining the characteristics of both groups. This implies that many cases of the newly poor tend to enter and exit from poverty repeatedly. Poverty entry rate was at a high level right after the economic crises, then was a downturn and remained fairly stable since 2000. However, the young, the high-educated, and even the professional are on the rise as a new poverty group. The major reason people become poor is temporary job loss. This factor is confirmed again by multi-variate analyses. In building anti-poverty policies, it is important to distinguish the long-term poor from the short-term poor. For the long-term poor, virtually the only affective policy will be income support. On the other hand, a labor-market strategy for jos security will be more effective for the short-term poor. The characteristics and determinants of poverty entry may affect poverty duration and exit in the future. Future research will be needed to investigate the relationship among these factors.

• Journal of Bio-Environment Control
• /
• v.23 no.2
• /
• pp.95-108
• /
• 2014

The objective of this study is to develop the plastic greenhouse models which are structurally safe under the weather condition of Seongju and have the dimensions suitable for oriental melon cultivation as well. To grasp the structural features of greenhouses in Seongju, the field survey was conducted on 406 farmhouses which included 2,068 greenhouses. The field survey showed that the roof shape of arch type accounted for the highest rate, but recently even span or peach type became more popular and the width and height of greenhouse tended to increase as the period of use was short. The relationship of the width, ridge height and eaves height were established based on field survey data. Using climate data of Gumi adjacent to Seongju, the regressions were determined for the design wind speed and design snow depth depending on recurrence period. To design the greenhouse models against weather disasters in Seongju, the optimal design loads are 23.7 cm of snow depth and $33.8m{\cdot}s^1$ of wind speed. As the design results, four models of single-span greenhouse, two models of double-span greenhouses including extension were developed.

• Journal of Korea Water Resources Association
• /
• v.43 no.11
• /
• pp.995-1009
• /
• 2010

Accelerated soil erosion due to extreme climate change, such as increased rainfall intensity, and human-induced environmental changes, is a widely recognized problem. Existing soil erosion models are generally based on the gross erosion concept to compute annual upland soil loss in tons per acre per year. However, such models are not suitable for event-based simulations of erosion and deposition in time and space. Recent advances in computer geographic information system (GIS) technologies have allowed hydrologists to develop physically based models, and the trend in erosion prediction is towards process-based models, instead of conceptually lumped models. This study aims to propose an effective and robust distributed rainfall-sediment yield-runoff model consisting of basic element modules: a rainfall-runoff module based on the kinematic wave method for subsurface and surface flow, and a runoff-sediment yield-runoff model based on the unit stream power method. The model was tested on the Cheoncheon catchment, upstream of the Yongdam dam using hydrological data for three extreme flood events due to typhoons. The model provided acceptable simulation results with respect to both discharge and sediment discharge even though the simulated sedigraphs were underestimated, compared to observations. The spatial distribution of erosion and deposition demonstrated that eroded sediment loads were deposited in the cells along the channel network, which have a short overland flow length and a gentle local slope while the erosion rate increased as rainfall became larger. Additionally, spatially heterogeneous rainfall intensity, dependant on Thiessen polygons, led to spatially-distinct erosion and deposition patterns.

• The Korean Journal of Applied Statistics
• /
• v.23 no.2
• /
• pp.219-234
• /
• 2010

In index investing according to KOSPI, we estimate Value at Risk(VaR) from the extreme losses of the daily returns which are obtained from KOSPI. To this end, we apply Block Maxima(BM) model which is one of the useful models in the extreme value theory. We also estimate the extremal index to consider the dependency in the occurrence of extreme losses. From the back-testing based on the failure rate method, we can see that the model is adaptable for the VaR estimation. We also compare this model with the GARCH model which is commonly used for the VaR estimation. Back-testing says that there is no meaningful difference between the two models if we assume that the conditional returns follow the t-distribution. However, the estimated VaR based on GARCH model is sensitive to the extreme losses occurred near the epoch of estimation, while that on BM model is not. Thus, estimating the VaR based on GARCH model is preferred for the short-term prediction. However, for the long-term prediction, BM model is better.