• 한국소음진동공학회논문집
• /
• 제18권9호
• /
• pp.972-981
• /
• 2008

The blower which is installed in a FCEV(fuel cell electric vehicle) may cause noise due to misalignment and unbalance of mechanical components that rotate at high speed. One of the key points in efforts to minimize the noise radiation from a blower is the knowledge of the main radiating component and the relation between the surface vibration of a blower and the sound pressure. In this research, the blower model is developed based on FEM(finite element method). FE(finite element) model is reliable by correlation of frequencies and MAC(modal assurance criterion) values between EMA(experimental modal analysis) and FEA(finite element analysis). This model is applied to predict the vibration of a blower by using inverse force identification method and predict the radiating noise by using BEM(boundary element method). Comparing the frequencies of resonance and those mode shapes between EMA and FEA, a structural modification of the FE model is evaluated for reducing the parameters of the blower noise.

• Structural Engineering and Mechanics
• /
• 제31권4호
• /
• pp.439-451
• /
• 2009

The paper deals with the problem related to the modelling of riveted assemblies for crashworthiness analysis of full-scale complete aircraft structures. Comparisons between experiments and standard FE computations on high-energy accidental situations onto aluminium riveted panels show that macroscopic plastic strains are not sufficiently localised in the FE shells connected to rivet elements. The main reason is related to the structural embrittlement caused by holes, which are currently not modelled. Consequently, standard displacement FE models do not succeed in initialising and propagating the rupture in sheet metal plates and along rivet rows as observed in the experiments. However, the literature survey show that it is possible to formulate super-elements featuring defects that both give accurate singular strain fields and are compatible with standard displacement finite elements. These super-elements can be related to the displacement model of the hybrid-Trefftz principle of the finite element method, which is a kind of domain decomposition method. A feature of hybrid-Trefftz finite elements is that they are mainly used for elastic computations. It is thus proposed to investigate the possibility of formulating a hybrid displacement finite element, including the effects of a hole, dedicated to crashworthiness analysis of full-scale aeronautic structures.

• 한국농공학회:학술대회논문집
• /
• 한국농공학회 2000년도 학술발표회 발표논문집
• /
• pp.246-252
• /
• 2000

This paper presents an explanatory study of combining the finite element and boundary element methods to achieve an efficient and accurate analysis of frame structure containing shear wall. This model analyzes the frame by finite element method and the shear wall by boundary element method. The purpose of this study is the specific case that boundary element is surrounded by finite element. If material properties of shear wall are relatively the very smaller than it of frame structure, the displacement shape of each node is calculated exactly. And if the solution of displacement is the larger, the displacement shape is approximated more accurately.

• 한국정밀공학회지
• /
• 제13권9호
• /
• pp.165-173
• /
• 1996

The dimensional accuracy of a final product is mainly affected by elastic die deformation during the forging and elastic recovery after the ejection in cold forging process. The investigations on elastic recovery are not so much as those of elastic die deformation. The elastic recovery can be determined by using the elastic-plalstic finite element analysis, but, this method has some limits such as poor conver- gence and long computational time, etc. In this paper, a theoretical analysis for predicting the elastic recovery of a final product in axi-symmetric forging process by using the rigid-plastic finite element method is presented. The rigid-plastic finite element analysis of a cold forward extrusion process involving loading, ejecting process is accomplished by rigid-plastic FE code, DEFORM. The effect of elastic die deformation on the final product dimenmsion is also considered. The calculated elastic recovery is compared is compared with the analysis result of elastic-plastic FE code. ABAQUS.

• Structural Engineering and Mechanics
• /
• 제12권2호
• /
• pp.201-214
• /
• 2001

Four finite element (FE) models are examined to find the one that best estimates moment-rotation characteristics of top- and seat-angle with double web-angle connections. To efficiently simulate the real behavior of connections, finite element analyses are performed with following considerations: 1) all components of connection (beam, column, angles and bolts) are discretized by eight-node solid elements; 2) shapes of bolt shank, head, and nut are precisely taken into account in modeling; and 3) contact surface algorithm is applied as boundary condition. To improve accuracy in predicting moment-rotation behavior of a connection, bolt pretension is introduced before the corresponding connection moment being surcharged. The experimental results are used to investigate the applicability of FE method and to check the performance of three-parameter power model by making comparison among their moment-rotation behaviors and by assessment of deformation and stress distribution patterns at the final stage of loading. This research exposes two important features: (1) the FE method has tremendous potential for connection modeling for both monotonic and cyclic loading; and (2) the power model is able to predict moment-rotation characteristics of semi-rigid connections with acceptable accuracy.

