http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
용탕주조법을 이용한 금속복합재료 제조공정의 열전달 해석
정창규,정성욱,남현욱,한경섭,Jung, Chang-Kyu,Jung, Sung-Wook,Nam, Hyun-Wook,Han, Kyung-Seop 대한기계학회 2002 大韓機械學會論文集A Vol.26 No.10
A finite element model is developed for the process of squeeze casting of metal matrix composites (MMCs) in cylindrical molds. The fluid flow and the heat transit. are fundamental phenomena in squeeze casting. To describe heat transfer in the solidification of molten aluminum, the energy equation is written in terms of temperature and enthalpy are applied in an axisymmetric model which is similar to the experimental system. A one dimensional flow model simulates the transient metal flow. A direct iteration technique was used to solve the resulting nonlinear algebraic equations, using a computer program to calculate the enthalpy, temperature and fluid velocity. The cooling curves and temperature distribution during infiltration and solidification were calculated fer pure aluminum. Experimentally, the temperature was measured and recorded using thermocouple wire. The measured time-temperature data were compared with the calculated cooling curves. The resulting agreement shows that the finite element model can accurately estimate the solidification time and predict the cooling process.
pSDF와 이진 결합 변환 상관기를 이용한 광 패턴 인식에 관한 연구
정창규,조동래,길상근,박한규,Jung, Chang-Kyoo,Cho, Dong-Rae,Gil, Sang-Keun,Park, Han-Kyu 대한전자공학회 1990 전자공학회논문지 Vol. No.
본 논문에서는 pSDF(projection synthetic discriminant function)공간 불변 필터 개념을 적용하여 기준 이미지를 구현하고, 공간 평면 상관기인 이진 결합 변환 상관기를 이용하여 동일 클래스 인식 및 클래스 판별을 위한 광 패턴 인식을 수행하였다. 컴퓨터 시뮬레이션 결과, 이진 결합 변환 상관기가 기존의 결합 변환 상관기보다 상관 첨두치 세기, 상관 첨두치 세기대 부로브비, 신호대 잡음비, 상관폭 부분에서 뛰어난 상관 특성을 보였다. 이진 결합 변환 상관기를 이용한 광 패턴 인식 실험을 한 결과, 동일 클래스 인식인 경우 $4.1~9.6{\%}$ 오차 범위 내에서 상관 첨두지 세기가 일정하게 나타났으며, 클래스 판별인 경우 두 클래스간의 상관 첨두지 세기가 2배 차이가 나서 판별 능력이 우수함을 알 수 있었다. In this paper, pSDF-based referance image is realized. Using BJTC (binary joint transform correlator) as the spatial plane correlator, optical pattern recognition for interclass identification and interclass discrimination is performed. Computer simulation shows that the correlation performance of BJTC is superior to that of JTC. Experimental results using BJTC reveal that correlation peak intensity is constant within the error rang from $4.1{\%}\to\9.6{\%}$ in interclass identification and correlation peak intensity of one class is two times higher than that of the other class in interclass discrimination, which indicates its superiority in discrimination sensitivity.
CAE 데이터베이스를 이용한 연주기 몰드의 탕면 변동 예측
정창규(Chang-Kyu Jung),이승환(Seunghwan Lee),김연호(Youn Ho Kim) 대한기계학회 2013 대한기계학회 춘추학술대회 Vol.2013 No.12
A neural network program predicting the molten steel surface fluctuation of continuous casting process from a CAE database was developed. The data sets were built from the results of flow simulation in the casting mold. Total 48 simulation cases were carried out with different input parameters. Among them, 45 cases were used for training the data mining model and 3 cases were used for validating the model. The developed neural network program predicts the maximum vertical displacement of the molten steel surface. The prediction process automatically carries out with previously stored data in the database and the program modifies the database with newly calculated results. The predicted values from neural network method showed less error than those from regression analysis method did.