• Journal of Korea Foundry Society
• /
• v.41 no.6
• /
• pp.521-527
• /
• 2021

This study presents a process management method for the detection of casting defects during in high-pressure die casting based on machine learning. The model predicts the defects of the next cycle by extracting the features appearing over the previous cycles. For design of the gearbox, the proposed model detects shrinkage defects with data from three cycles in advance with 98.9% accuracy and 96.8% recall rates.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.16 no.5
• /
• pp.164-170
• /
• 2008

A typical clutch system for automotive manual transmissions transfers hydraulic pressure generated by driver's pedal manipulation to the clutch diaphragm spring. The foot effort history during the period of push is different than the period of the clutch pedal's return. The effort or load difference is called clutch foot effort hysteresis. It is known that the hysteresis is caused by friction. The frictional force and moment are produced between various component contact points such as between the rubber seal and the inner wall inside the hydraulic cylinder and between the diaphragm spring and the pressure plate, etc. Understanding the clutch pedal foot effort hysteresis is essential for a clutch release system design and analysis. The dynamic model for a clutch release system is developed for the foot effort hysteresis prediction and a simulation analysis is performed to propose a tool for analysing a clutch system.

• Journal of Advanced Marine Engineering and Technology
• /
• v.27 no.4
• /
• pp.467-478
• /
• 2003

Automotive shock absorber may not be regarded as only one(simple) damping machine because it is composed of many components, and shows non-linear damping characteristics. No matter how advanced form of shock absorber is developed, the oil shock absorber can not be neglected. because their structures are based on the oil shock absorber. Therefore it is essential to accurately analyze the dynamic characteristics of oil shock absorber. It stands mainly roi damper valve tuning which nowadays is still exhaustively done by means of ride work. In this study, damping mechanism and dynamic characteristics for oil shock absorber of twin tube type are analyzed, based on the mathematical model considering internal flow and pressure. For the reliability of numerical prediction. the database is constructed within the limit of adequate reliability. Finally, the programmed system that gives out necessary specification by inputting damping specification and tolerance is to be constructed.

• Proceedings of the Korean Nuclear Society Conference
• /
• 1997.05a
• /
• pp.409-414
• /
• 1997

Reflood model in RELAP5/MOD3.1 are modified to improve the unrealistic prediction results of the model. In the new method, the modified Zuber pool boiling critical heat flux (CHF) correlation is adopted. The reflood drop size is characterized by the use of We=1.5 and the minimum drop size of 0.0007 m for $p^{*}\;{\leq}\;0.025$. To describe the wall to vapor heat transfer at low pressure and low flow condition, the Webb-Chen correlation is utilized . The suggested method has been verified through the simulations of the Lehigh University rod bundle reflood tests. Through sensitivity study it is shown that the effect of drag coefficients is dominant in the reflood model. It is proved that the present modifications result in much more improved quench behavior and accurate wan and vapor temperature predictions.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2011.11a
• /
• pp.985-987
• /
• 2011

This paper includes a development of a starting time prediction model for a derivative engine. For this derivative engine design, a new map expansion method, Modified Pump Scaling Law(MPS), has been applied and expand the maps to sub-idle range. From loss characteristics of the reference engine, loss models for the derivative engine have been developed considering different pressure, temperature, and engine configurations. Starting time predictions of the derivative engine shows preferable results comparing test results.

• Proceedings of the Korean Nuclear Society Conference
• /
• 1997.10a
• /
• pp.546-551
• /
• 1997

This paper presents a prediction of critical heat flux (CHF) in bubbly flow regime using dry-spot model proposed recently by authors for pool and flow boiling CHF and existing correlations for forced convective heat transfer coefficient, active site density and bubble departure diameter in nucleate boiling region. Without any empirical constants always present in earlier models, comparisons of the model predictions with experimental data for upward flow of water in vertical, uniformly-heated round tubes are performed and show a good agreement. The parametric trends of CHF have been explored with respect to variations in pressure, tube diameter and length, mass flux and inlet subcooling.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.26 no.6
• /
• pp.823-832
• /
• 2002

