• 전자공학회논문지 IE
• /
• v.48 no.2
• /
• pp.19-25
• /
• 2011

This paper presented neural network modeling method using vehicle data to predict fuel consumption. To acquire data for training and testing the proposed neural network, medium-class gasoline vehicle drove at downtown and parameters measured include speed, engine rpm, throttle position sensor (TPS), and mass air flow (MAF) as input data, and fuel consumption as target data from OBD-II port. Multi layer perception network was used for nonlinear mapping between the input and the output data. It was observed that the neural network model can predict the vehicle quite well with mean squared error was $1.306{\times}10^{-6}$ for the fuel consumption.

• Fibers and Polymers
• /
• v.1 no.2
• /
• pp.103-110
• /
• 2000

Flow analysis of profile extrusion is essential for design and production of a profile extrusion die. Velocity, pressure, and temperature distribution in an extrusion die are predicted and compared with the experimental results. A two dimensional numerical method is proposed for three dimensional analysis of the flow field within the profile extrusion die by applying a modified cross-sectional numerical method. Since the cross-sectional shape of the die is varied gradually, it is assumed that the pressure is constant within a cross-sectional plane that is perpendicular to the flow direction. With this assumption, the velocity component in the cross-sectional direction is neglected. The exact cross-sectional shape at any position is calculated based on the geometry of standard cross-sections. The momentum and energy equations are solved with proper boundary conditions at a cross-section and then the same calculation is carried out for the next cross-section using the current calculated values. An L-shaped profile extrusion die is produced and employed for experimental investigation using a commercially available polypropylene. Numerical prediction for the varying cross-sectional shape provides better results than the previous studies and is in good agreement with the experimental results.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.13 no.4
• /
• pp.1-8
• /
• 2009

Accumulated slag mass was predicted to estimate accurate performance of kick motor (KM) system. The validation of numerical analysis was performed with mass flow rate measured at the 4th ground test of the KM. The study described here includes the internal flow field of KM at various time steps during burning. Slag mass accumulation was analyzed through the aluminum oxide particle paths to predict slag mass deposition. Numerical analysis to solve both flow field and droplet accumulation was performed with Fluent 6.3 program. Analysing the effects of the acceleration, starting position and diameters of the aluminum oxide particles, total slag mass accumulation was obtained.

• Journal of the Society of Naval Architects of Korea
• /
• v.48 no.1
• /
• pp.15-22
• /
• 2011

In this paper, cavitation patterns of model tests were compared with those of full-scale measurement for a propeller of crude oil carrier which was suffered from erosions on suction side of blade tip region. Cavitation tests were performed at design and ballast draft using model and full scale nominal wakes. A model ship and wire mesh method was used for the simulation of wake patterns of model nominal wakes. For the prediction of full-scale wake patterns, a RANS solver(Fluent 6.3) was used and wire mesh method was used for the simulation of the full scale wakes. Comparison results show that cavitation patterns using predicted full-scale wake patterns are closer to cavitation patterns of full-scale measurement at ballast draft condition. Also, cloud cavitations were observed on the position of eroded area at both full-scale measurement and cavitation tests using simulated full-scale wake patterns.

• The KSFM Journal of Fluid Machinery
• /
• v.7 no.5 s.26
• /
• pp.20-28
• /
• 2004

Static and dynamic structural analyses of a turbine bladed-disk for a liquid rocket turbopump are performed to investigate the safety level of strength and vibration at design point. During operation, turbopump is exposed to various external loads. Therefore, the effects of them should be carefully considered and properly modeled. First, due to the high rotational speed of the turbopump, effects of centrifugal forces are considered in the structural analysis. Thermal load caused by severe temperature differences is also considered. A three dimensional finite element method (FEM) is used for linear and nonlinear structural analyses with modified Newton-Raphson iteration method. After the nonlinear solution is obtained from the structural analysis, dynamic characteristics are obtained as a function of rotational speed from the linearized eigenvalue analysis at an equilibrium position. From the analysis results, characteristics of stress distribution and vibration were thoroughly examined and investigated.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1021-1023
• /
• 2003

Atmospheric water vapor is a key variable in numerical weather prediction (NWP) models, but it is a crucial factor to limit the accuracy of high-precision GPS positioning technique. For both issues, knowledge about the amount of water vapor is extremely important. In this study, we perform a simulation study to utilize GPS signals through a developed tomographic scheme to retrieve 3D structure of atmospheric wet refractivity, which may be assimilated into NWP models for advancing forecasting or position calculation for improving GPS positioning accuracy. For the purpose of knowing the absolute accuracy of the developed tomographic method, a well-defined temporal and spatial varying state of atmospheric profile is utilized. Under such circumstance, several factors that may influence the retrievals can be easily examined and their impacts may be clearly quantified. They include the values of the positional dilution of precision (PDOP) factors of the GPS signals, ... etc. Based upon the use of a variety spectrum of adjustable factors, many interesting findings are obtained. For example, the more is the number of the observed GPS signals the better becomes the retrievals as expected. Also, the smaller is the PDOP value the better becomes the retrievals.

