• Journal of Korean Tunnelling and Underground Space Association
• /
• v.18 no.1
• /
• pp.13-30
• /
• 2016

This paper concerns the development of an optimized TBM segmental lining design system for a subsea tunnel. The subsea tunnel is normally laid down under the sea water and submarine ground which consists of soil or rock. The design system is the series of process which can predict segmental lining member forces by ANN (artificial neural network system), analyze suitable section for the designated ground, construction and tunnel conditions. Finally, this lining design system aims to be connected with a BIM system for designing the subsea tunnel automatically. The lining member forces are predicted based on the ANN which was calculated by a FEM (finite element analysis) and it helps designers determine its segmental lining dimension easily without any further FE calculations.

• Journal of Korea Multimedia Society
• /
• v.21 no.4
• /
• pp.450-463
• /
• 2018

Sound based machine fault diagnosis is the process consisting of detecting automatically the damages that affect the machines by analyzing the sounds they produce during their operating time. The collected sounds being inevitably corrupted by random disturbance, the most important part of the diagnosis consists of discovering the hidden elements inside the data that can reveal the faulty patterns. This paper presents a novel feature extraction methodology that combines various digital signal processing and pattern recognition methods for the analysis of the sounds produced by the drills. Using the Fourier analysis, the magnitude spectrum of the sounds are extracted, converted into two-dimensional vectors and uniformly normalized in such a way that they can be represented as 8-bit grayscale images. Histogram equalization is then performed over the obtained images in order to adjust their very poor contrast. The obtained contrast enhanced images will be used as the features of our diagnosis system. Finally, principal component analysis is performed over the image features for reducing their dimensions and a nonlinear classifier is adopted to produce the final response. Unlike the conventional features, the results demonstrate that the proposed feature extraction method manages to capture the hidden health patterns of the sound.

• The Journal of Information Systems
• /
• v.26 no.3
• /
• pp.63-90
• /
• 2017

Purpose This study proposes a novel system trading model using case-based reasoning (CBR) based on absolute similarity threshold. The proposed model is designed to optimize the absolute similarity threshold, feature selection, and instance selection of CBR by using genetic algorithm (GA). With these mechanisms, it enables us to yield higher returns from stock market trading. Design/Methodology/Approach The proposed CBR model uses the absolute similarity threshold varying from 0 to 1, which serves as a criterion for selecting appropriate neighbors in the nearest neighbor (NN) algorithm. Since it determines the nearest neighbors on an absolute basis, it fails to select the appropriate neighbors from time to time. In system trading, it is interpreted as the signal of 'hold'. That is, the system trading model proposed in this study makes trading decisions such as 'buy' or 'sell' only if the model produces a clear signal for stock market prediction. Also, in order to improve the prediction accuracy and the rate of return, the proposed model adopts optimal feature selection and instance selection, which are known to be very effective in enhancing the performance of CBR. To validate the usefulness of the proposed model, we applied it to the index trading of KOSPI200 from 2009 to 2016. Findings Experimental results showed that the proposed model with optimal feature or instance selection could yield higher returns compared to the benchmark as well as the various comparison models (including logistic regression, multiple discriminant analysis, artificial neural network, support vector machine, and traditional CBR). In particular, the proposed model with optimal instance selection showed the best rate of return among all the models. This implies that the application of CBR with the absolute similarity threshold as well as the optimal instance selection may be effective in system trading from the perspective of returns.

• Journal of Advanced Navigation Technology
• /
• v.22 no.3
• /
• pp.233-239
• /
• 2018

Since solar power generation is intermittent depending on weather conditions, it is necessary to predict the accurate generation amount of solar power to improve the efficiency and economical efficiency of solar power generation. This study proposes a short - term deep learning prediction model of solar power generation using meteorological data from Mokpo meteorological agency and generation data of Yeongam solar power plant. The meteorological agency forecasts weather factors such as temperature, precipitation, wind direction, wind speed, humidity, and cloudiness for three days. However, sunshine and solar radiation, the most important meteorological factors for forecasting solar power generation, are not predicted. The proposed model predicts solar radiation and solar radiation using forecast meteorological factors. The power generation was also forecasted by adding the forecasted solar and solar factors to the meteorological factors. The forecasted power generation of the proposed model is that the average RMSE and MAE of DNN are 0.177 and 0.095, and RNN is 0.116 and 0.067. Also, LSTM is the best result of 0.100 and 0.054. It is expected that this study will lead to better prediction results by combining various input.

• Journal of Advanced Navigation Technology
• /
• v.22 no.5
• /
• pp.478-484
• /
• 2018

This study predicts solar radiation, solar radiation, and solar power generation using hourly weather data such as temperature, precipitation, wind direction, wind speed, humidity, cloudiness, sunshine and solar radiation. I/O pattern in supervised learning is the most important factor in prediction, but it must be determined by repeated experiments because humans have to decide. This study proposed four input and output patterns for solar and sunrise prediction. In addition, we predicted solar power generation using the predicted solar and solar radiation data and power generation data of Youngam solar power plant in Jeollanamdo. As a experiment result, the model 4 showed the best prediction results in the sunshine and solar radiation prediction, and the RMSE of sunshine was 1.5 times and the sunshine RMSE was 3 times less than that of model 1. As a experiment result of solar power generation prediction, the best prediction result was obtained for model 4 as well as sunshine and solar radiation, and the RMSE was reduced by 2.7 times less than that of model 1.