A modified $textsc{k}$-$\varepsilon$model is proposed for calculation of transitional boundary-layer flows with changing pressure gradient. In order to develop the model for this problem, the flow is divided into three regions; pre-transition region, transition region and fully turbulent region. The effect of pressure gradient is taken into account in stream-wise intermittency factor, which bridges the eddy-viscosity models in the pre-transition region and the fully turbulent region. From intermittency data in various flows, Narashima＇s intermittency function, F(${\gamma}$), has been found to be proportional to $\chi$$^{n}$ according to the extent of pressure gradient. Three empirical correlations of intermittency factor being analyzed, the best one was chosen to calculate three benchmark cases of bypass transition flows with different free-stream turbulence intensity under arbitrary pressure gradient. It was found that the variations of skin friction and shape factor as well as the profiles of mean velocity in the transition region were very satisfactorily predicted.

• Nuclear Engineering and Technology
• /
• v.55 no.12
• /
• pp.4607-4616
• /
• 2023

We implement machine learning regression models to predict peak pressures of primary and secondary systems, a major safety concern in Loss Of Condenser Vacuum (LOCV) accident. We selected the Multi-dimensional Analysis of Reactor Safety-KINS standard (MARS-KS) code to analyze the LOCV accident, and the reference plant is the Korean Optimized Power Reactor 1000MWe (OPR1000). eXtreme Gradient Boosting (XGBoost) is selected as a machine learning tool. The MARS-KS code is used to generate LOCV accident data and the data is applied to train the machine learning model. Hyperparameter optimization is performed using a simulated annealing. The randomly generated combination of initial conditions within the operating range is put into the input of the XGBoost model to predict the peak pressure. These initial conditions that cause peak pressure with MARS-KS generate the results. After such a process, the error between the predicted value and the code output is calculated. Uncertainty about the machine learning model is also calculated to verify the model accuracy. The machine learning model presented in this paper successfully identifies a combination of initial conditions that produce a more conservative peak pressure than the values calculated with existing methodologies.

• 한국신재생에너지학회:학술대회논문집
• /
• 2011.11a
• /
• pp.56.1-56.1
• /
• 2011

The solar and meteorological resources map is calculated using by one-layer solar radiation model (GWNU model), satellites data and numerical model output on the Korean peninsula. The Meteorological input data to perform the GWNU model are retrieved aerosol optical thickness from MODIS (TERA/AQUA), total ozone amount from OMI (AURA), cloud fraction from geostationary satellites (MTSAT-1R) and temperature, pressure and total precipitable water from output of RDAPS (Regional Data Assimilation and Prediction System) and KLAPS (Korea Local Analysis and Prediction System) model operated by KMA (Korea Meteorological Administration). The model is carried out every hour using by the meteorological data (total ozone amount, aerosol optical thickness, temperature, pressure and cloud amount) and the basic data (surface albedo and DEM). And the result is analyzed the distribution in time and space and validated with 22 meteorological solar observations. The solar resources map is used to the solar energy-related industries and assessment of the potential resources for solar plant. The National Institute of Meteorological Research in KMA released $4km{\times}4km$ solar map in 2008 and updated solar map with $1km{\times}1km$ resolution and topological effect in 2010. The meteorological resources map homepage (http://www.greenmap.go.kr) is provided the various information and result for the meteorological-solar resources map.

• Fibers and Polymers
• /
• v.6 no.3
• /
• pp.269-276
• /
• 2005

In this study, cross-cultural comparison of sound sensation for Korean traditional silk fabrics between Korea and America was performed and prediction models for sound sensation by objective measurements including sound parameters such as level pressure of total sound (LPT), Zwicker's psychoacoustic characteristics, and mechanical properties by Kawabata Evaluation System were established for each nation to explore the objective parameters explaining sound sensation of the Korean traditional silk. As results, Koreans felt the silk fabric sounds soft and smooth while Americans were revealed as perceiving them hard and rough. Both Koreans and Americans were pleasant with sounds of Gongdan and Newttong and especially Newttong was preferred more by Americans in terms of sound sensation. In prediction models, some of subjective sensation were found as being related mainly with mechanical properties of traditional silk fabrics such as surface and compressional characteristics.