• 전력전자학회:학술대회논문집
• /
• 전력전자학회 2004년도 전력전자학술대회 논문집(1)
• /
• pp.402-405
• /
• 2004

In this paper, the parallel operation of the IGBT and power stack for easy capacity enlargement series in the medium power capacity inverter system of the 660V voltage class is described. The parallel operation of the IGBT and power stack for 1.5MVA medium power inverter system's design is applied. The results of the parallel operation are described in this paper. The designed stack capacity for parallel operation is 800kVA class. For 1.5MVA inverter system, the 800kVA stack is applied with 2 parallel configurations. The 800kVA stack is designed with 3 parallel configurations of the IGBT Module. In this paper, the feasibility for easy capacity enlargement series in the medium power inverter by applying the parallel operation of the IGBT and power stack is verified. The experimental results show the good characteristics for the parallel operation of the IGBT and power stack.

• 정보처리학회 논문지
• /
• 제13권9호
• /
• pp.413-420
• /
• 2024

RISC-V는 오픈소스 명령어 집합 아키텍처로, 다양한 하드웨어 구현에서 사용되며, RV64I 기본 명령어 집합과 16개의 표준 확장을 통해 시스템 요구 사항에 맞게 유연하게 확장할 수 있다. 현재 RISC-V 아키텍처에서는 반환 주소를 보호하기 위해 쉐도우 스택 기법을 사용하고 있다. 본 논문에서는 RISC-V 아키텍처에서 컴팩트 쉐도우 스택 메커니즘과 병렬 쉐도우 스택 메커니즘의 성능을 SPEC CPU 2017 및 beebs 벤치마크를 사용하여 비교하였다. 실험 결과, 병렬 쉐도우 스택 메커니즘이 컴팩트 쉐도우 스택 메커니즘보다 더 높은 오버헤드를 보이는 것으로 나타났다. 이는 RISC-V 아키텍처의 한계로 인해 병렬 메커니즘의 효율성이 떨어짐을 시사하며, 따라서 컴팩트 쉐도우 스택이 RISC-V 아키텍처에 더 적합함을 보여준다. 또한 본 논문에서 기존 RISC-V 쉐도우 스택의 보안상 한계를 파악하고, RISC-V의 안전한 수행 환경을 보장하기 위해 쉐도우 스택 메커니즘의 성능과 보안성을 향상시키는 방향을 제시한다.

• 전자공학회논문지S
• /
• 제36S권2호
• /
• pp.88-95
• /
• 1999

스텍필터는 신호복원성이 뛰어난 디지털 비선형 필터의 일종이다. 그러나 기존의 적응형 설계기법을 이요하면 순차적인 특성 때문에 윈도우의 크기가 큰 스텍필터의 설계시 많은 계산량으로 그 응용에 한계가 있었다. 본 논문에서는 평균적대오차를 최소화하는 최적의 스텍필터를 설계하기 위한 병렬구조의 기법을 제안하였고 제안된 기법을 사용하면 적응특성 즉 학습의 반복적인 특성을 가지면서 최적의 필터로 수령하는데 필요한 반복 횟수를 줄임으로써 기존의 설계기법보다 설계시간을 단축할 수 있다는 사실과 제안된 기법이 최적의 스텍필터로 수렴한다는 사실을 증명하였다.

• 한국자동차공학회논문집
• /
• 제21권5호
• /
• pp.47-53
• /
• 2013

The wide application of disc springs to the designing of mechanical products with space limit is mainly attributable to their property of sustaining large axial load with small displacement. Due to the impediments in expecting the final results caused by the significant differences existing between a single unit and a stacked form, the force-displacement characteristics of a single disc spring and stacked disc springs are mainly examined in this study. In particular, the hysteresis of the series stack and the parallel stack will be investigated through the FE analysis and the analytical results will finally be compared with the acquired experimental data. In the final result, the analytical results were in accordance with the experimental data.

• 한국수소및신에너지학회논문집
• /
• 제21권5호
• /
• pp.405-411
• /
• 2010

A 5kW class SOFC cogeneration system consisted of a hot box part, a cold BOP (balance of plant) part, and a hot water reservoir. The hot box part contained a stack, a fuel reformer, a catalytic combustor, and heat exchangers. A cold BOP part was composed of blowers, pumps, a water trap, and system control units. A 5kW stack was designed to integrate 2 sub-modules. In this paper, the 5kW class SOFC system was operated using 2 short stacks connected in parallel to test the sub-module and the system. A short stack had 15 cells with $15{\times}15 cm^2$ area. When a natural gas was used, the total power was about 1.38 kW at 120A. Because the sub-modules were connected in parallel and current was loaded using a DC load, voltages of sub-modules were same and the currents were distributed according to the resistance of sub-modules. The voltage of the first stack was 11.46 V at 61A and the voltage of the second stack was 11.49V at 59A.

