• 한국산학기술학회논문지
• /
• 제20권12호
• /
• pp.94-101
• /
• 2019

본 연구에서는 초음파 용접기의 진동에너지를 모재에 전달하는 금속 용접용 혼의 형상설계를 위해 해석적 기법을 적용하였으며, 20kHz 영역에서 사용되는 한파장 바 형상을 갖는 용접용 혼의 형상설계를 위해 FEM(Finite Element Method)을 사용하였다. 봉의 종진동 이론을 Ansys APDL(Ansys Parametric Design Language)에 적용하여 혼의 형상을 최적화하였다. 적절한 혼의 형상을 도출하기 위해 진동의 입력면과 출력면의 면적비와 혼의 축방향 길이비를 이용하여 총 25가지의 모델을 설계하였으며, Modal 해석과 Harmonic 해석을 통해 균일한 진동특성을 확보하고자 하였다. 스텝 바혼의 끝단부(팁)의 균일도 90%이상을 확보하기 위해 Harmonic 해석을 통해 길이비, 면적비를 결정하였고 이에 따라 혼의 총길이 130mm, 스텝 길이 65mm, 출력단 면적 28.79mm의 혼을 설계하였다. 설계치수를 바탕으로 Titanium (Ti-6Al-4V)-GR5 재료를 이용하여 혼을 제작하였다. 또한, 별도로 구성된 초음파 용접시스템을 이용하여 혼의 끝단부의 진동 진폭률 및 변위특성을 평가하였다. 최적화된 혼의 진폭률은 51%까지 향상됨을 확인하였다. 최종적으로 균일한 종방향 스텝 바혼을 설계하였으며, 끝단부(팁)의 균일도 97.4%이상을 확보하였다.

• 한국공간구조학회논문집
• /
• 제20권2호
• /
• pp.51-58
• /
• 2020

Recently, deep learning that is the most popular and effective class of machine learning algorithms is widely applied to various industrial areas. A number of research on various topics about structural engineering was performed by using artificial neural networks, such as structural design optimization, vibration control and system identification etc. When nonlinear semi-active structural control devices are applied to building structure, a lot of computational effort is required to predict dynamic structural responses of finite element method (FEM) model for development of control algorithm. To solve this problem, an artificial neural network model was developed in this study. Among various deep learning algorithms, a recurrent neural network (RNN) was used to make the time history response prediction model. An RNN can retain state from one iteration to the next by using its own output as input for the next step. An eleven-story building structure with semi-active tuned mass damper (TMD) was used as an example structure. The semi-active TMD was composed of magnetorheological damper. Five historical earthquakes and five artificial ground motions were used as ground excitations for training of an RNN model. Another artificial ground motion that was not used for training was used for verification of the developed RNN model. Parametric studies on various hyper-parameters including number of hidden layers, sequence length, number of LSTM cells, etc. After appropriate training iteration of the RNN model with proper hyper-parameters, the RNN model for prediction of seismic responses of the building structure with semi-active TMD was developed. The developed RNN model can effectively provide very accurate seismic responses compared to the FEM model.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 추계학술대회논문집
• /
• pp.1120-1124
• /
• 2007

Recently in the automotive brake industry brake squeal noise has become one of the top automotive quality warranty issues. The contact pressure is used to predict friction coupling in the brake squeal analysis. The formulation of friction coupling has performed by nonlinear static analysis prior to the complex eigenvalue analysis. This paper proposes a validation methodology of squeal analysis using modal testing and contact analysis and examines the effect of predicted contact pressure that leads to the discrepancy between unstable complex mode and squeal frequency. This studies compose a three step validation process : examining the modal characteristics of component and assembly loaded contact pressure using modal testing and FEM analysis and verifying the contact pressure distribution using nonlinear static analysis and experiment. Finally, the unstable modes from complex eigenvalue analysis and realistic squeal frequency from the noise dynamometer are investigated.

• 한국소음진동공학회논문집
• /
• 제21권6호
• /
• pp.499-505
• /
• 2011

This paper presents optimal design of controllable magnetorheological suspension unit for a tracked vehicle. As a first step, a double-rod type MR suspension unit is designed on the basis of the Bingham model of commercially available MR fluid, and its damping characteristics are evaluated with respect to the intensity of the magnetic field. Subsequently, the governing equation of motion of the MR suspension system featuring the MR valve is established. Then, the optimization problem to find optimal geometric dimensions of the MR supension unit is formulated by considering an objective function which is related to damping torque and control energy. The first order optimization method intergrated with a commercial finite element method(FEM) software is adopted to obtain optimal solution of the system. The performance characteristics of the optimized MR susepnsion unit is then evaluated and compared with initial one.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2007년도 춘계학술대회 논문집
• /
• pp.160-164
• /
• 2007

In the steel industry, there is a need to produce large forged parts for the automobile industries, the flight and shipping industries ad military industries. In the steel-industry application, a cogging technique for cast ingots is required, because the major parts are needed as one large body in order to obtain higher quality. Therefore, cogging process is the primary step in manufacturing of practically large open-die forging. In the cogging process, internal voids have to be eliminated as defects, The present work is concerned with the elimination of the internal voids in large ingots so as obtain sound products. In this study, hot compression tests were carried out to obtain the flow stress of cast microstructure at different temperature and strain rates. The FEM analysis are performed to investigate the overlap defect of cast ingots during cogging stage. The measure flow stress data were used to simulate the cogging process of cast ingot using the practical material properties. Also the analysis of void closure are performed by using the $DEFORM^{TM}$-3D. The calculated results of void closure behavior are compared with the measured results before and after cogging, which are scanned by the X-ray scanner. From this result, the criteria for deformation amounts effect on the void closure can be investigated by the comparison of practical experiment and numerical analysis.

• 전기전자학회논문지
• /
• 제6권2호
• /
• pp.136-145
• /
• 2002

정상상태에서 소자 내부의 격자온도 분포를 해석할 수 있는 시뮬레이터를 개발하였다. Slotboom 변수를 사용하여 열흐름 방정식을 이산화하였고, 요소내에서 열전도율의 적분값을 해석적으로 구할 수 있는 방법을 제안하였다. 제안된 방법의 타당성을 검증하기 위해 $N^+P$ 접합 다이오드와 BJT에 대해 모의실험을 수행하였고 DAVINCI와 MEDICI의 결과와 비교하였다. $N^+P$ 접합 다이오드에서 순방향 인가전압이 1.4[V]일 때 격자 온도분포는 DAVINCI의 결과와 2%의 상대오차를 보였으며 BJT에서 컬렉터 전압이 5.0[V]이고 베이스 전압이 0.5[V]일 때 격자 온도분포는 MEDICI의 결과와 3%의 상대오차를 보였다. BANDIS에서 제안된 열전도율의 적분방법을 사용하는 경우 수렴을 위해 평균 3.45회의 행렬 연산이 필요하나 DAVINCI에서는 평균 5.1회의 행렬 연산이 필요하였고 MEDICI는 평균 4.3회 행렬연산이 필요하였다.