• Computers and Concrete
• /
• 제23권6호
• /
• pp.399-408
• /
• 2019

Structural health monitoring is important for the safety of lives and asset management. In this study, numerical models were developed for the piezoresistive behavior of smart concrete based on finite element (FE) method. Finite element models were calibrated with experimental data collected from compression test. The compression test was performed on smart concrete cube specimens with 75 mm dimensions. Smart concrete was made of cement CEM II 42.5 R, silica fume, fine and coarse crushed limestone aggregates, brass fibers and plasticizer. During the compression test, electrical resistance change and compressive strain measurements were conducted simultaneously. Smart concrete had a strong linear relationship between strain and electrical resistance change due to its piezoresistive function. The piezoresistivity of the smart concrete was modeled by FE method. Twenty-noded solid brick elements were used to model the smart concrete specimens in the finite element platform of Ansys. The numerical results were determined for strain induced resistivity change. The electrical resistivity of simulated smart concrete decreased with applied strain, as found in experimental investigation. The numerical findings are in good agreement with the experimental results.

• 한국음향학회지
• /
• 제33권3호
• /
• pp.191-199
• /
• 2014

파이프, 평판과 같이 단면의 형상이 길이 방향으로 일정한 도파관 구조물을 따라 전파되는 진동의 반사 및 투과 특성은 여러 공학 분야에서 응용되는 중요한 주제이다. 도파관에 조인트 또는 균열 등의 국부적 불연속이 있는 경우, 스펙트럴 요소(spectral element)와 유한 요소(finite elment)를 결합한 SE/FE 방법이 주로 사용되고 있다. 그러나 이 방법은 보 이론에 기반한 스펙트럴 요소가 사용되므로 저주파수 대역 해석에 국한되는 단점이 있다. 고주파수 대역 해석에는 스펙트럴 수퍼 요소(spectral super element)와 유한 요소를 결합한 SSE/FE 방법이 제안되었으나 유한요소와 스펙트럼 요소의 연성으로 인해 많은 연산 시간이 요구된다. 이러한 문제점을 개선하고자, 본 연구에서는 국부적 불연속 구간의 단면이 일정한 경우에 대해 국부적 불연속 구간을 스펙트럴 수퍼 요소로 대체한 SSE/SSE 연성 해석을 시도하였다. 적용 모델로는 국부적 결함을 가진 레일의 파동 반사 및 투과, 그리고 주기적 보강재를 가진 평판의 진동전파에 대해 적용하였다. 결함을 가진 레일의 해석 예를 통해, 본 논문에서 사용한 SSE/SSE 방법과 기존의 SSE/FE 방법의 성능을 비교하였다. 보강재를 가진 평판의 예를 통해서는 반복 구조를 가진 도파관의 파동 전파 특성 해석에 SSE/SSE 방법이 유용함을 확인하였다.

• 대한전기학회논문지:전기기기및에너지변환시스템부문B
• /
• 제49권7호
• /
• pp.457-461
• /
• 2000

The finite element method (FEM) is suitable for the analysis of a complicated region that includes nonlinear materials, whereas the boundary element method (BEM) is naturally effective for analyzing a very large region with linear characteristics. Therefore, considering the advantages in both methods, a novel algorithm for the alternate application of the FEM and BEM to magnetic field problems with the open boundary is presented. This approach avoids the disadvantages of the typical numerical methods with the open boundary problem such as a great number of unknown values for the FEM and non-symmetric matrix for the Hybrid FE-BE method. The solution of the overall problems is obtained by iterative calculations accompanied with the new acceleration method.

• 동력기계공학회지
• /
• 제11권4호
• /
• pp.78-85
• /
• 2007

In order to analyze accurately the vibration of a structure by using the finite element method (FEM), we have to model a analytical structure as a numerical model with many degrees-of-freedom. However, in this case, the FEM needs much computation time and storage. The authors developed the finite element-transfer stiffness coefficient method (FE-TSCM) for overcoming the drawback of the FEM. In this paper, the authors apply the FE-TSCM to the in-plane vibration analysis of general plates with various shapes. Two numerical examples, a rectangular plate and a triangular plate, are used to compare the results of the FE-TSCM and the FEM. Through the numerical calculation, we confirm that the FE-TSCM can be applied to the in-plane free or forced vibration analysis of the general plates with various shapes and is effective to in-plane vibration analysis of general plates.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2000년도 춘계학술대회논문집
• /
• pp.1583-1588
• /
• 2000

The dynamic characteristics of a HDD suspension system are investigated by finite element analysis and experimental modal analysis. A finite element model of the suspension Type850 was developed for unloaded case. The calculated vibration modes were compared with measurements and agree well in shape and frequency except some local modes. Local thickness and Young's modulus of the finite element model are updated by modal tuning method to develop the precise FE model. A sensitivity matrix of the natural frequencies for some design variables was calculated using finite difference method. Most natural frequencies calculated by the tuned FE model coincide with the measurements and the errors between them are less than 2%.