http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Tatsuto Kinjo,Tomonobu Senjyu,Naomitsu Urasaki,Hideki Fujita 전력전자학회 2004 ICPE(ISPE)논문집 Vol.- No.-
Utilizations of renewable energy are coming up from view points of environmental conservation and depletion of fossil fuel. However, the generated power from renewable energies is always fluctuating due to environmental status. Energy storage system is indispensable to compensate these fluctuating components. Superconducting magnetic energy storage (SMES) unit is useful for the compensation of fluctuating power since one is capable of controlling both active and reactive power simultaneously and quickly.<br/> This paper presents the series and parallel compensation using SMES (SPC-SMES), which consist of series inverter, parallel inverter, two-quadrant DC-DC converter, and superconductingcoil. The control systems for the active and reactive power of SPC-SMES are also proposes, and are verified the effectiveness through an instantaneous value analysis based computer simulations for case of wind power generation system.
GPU-based Acceleration of Minimum Vertex Covering for Large-scale Graphs
Kouta Toume,Hideki Kinjo,Morikazu Nakamura 대한전자공학회 2015 ITC-CSCC :International Technical Conference on Ci Vol.2015 No.6
Minimum vertex cover problems are one of the fundamental problems in graph theory and are known to be NP-hard. For structural data mining, we have proposed a GPU-based algorithm for minimum vertex cover problems. To derive sufficient parallelism for the many cores platform, the algorithm employs a distributed fashion for parallel operations, that is, GPU cores can basically perform their elemental operation only with their own data. Through the experimental evaluation, we demonstrate that our GPU algorithm implemented on both NVIDIA Fermi and Kepler GPU architectures is quite faster than a CPU version and the speedup becomes much evident when the graph size is enlarged.