• 대한기계학회논문집A
• /
• 제20권4호
• /
• pp.1225-1232
• /
• 1996

In this paper the vibration and power flow analysis for the branched piping system conveying fluid are performed by wave approach. The uniform straight pipe element conveying fluid is formulated using the dynamic stiffness matrix by wave approach. The branched piping system conveying fluid can be easily formulated with considering of simple assumptions of displacements at the junction and continuity conditions of the pipe internal flow. The dynamic stiffness matrix for each uniform straight pipe element can be assembled by using the global assembly technique using in conventional finite element method. The computational method proposed in this paper can easily calculate the forced responses and power flow of the branched piping system conveying fluid regardless of finite element size and modal properties.

• 신재생에너지
• /
• 제16권1호
• /
• pp.15-24
• /
• 2020

For mechanical load testing of wind turbines, capture matrix is constructed for various range of wind speeds according to the international standard IEC 61400-13. The conventional method wastes considerable amount of data by its invalid data policy -segment data into 10 minutes then remove invalid ones. Previously, we have suggested an alternative way to save the total amount of data to build a capture matrix, but the efficient selection of data has been still under question. The paper introduces optimization algorithms to construct capture matrix with less data. Heuristic algorithm (simple stacking and lowest frequency first), population method (particle swarm optimization) and Q-Learning accompanied with epsilon-greedy exploration are compared. All algorithms show better performance than the conventional way, where the distribution of enhancement was quite diverse. Among the algorithms, the best performance was achieved by heuristic method (lowest frequency first), and similarly by particle swarm optimization: Approximately 28% of data reduction in average and more than 40% in maximum. On the other hand, unexpectedly, the worst performance was achieved by Q-Learning, which was a promising candidate at the beginning. This study is helpful for not only wind turbine evaluation particularly the viewpoint of cost, but also understanding nature of wind speed data.

• 수산해양교육연구
• /
• 제21권2호
• /
• pp.325-334
• /
• 2009

The aims of this study is to suggest the strategic development plan of Ulsan Port using SWOT/AHP method. In order to do this, 16 detailed factors for SWOT Matrix were identified both from previous studies and from brainstorming professionals of the port and these made into a 3 level hierarchy structure. AHP method identified the relative weight of SWOT level as threat(0.311), opportunity(0.259), strength(0.218) and weakness(0.212), also the composite relative weight of detailed factors as recession of global economy(0.125), strengthening of economic function in port(0.085), insufficiency of port facilities(0.080) and a government regulation of port management(0.075). As a result, the study suggests S/T strategy, W/T strategy, S/O strategy, W/O strategy as a strategic development plan of the Ulsan Port.

• 한국전산구조공학회논문집
• /
• 제20권6호
• /
• pp.731-742
• /
• 2007

본 연구에서는 새로운 비선형해석 알고리즘인 적응형 Newton-Raphson 반복기법을 제안한다. 제안된 기법은 기존 Newton-Raphson 기법을 근간으로 적응형 부구조물화 기법을 이용하여 강성등가하중을 구하고, 이미 역행렬이 계산되어 있는 초기강성행렬에 강성등가하중을 적용하여 보정변위를 구하는 것으로 요약된다. 제안된 알고리즘의 가장 큰 특징은 하중 구간의 수에 관계없이 구조물 강성행렬에 대한 역행렬 계산을 단 한번만 수행한다는 것이다. 제안된 기법의 효율성은 강성행렬 및 역행렬 계산 후 부재강성행렬이 변경된 부재들이 연결된 자유도 수와 전체 자유도 수의 비율에 직접 관계된다. 이 비율에 따라 제안된 기법을 기존 비선형해석 기법과 보완적으로 사용함으로써 전체 비선형해석 효율을 향상시킬 수 있다.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2002년도 추계종합학술대회
• /
• pp.226-230
• /
• 2002

본 논문은 영상 모자이킹을 위한 사영 변환 행렬의 정밀도를 개선하는 방법을 제안한다. Shift Theorem을 이용하여 영상간의 전역 이동 성분을 추출하고 구해진 이동 성분을 이용하여 영상간의 일치점을 찾음으로서 기존의 일치점 검출의 문제점을 해결하였다. 일치점 집합을 RANSAC 알고리즘을 적용하여 신뢰도가 높은 일치점을 추출함으로서 잡음의 영향에 강인하도록 하였으며 일치점의 좌표를 정규화 하여 사영변환 행렬의 정밀도를 개선하였다.