• 한국정보과학회논문지:시스템및이론
• /
• 제26권10호
• /
• pp.1185-1193
• /
• 1999

최적 스택 필터는 시그널 또는 영상의 임의의 특성 정보를 보존하고자 하는 요구조건에 의해 강제된 구조적 제약 하에서 최대의 잡음제거 효과를 얻을 수 있다. 그리고 임계치 분할 특성과 양의 부울 함수에 기반한 이진 영역에서의 처리 특성은 이 필터가 높은 병렬성을 갖고 있음을 보여준다. 본 논문에서는 두 개의 병렬 계산 모델 MCC(Mesh-Connected Computer)와 CCC(Cube-Connected Computer)에서 최적 스택 필터를 위한 1차원 병렬 알고리즘을 개발한다. 최적 스택 필터의 실행 시간은 주로 이진 median 연산에 의해 결정되고 본 논문에서 제안된 알고리즘은 선형 분리성에 의해 이 연산을 구현한다. 이를 바탕으로, M 레벨의 1-D 시그널의 길이가 L이고 윈도우 폭이 N이라고 가정할 때, 제안된 알고리즘은 {{{{root M times root M`` MCC에서 O(L sqrt{M}`) 시간에 그리고 M 개의 PE를 갖는 CCC에서 O(L log M)시간에 수행될 수 있다. 또한 잡음을 더욱 효과적으로 제거하기 위해 윈도우 폭 N을 증가시킬 때, 제안된 병렬 알고리즘의 계산 시간은 일정하게 유지됨을 보인다.Abstract An optimal stack filter achieves the maximum noise attenuation under the structural constraints imposed by the requirement of preserving certain signal or image features. And the filter provides a high parallelism due to the principles of threshold decomposition and binary processing based on positive Boolean functions(PBFs). In this paper, we develop an one-dimensional parallel algorithm for the optimal stack filter on two parallel computation models, MCC(Mesh-Connected Computer) and CCC(Cube-Connected Computer). The running time of the optimal stack filter depends mainly on the binary median operation and our algorithm realizes this operation by the linear separability. Based on this scheme, our parallel algorithm can be performed in {{{{O(L sqrt{M}`) MCC and inO(L log M) time on CCC with M PEs, when the length of M``-valued 1-D signal is L`` and window width is N`` Also, we show that the computation time of our parallel algorithm keeps constant when the window width N increases in order to achieve the best noise attenuation.

• 한국전기전자재료학회논문지
• /
• 제16권1호
• /
• pp.77-82
• /
• 2003

The ac loss is an important issue in the design of high-Tc superconducting power devices such as transformers and cables. In these devices many Bi-2223 tapes are closely stacked together and exposed to alternating magnetic fields that can have different orientations with respect to a tape. In such arrangement the magnetization loss is influenced by the screening current induced in adjacent tapes and thus different from that in a single tape. This stacking effect was experimentally investigated by measuring the magnetization loss in a stack, which consists of a number of tapes. First the magnetization loss in the single tape was measured in order to confirm the reliability of the loss data measured in the stack. The results for the single tape coincide well will the loss characteristics described in other previous works. For the stack In parallel and longitudinal magnetic fields the measured loss is Independent of both the number of tapes and stacking type. The longitudinal magnetization loss Is well explained rather by the slab model for decoupled filaments. For the tall stack in perpendicular field the measured loss at low fields is greatly decreased, compared to the loss of the single tape. However the loss at high fields is unaffected. These loss behaviors in the tall stack are well described by the slab model for full coupling.

• 한국초전도ㆍ저온공학회논문지
• /
• 제16권2호
• /
• pp.42-46
• /
• 2014

In this study we investigated ac transport current losses in the face to face stack for the anti-parallel current flow, and compared the electromagnetic properties with those of the single SC tape as well as those of the same stack for the parallel current path. The gap between the SC tapes in the stack varied in order to verify the electromagnetic influence of the neighbors when current flows in opposite direction, and the model was implemented in the finite element method program by the commercial software, COMSOL Multiphysics 4.2a. Conclusively speaking, the loss was remarkably decreased for the anti-parallel current case, which is attributed the magnetic flux compensation between the SC layers due to the opposite direction of the current flows. As the gap between SC tapes was increased, the loss mitigation became less effective. Besides, the current density distribution is very flat cross the sample width for the narrower gap case, which is believed to be benefit for the power electric system. These results are all in good agreement with those predicted theoretically for an infinite bifilar stack.

• 대한전기학회논문지:전기기기및에너지변환시스템부문B
• /
• 제52권7호
• /
• pp.315-319
• /
• 2003

We improved quench properties of a superconducting fault current limiter (SFCL) based on YBCO thin films by their serial and parallel combinations. The SFCL consisted of 6 switching elements fabricated of 4 inch-diameter YBCO thin films. The quench currents of the switching elements were distributed between 33.9 A and 35.6 A. Simple serial connection resulted in imbalanced power dissipation between switching elements even at the quench current difference of 0.6 A. On the other hand, $2{\times}2$ and $3{\times}2$ stack combinations produced simultaneous quenches. The $3{\times}2$ stack combination showed better simultaneous quench behavior than the $2{\times}2$ stacks. This is suggested to be because the currents between switching elements in parallel connection of the $3{\times}2$ stacks were more effectively redistributed than the $2{\times}2$ stacks.

• 한국유체기계학회 논문집
• /
• 제15권5호
• /
• pp.11-19
• /
• 2012

The main function of fuel cell manifold is to render reactants distribution as uniform as possible into a fuel cell stack. The purpose of this study is to numerically investigate the effects of stack manifold design on reactants distribution within a fuel cell stack. Four manifold designs with different manifold entrance shapes (expansion or diffuser) and different values of the extra width between the cell outer channel and manifold side wall are considered and applied to the fuel cell stack consisting of 50 cells. Since the fuel cell stack geometry involves several millions of grid points for numerical calculations, a parallel computing methodology is employed to substantially reduce the computational time and overcome the memory requirement. The numerical simulations are carried out and calculated results clearly demonstrate that both the manifold entrance shape and extra width have a substantial influence on manifold performance, controlling the degree of flow separation and entrance length for fully developed flow in the manifold channel. Finally, we suggest the optimum design of fuel cell manifold based on the simulation results.