• Journal of Internet Computing and Services
• /
• v.25 no.5
• /
• pp.121-130
• /
• 2024

Tabular Data is a mixture of numerical and categorical data, and machine learning models have been evaluated to be more suitable than generative models in performing learning using such tabular data. This evaluation is because the generative model had a problem of excessively increasing parameters or not finding the direction of learning due to the numerical multimodal distribution and categorical frequency imbalance, which are characteristics of Tabular Data. However, as data gradually becomes big data and becomes real-time, existing machine learning models have shown limitations in their application. In this paper, as a methodology for applying generative models to tabular data, we propose RGGAN (Reinforced Generator GAN), a reinforced generator adversarial neural network that Clustering sampling that leverages conjugate prior distributions and the loss function improved with Gower coefficients and mutual information. As a result of measuring the AUC by detecting fraudulent transactions in the IEEE-CIS Fraud Detection Dataset by constructing an anomaly detector with the discriminators learned from the RGGAN proposed in this paper, it showed a performance improvement effect of 1-7% over the existing generative models, proving that the proposed model is effective for learning tabular data and also effective in detecting fraudulent transactions.

• Journal of Advanced Marine Engineering and Technology
• /
• v.33 no.4
• /
• pp.502-508
• /
• 2009

To obtain high power, diesel engines continuously increase combustion pressure and mean effective pressure each cylinder, and the excitation sources and noisy sources are increased, too. Moreover, to reduce the costs, shipyards make hull structures weaker than before. As above reasons, it is more difficult to control the vibration phenomenon nowadays. In this study, it was investigated why diesel generator sets reached the vibration allowable limits during the FAT and heavy vibration phenomenon of diesel generator sets using ODS test during onboard tests. Also, it is found out the stiffness of deck and common bed using the test result of their structural impedance. To find out the vibratory characteristics of diesel generator sets, model tests were carried out. From the sensitivity analysis after above tests, it was selected points to be reinforced and studied troubleshooting to solve heavy vibration phenomenon of diesel generator sets.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2009.04a
• /
• pp.504-508
• /
• 2009

To obtain high power, diesel engines continuously increase combustion pressure and mean effective pressure each cylinder, and the excitation sources and noisy sources are increased, too. Moreover, to reduce the costs, shipyards make hull structures weaker than before. As above reasons, it is more difficult to control the vibration phenomenon nowadays. In this study, it was investigated why diesel generator sets reached the vibration allowable limits during the FAT and heavy vibration phenomenon of diesel generator sets using ODS test during onboard tests. Also, it is found out the stiffness of deck and common bed using the test result of their structural impedance. To find out the vibratory characteristics of diesel generator sets, MODAL tests were carried out. From the sensitivity analysis after above tests, it was selected points to be reinforced and studied troubleshooting to solve heavy vibration phenomenon of diesel generator sets.

• Transactions of the Korean Society of Pressure Vessels and Piping
• /
• v.4 no.2
• /
• pp.29-33
• /
• 2008

As the measures of nuclear power plant utilities and manufacturers to reduce the defects of tube expansion region during manufacturing steam generators, many types of NDEs(Non-Destructive Examinations) are conducted to inspect the expansion region. The expansion region of tube is subject to degrade because of stress concentration induced by tube expansion, sludge pile and high temperature. So the inspections for tube expansion region have been reinforced. Liquid penetrant test, helium leak test, Bobbin profile test and hydraulic test are performed to confirm the integrity of tube expanded by hydraulic expansion method. Liquid penetrant test and helium leak test are used to inspect seal weld region on tubesheet end part. Bobbin Profile test is used to inspect fully the expanded region of steam generator tube. Hydraulic test finally verifies the integrity of seal weld region on tubesheet end part.

• Journal of Korean Association for Spatial Structures
• /
• v.17 no.1
• /
• pp.49-58
• /
• 2017

The result of the previous work leads to the idea that the inner area of the hyperbolic shell generator should be minimized for the cooling tower with higher first natural frequency. In this study the inner area of the hyperbolic shell generator was graphically established under varying height of the throat and angle of the base lintel. From the graph, several shell geometries were selected and analysed in the aspect of the natural frequency. Three representative towers reinforced differently due to different first natural frequencies were analysed non-linearly and evaluated using a damage indicator based on the change of natural frequencies. The results demonstrated that the damage behaviour of the tower reinforced higher due to a lower first natural frequency was not necessarily advantageous than the others.

