• Journal of the Korean Statistical Society
• /
• 제29권2호
• /
• pp.187-199
• /
• 2000

In this paper, we deal with the asymptotic properties of the least absolute deviation estimators in the nonlinear time series regression model. For the sinusodial model which frequently appears in a time series analysis, we study the strong consistency and asymptotic normality of least absolute deviation estimators. And using the derived limiting distributions we show that the least absolute deviation estimators is more efficient than the least squared estimators when the error distribution of the model has heavy tails.

• 설비공학논문집
• /
• 제29권10호
• /
• pp.515-529
• /
• 2017

EnergyPlus, which is widely used in various fields, provides Simple Window Model, a window model that can be used practically. However, the results of building load using the model are different from those of the standard model. The main cause of the deviation by Simple Window Model was analyzed to be due to the assumption that all windows were considered as single layer. The purpose of this study is to propose a window model that improves the cause of deviation by Simple Window Model and can be easily calculated from the algebraic relations. The proposed window model solved the heat balance equation algebraically by using seven window characteristic coefficients. The coefficient relationships consisted of the heat transmission coefficient and solar heat gain coefficient as input parameters make practical use and calculation possible. As a result of comparing the deviation between each window model by implementing the dynamic analysis method, the proposed window model showed that the deviation of the total heating/cooling energy consumption was reduced to 1/3 compared to Simple Window Model for one year. Although the maximum energy consumption did not show any significant improvement, the indoor temperature evaluation showed significantly reduced deviation.

• 대한인간공학회지
• /
• 제26권1호
• /
• pp.71-77
• /
• 2007

Shoe's size and shape are determined by the last that takes shape of foot because last is the mold of shoe in development and manufacturing process. Then adaptation between foot and shoe is dependent on the last. In mass shoe production, model size is developed in the first place, other sized lasts are made through the grading process based on model size. The most important factor in grading system is grading deviation that must be same amount induced from foot measuring database. At present, most of the last manufacturing companies in korea using 260mm as a standard foot model size. When length grading deviation is 5mm, the ball girth grading deviation is 3.7mm and the ball width grading deviation is 1.2mm. I verified existing grading system by comparing grading results with foot measuring data. Also, I proposed reasonable grading deviation and application method of grading system. From the analysis of foot measuring database, reasonable grading deviations are 1.22mm in ball width and 0.84mm in ankle height in case of length grading deviation is 5mm. I confirmed that the current grading system is very accurate. When we grade last from 230mm to 290mm by current grading system based on model size 260mm, there is grading error over 1mm in the front outside area of foot. This error level of 1mm is no problem in normal walking shoe's last, but it induces adaptation problems in sports and special purposed shoe's last. Therefore using of three standard model size is recommended in grading men's last for reducing grading deviation error under the level of 1mm. It is specifically described as 235mm in 225-245mm, 260mm in 250-270mm, 285mm in 275-295mm. According to the above recommended grading system, it is enough to measure only three foot sizes in case of foot measuring project for men's last development.

• Journal of applied mathematics & informatics
• /
• 제18권1_2호
• /
• pp.649-656
• /
• 2005

In this paper we consider a fuzzy least absolute deviation method in order to construct fuzzy linear regression model with fuzzy input and fuzzy output. We also consider two numerical examples to evaluate an effectiveness of the fuzzy least absolute deviation method and the fuzzy least squares method.

• 설비공학논문집
• /
• 제20권7호
• /
• pp.478-485
• /
• 2008

The extreme steam temperature deviation experienced in the superheater of a tangentially fired boiler can seriously affect its economic and safe operation. This temperature deviation is one of the main causes of boiler tube failures. The steam temperature deviation is mainly due to the thermal load deviation in the lateral direction of the superheater. The thermal load deviation consists of several causes. One of the causes is the non-uniform heat flow distribution of burnt gas on the superheater tube system. This distribution is very difficult to measure in situ using direct experimental techniques. So, we need thermal load model to estimate the tube temperature. In this paper, we propose a thermal load distribution model by using CFD analysis and plant data. We successfully predict the tube temperature and the steam flow rate in a final superheater system from the thermal load model and one dimensional heat-flow system analysis. The proposed model and analysis method would be valuable in preventing the frequent tube failure of the final superheater tubes.

• Korean Journal of Mathematics
• /
• 제7권2호
• /
• pp.301-313
• /
• 1999

In this paper we obtain similar limit theorems of the Generalized Curie - Weiss model for a new class Hamiltonian. We expressed the saddlepoint approximation by large deviation rate and then obtain the limit theorems.

• 대한설비공학회:학술대회논문집
• /
• 대한설비공학회 2007년도 동계학술발표대회 논문집
• /
• pp.589-593
• /
• 2007

In the mechanical ventilation system, it is a fundamental condition to distribute the air equally to the each room. In this study, distribution performance of the air distributor which generally connected to a circular duct was investigated by simulation and experiment. In the first CFD analysis, maximum air flow rate deviation was an 63% in the air distributor model. After numbers of model modification and simulation, maximum flow rate deviation was reduced to 19% in the final simulation model. An air distributor which used in the experiment was produced by using data obtained from the final analysis. When experimental result was compared with analysis result, there was a deviation difference as much as 9%.

• 한국도로학회논문집
• /
• 제18권4호
• /
• pp.63-68
• /
• 2016

PURPOSES : The nonlinear model of fatigue cracking is typically used for determining the maintenance period. However, this requires that the model parameters be known. In this study, the particle filter (PF) method was used to determine various statistical parameters such as the mean and standard deviation values for the nonlinear model of fatigue cracking. METHODS : The PF method was used to determine various statistical parameters for the nonlinear model of fatigue cracking, such as the mean and standard deviation. RESULTS : On comparing the values obtained using the PF method and the least square (LS) method, it was found that PF method was suitable for determining the statistical parameters to be used in the nonlinear model of fatigue cracking. CONCLUSIONS : The values obtained using the PF method were as accurate as those obtained using the LS method. Furthermore, reliability design can be applied because the statistical parameters of mean and standard deviation can be obtained through the PF method.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2011년 춘계학술대회논문집
• /
• pp.230-235
• /
• 2011

COMS (Communication, Ocean and Meteorological Satellite) is a geostationary satellite and was developed by KARI for communication, ocean and meteorological observations. COMS was tested under vacuum and very law temperature conditions in order to correlate thermal model and to verify thermal design. The test was performed by using KARI large thermal vacuum chamber. The COMS S/C thermal model was successfully correlated versus the 2 thermal balance test phases. After model correlation, temperatures deviation of all individual unit were less than $5^{\circ}C$ and global deviation and standard deviation also satisfied the requirements, less than $2^{\circ}C$ and $3^{\circ}C$. The final flight prediction was performed by using the correlated thermal model.

• 한국전산유체공학회지
• /
• 제16권3호
• /
• pp.59-65
• /
• 2011

COMS (Communication, Ocean and Meteorological Satellite) is a geostationary satellite and was developed by KARI for communication, ocean and meteorological observations. COMS was tested under vacuum and very low temperature conditions in order to correlate thermal model and to verify thermal design. The test was performed by using KARI large thermal vacuum chamber. The COMS S/C thermal model was successfully correlated versus the 2 thermal balance test phases. After model correlation, temperatures deviation of all individual units were less than $5^{\circ}C$ and global deviation and standard deviation also satisfied the requirements, less than $2^{\circ}C$ and $3^{\circ}C$. The final flight prediction was performed by using the correlated thermal model.