• Journal of Industrial Technology
• /
• v.39 no.1
• /
• pp.41-45
• /
• 2019

Hydrogen sulfide ($H_2S$) is mainly produced along with methane and hydrocarbons in many gas fields as well as hydrodesulfurization processes of crude oils containing sulfur compounds and the emission of $H_2S$ has a considerable effect on both environmental problem and human health aspects due to formation of, e.g. acid rain and smog. In recent years, ionic liquids (ILs) have been proposed as the most promising solvents for $CO_2$ and hazardous pollutants capture, such as $H_2S$ and sulfur dioxide ($SO_2$). In this work, we demonstrate the use of the predictive COSMO-SAC model for the prediction of Henry's law constant of $H_2S$ in ILs. Furthermore, the method is used to screen for potential IL candidates for $H_2S$ capture from a set of 2,624 ILs formed from 82 cations and 32 anions. The effects of cation on the Henry's law constant of $H_2S$ such as (i) the variation of the alkyl chain length on cation, (ii) the substituent of methyl group ($-CH_3$) for H in C(2) position and (iii) the change of ring structure for cation family are clearly predicted by COSMO-SAC model.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.7 no.3
• /
• pp.435-451
• /
• 2015

Evaluation of the performance of aging structures is essential in the oil and gas industry, where the inaccurate prediction of structural performance can have significantly hazardous consequences. The effects of structure failure due to the significant reduction in wall thickness, which determines the burst strength, make it very complicated for pipeline operators to maintain pipeline serviceability. In other words, the serviceability of gas pipelines and elbows needs to be predicted and assessed to ensure that the burst or collapse strength capacities of the structures remain less than the maximum allowable operation pressure. In this study, several positions of the corrosion in a subsea elbow made of API X42 steel were evaluated using both design formulas and numerical analysis. The most hazardous corrosion position of the aging elbow was then determined to assess its serviceability. The results of this study are applicable to the operational and elbow serviceability needs of subsea pipelines and can help predict more accurate replacement or repair times.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.12 no.1
• /
• pp.75-82
• /
• 2000

This paper presents the predicted results by CFD technique of air flow and contaminant distribution in a full-scale kitchen opened to a living room, ventilated by a exhaust hood. To analyze the characteristics of the indoor environment, the concept of contaminant index was defined. In this study, the locations of the gas range and the window were chosen as the parameters to investigate the indoor environment. The values of the contaminant index for several layout of the gas range and the window were calculated and compared. When the gas range is installed along the wall with specified window location, its position in relation to the wall has unnoticed effect on contaminant infer. Once the location of the gas range is fixed, the indoor air quality may deteriorate by the proximity of the window to the gas range. This is due to the shorter distance that external fresh air must travel within the kitchen before it reaches the exhaust fan.

• Nuclear Engineering and Technology
• /
• v.41 no.9
• /
• pp.1181-1190
• /
• 2009

Knowing more about the Local Power Density (LPD) at the hottest part of a nuclear reactor core can provide more important information than knowledge of the LPD at any other position. The LPD at the hottest part needs to be estimated accurately in order to prevent the fuel rod from melting in a nuclear reactor. Support Vector Machines (SVMs) have successfully been applied in classification and regression problems. Therefore, in this paper, the power peaking factor, which is defined as the highest LPD to the average power density in a reactor core, was estimated by SVMs which use numerous measured signals of the reactor coolant system. The SVM models were developed by using a training data set and validated by an independent test data set. The SVM models' uncertainty was analyzed by using 100 sampled training data sets and verification data sets. The prediction intervals were very small, which means that the predicted values were very accurate. The predicted values were then applied to the first fuel cycle of the Yonggwang Nuclear Power Plant Unit 3. The root mean squared error was approximately 0.15%, which is accurate enough for use in LPD monitoring and for core protection that uses LPD estimation.