• Nuclear Engineering and Technology
• /
• v.27 no.3
• /
• pp.327-335
• /
• 1995

The normal shutdown and maintenance period of a nuclear power plant can be remarkably shortened when the examination and maintenance works in steam generator tubes are simultaneously carried out with refueling job. There are nozzle dams to Hock the coolant How from reactor to steam generator. Workers are reluctant to install nozzle dam because of the high radiation exposure and the limited working space in steam generator. Moreover, the heavy weight of present nozzle dam makes it installation and removal works much difficult. In this paper, a lighter KAERI nozzle dam with increased flexural rigidity-to-weight was designed and manufactured by changing the structure design of the present nozzle dam and by selecting new material, carbon fiber-reinforced plastic.

• ETRI Journal
• /
• v.32 no.6
• /
• pp.972-975
• /
• 2010

In this letter, we consider the lossy coding of a non-uniform binary source based on GF(q)-quantized low-density generator matrix (LDGM) codes with check degree $d_c$=2. By quantizing the GF(q) LDGM codeword, a non-uniform binary codeword can be obtained, which is suitable for direct quantization of the non-uniform binary source. Encoding is performed by reinforced belief propagation, a variant of belief propagation. Simulation results show that the performance of our method is quite close to the theoretic rate-distortion bounds. For example, when the GF(16)-LDGM code with a rate of 0.4 and block-length of 1,500 is used to compress the non-uniform binary source with probability of 1 being 0.23, the distortion is 0.091, which is very close to the optimal theoretical value of 0.074.

• Structural Engineering and Mechanics
• /
• v.51 no.6
• /
• pp.1047-1065
• /
• 2014

The article presents the methodology for finding material damping capacity at higher frequency and at relatively lower amplitudes. The Lamb wave dispersion theory and loss less finite element model is used to find the damping capacity of composite materials. The research has been focused on high frequency applications materials. The method was implemented on carbon fiber reinforced polymer (CFRP) and glass fiber reinforced polymer (GFRP) plates. The Lamb waves were generated using ultrasonic pulse generator setup. The hybrid method has been explored in this article and the results have been compared with bandwidth methods available in the literature.

• Computers and Concrete
• /
• v.1 no.1
• /
• pp.15-30
• /
• 2004

In this study, a design process for reinforced concrete structures using the nonlinear FEM analysis is developed. Instead of using the nonlinear analysis to evaluate the required performance after design process, the nonlinear analysis is applied before designing the reinforcement arrangement inside the RC structures. An automatic reinforcement generator for computer aided reinforcement agreement is developed for this purpose. Based on a nonlinear FEM program for analyzing the reinforced concrete structure, a smart fictitious material model of steel, is proposed which can self-adjust the reinforcement to the required amount at the cracking location according to the load increment. Using this tool, the reinforcement ratio required at design load level can be decided automatically. In this paper, an example of RC beam with opening is used to verify the proposed process. Finally, a trial design process for a real size underground RC LNG tank is introduced.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.19 no.5
• /
• pp.76-81
• /
• 2011

Exhaust gas recirculation is the well-known and widely used NOx reduction technology for diesel engines. More effective EGR cooler has been developed and applied to diesel engines to meet the reinforced emission regulation. However, the contaminated EGR cooler by diesel exhaust gas reduces the performance of the engine and NOx reduction rate. The buildup of deposits in EGR coolers cause significant degradation in heat transfer performance, often on the order of 20~30%. Deposits also increase pressure drop across coolers and thus may degrade engine efficiency under some operation conditions. In this study, as a solution for this problem, DOC assisted EGR cooler is designed and then investigated to reduce fouling and its impact on cooler performance. A single channel EGR cooler fouling test apparatus and soot particle generator were developed to represent the real EGR cooler and exhaust gas of diesel engine. EGR cooler effectiveness of the case with catalyst of pt 30g/ft3 decreased just up to 5%. This value was 45% less compared to the case without catalyst which decreased up to 9% after 10hours experiments.