• Journal of the Korean Chemical Society
• /
• v.45 no.5
• /
• pp.422-428
• /
• 2001

Retention relative times(RRTs) of PAH molecules and their derivatives in gas chromatography are trained and predicted in testing sets using a multiple linear regression(MLR) and an artificial neural network(ANN). The main descriptors of PAHs and their derivatives in QSRR are the square root of molecular weight(sqmw), molecular connectivity($^1{\chi}_v$), molecular dipole moment(D) and length-to-breadth ratios(L/B). The results of MLR shows that a heavy molecule has a propensity for long retention time. L/B closely related with slot model is a good descriptor in MLR. On the other hand, ANN which is not effected by the linear dependencies among the descriptors were exclusively based on molecular weight and molecular dipole moment. The variances which shows the accuracy of prediction for retention times in testing sets are 1.860, 0.206 for MLR and ANN, respectively. It was shown that ANN can exceed the MLR in prediction accuracy.

• Communications for Statistical Applications and Methods
• /
• v.18 no.6
• /
• pp.859-869
• /
• 2011

Interest has increased in academic research regarding key factors that drive box-office success as well as the ability to predict the box-office success of a movie from a commercial perspective. This study analyzed the relationship between key success factors of a movie and box office records based on movies released in 2010 in Korea. At the pre-production investment decision-making stage, the movie genre, motion picture rating, director power, and actor power were statistically significant. At the stage of distribution decision-making process after movie production, among other factors, the influence of star actors, number of screens, power of distributors, and social media turned out to be statistically significant. We verified movie success factors through the application of a Multinomial Logit Model that used the concept of choice probabilities. The Multinomial Logit Model resulted in a higher level of accuracy in predicting box-office success compared to the Artificial Neural Network and Discriminant Analysis.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.5
• /
• pp.489-495
• /
• 2014

A micro-genetic algorithm (MGA) is one of the improved forms of a genetic algorithm. It is used to reduce the number of iterations and the computing resources required by using small populations. The efficiency of MGAs has been proved through many problems, especially problems with 3-5 design variables. This study proposes an optimization algorithm based on the sequential design of experiments (SDOE) and an MGA. In a previous study, the authors used the SDOE technique to reduce trial-and-error in the conventional approximate optimization method by using the statistical design of experiments (DOE) and response surface method (RSM) systematically. The proposed algorithm has been applied to various mathematical examples and a structural problem.

• Journal of Fashion Business
• /
• v.6 no.5
• /
• pp.136-150
• /
• 2002

In this study, we proposed a method to be standing customers as the supporting system for the improvement of fashion garment industry which was the marginal growth getting into full maturity of market. As for the customer creation method of Fashion garment company is developing a marketing program to be standing customer as customer scoring to estimate a existing customer‘s buying power, and figure out minimum fixed sales of company to use a future purchasing predict. This study was a result of data from total sixty thousands data to be created for the 11 months from september. 2000 to July. 2001. The data is part of which the company leading the Korean fashion garment industry has a lot of a customer purchasing history data. But this study used only 48,845 refined purchased data to discriminate from sixty thousands data and 21,496 customer case with the exception of overlapping purchased data among of those. The software used to handle sixty thousands data was SAS e-miner. As the analysis process is put in to operation the analysis of the purchasing customer’s profile firstly, and the second come into basket analysis to consider the buying associations for Association goods, the third estimate the customer grade of Customer loyalty by 3 ways of logit regression analysis, decision tree, Artificial Neural Network. The result suggested a method to be estimate the customer loyalty as 3 independent variables, 2 coefficients. The 3 independent variables are total purchasing amount, purchasing items per one purchase, payment amount by one purchasing item. The 2 coefficients are royal and normal for customer segmentation. The result was that this model use a logit regression analysis was valid as the method to be estimate the customer loyalty.

• Journal of Korean Society on Water Environment
• /
• v.30 no.5
• /
• pp.491-502
• /
• 2014

As meteorology is the driving force for lake thermodynamics and mixing processes, the effects of climate change on the physical limnology and associated ecosystem are emerging issues. The potential impacts of climate change on the physical features of a reservoir include the heat budget and thermodynamic balance across the air-water interface, formation and stability of the thermal stratification, and the timing of turn over. In addition, the changed physical processes may result in alteration of materials and energy flow because the biogeochemical processes of a stratified waterbody is strongly associated with the thermal stability. In this study, a novel modeling framework that consists of an artificial neural network (ANN), a watershed model (SWAT), a reservoir operation model(HEC-ResSim) and a hydrodynamic and water quality model (CE-QUAL-W2) is developed for projecting the effects of climate change on the reservoir water temperature and thermal stability. The results showed that increasing air temperature will cause higher epilimnion temperatures, earlier and more persistent thermal stratification, and increased thermal stability in the future. The Schmidt stability index used to evaluate the stratification strength showed tendency to increase, implying that the climate change may have considerable impacts on the water quality and ecosystem through changing the vertical mixing characteristics of the reservoir.