• Earthquakes and Structures
• /
• 제14권6호
• /
• pp.589-598
• /
• 2018

An attempt is conducted to explore the relationship between the macroscopic global damage and the local damage of shear-type RC frames. A story damage index, which can be expressed as multi-variate functions of modal parameters, is deduced based on the tridiagonal matrix of the shear-type frame. The global damage model is also originated from structural modal parameters. Due to the connection of modal damage indexes, the relationship between the macroscopic global damage and the local story damage is reasonably established. In order to validate the derivation, a case study is carried out via an 8-story shear-type frame. The sensitivities of modal damage indexes to the location and severity of local story damages are studied. The evolution of the global damage is investigated as well. Results show that the global damage is sensitive to the degree of story damage, but it's not sensitive to its location. As the number of the damaged stories increases, more and more modes will be involved. Meanwhile, the global damage evolution curve changes from the concave shape to the S-type and then finally transforms into the convex shape. Through the proposed story damage, modal damage and global damage model, a multi-level damage assessment method is established.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 C
• /
• pp.1195-1198
• /
• 1999

This paper presents optimal placement of minimal set of phasor measurement units(PMU's) and observability of measurement system with PMU. By using the incidence matrix symbolic method which directly assigns measurement and pseudo-measurement to incidence matrix, it is much simpler and easier to analyze observability. The optimal PMU set is found through the simulated-annealing(SA) and the direct combinational method. The cooling schedule parameter which is suitable to the property of problem to solve is specified and optimal placement is proven by presented direct combinational method. Search spaces are limited within reasonable feasible solution region to reduce a unnecessary one in the SA implementation based on global search. The proposed method presents to save CPU time and estimate state vectors based on optimal PMU set.

• 한국자동차공학회논문집
• /
• 제5권6호
• /
• pp.44-52
• /
• 1997

As part of the effort to reduce the weight of powertrain, a hollow crankshaft has been designed. The mass reduction of the crankshaft changes the dynamic properties of the crankshaft such as moment of inertia, and torsional, bending stiffness. The purpose of this paper is to compare the dynamic behavior of the hollow crankshaft with that of the original, solid crankshaft. Global dynamic behavior of the crankshaft is analyzed bgy the transfer matrix method(TMM). The crankshaft has been modeled by 38 lumped mass and stiffness elements. The dynamic patameters of each lumped element are provided by Finite Element Method(FEM). The responses of the crankshaft from TMM are fed back as loading conditions to the Finite Element model to obtain dynamic stresses for critical areas of the crankshaft.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제13권4호
• /
• pp.267-280
• /
• 2009

In this paper we investigate a simple two-level additive Schwarz preconditioner for the P1 symmetric interior penalty Galerkin method of the Poisson equation on rectangular meshes. The construction is based on the decomposition of the global space of piecewise linear polynomials into the sum of local subspaces, each of which corresponds to an element of the underlying mesh, and the global coarse subspace consisting of piecewise constants. This preconditioner is a direct combination of the block Jacobi iteration and the cell-centered finite difference method, and thus very easy to implement. Explicit upper and lower bounds for the maximum and minimum eigenvalues of the preconditioned matrix system are derived and confirmed by some numerical experiments.

• Transactions on Control, Automation and Systems Engineering
• /
• 제4권4호
• /
• pp.324-329
• /
• 2002

In this paper, the guaranteed cost filtering design method for linear time delay systems with convex bounded uncertainties in discrete-time case is presented. The uncertain parameters are assumed to be unknown but belonging to known convex compact set of polytotype less conservative than norm bounded parameter uncertainty. The main purpose is to design a stable filter which minimizes the guaranteed cost. The sufficient condition for the existence of filter, the guaranteed cost filter design method, and the upper bound of the guaranteed cost are proposed. Since the proposed sufficient conditions are LMI(linear matrix inequality) forms in terms of all finding variables, all solutions can be obtained simultaneously by means of powerful convex programming tools with global convergence assured. Finally, a numerical example is given to check the validity of the proposed method.