File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Article: A GPU-based Transient Stability Simulation using Runge-Kutta Integration Algorithm

TitleA GPU-based Transient Stability Simulation using Runge-Kutta Integration Algorithm
Authors
KeywordsTransient stability simulation
Runge-Kutta algorithm
Graphics processing units (GPU)
Parallel computing
Issue Date2013
PublisherEngineering and Technology Publishing. The Journal's web site is located at http://www.ijsgce.com/
Citation
International Journal of Smart Grid and Clean Energy, 2013, v. 2 n. 1, p. 32-39 How to Cite?
AbstractGraphics processing units (GPU) have been investigated to release the computational capability in various scientific applications. Recent research shows that prudential consideration needs to be given to take the advantages of GPUs while avoiding the deficiency. In this paper, the impact of GPU acceleration to implicit integrators and explicit integrators in transient stability is investigated. It is illustrated that implicit integrators, although more numerical stable than explicit ones, are not suitable for GPU acceleration. As a tradeoff between numerical stability and efficiency, an explicit 4th order Runge-Kutta integration algorithm is implemented for transient stability simulation based on hybrid CPU-GPU architecture. The differential equations of dynamic components are evaluated in GPU, while the linear network equations are solved in CPU using sparse direct solver. Simulation on IEEE 22-bus power system with 6 generators is reported to validate the feasibility of the proposed method.
Persistent Identifierhttp://hdl.handle.net/10722/189050
ISSN

 

DC FieldValueLanguage
dc.contributor.authorQin, Zen_US
dc.contributor.authorHou, Yen_US
dc.date.accessioned2013-09-17T14:24:48Z-
dc.date.available2013-09-17T14:24:48Z-
dc.date.issued2013en_US
dc.identifier.citationInternational Journal of Smart Grid and Clean Energy, 2013, v. 2 n. 1, p. 32-39en_US
dc.identifier.issn2315-4462-
dc.identifier.urihttp://hdl.handle.net/10722/189050-
dc.description.abstractGraphics processing units (GPU) have been investigated to release the computational capability in various scientific applications. Recent research shows that prudential consideration needs to be given to take the advantages of GPUs while avoiding the deficiency. In this paper, the impact of GPU acceleration to implicit integrators and explicit integrators in transient stability is investigated. It is illustrated that implicit integrators, although more numerical stable than explicit ones, are not suitable for GPU acceleration. As a tradeoff between numerical stability and efficiency, an explicit 4th order Runge-Kutta integration algorithm is implemented for transient stability simulation based on hybrid CPU-GPU architecture. The differential equations of dynamic components are evaluated in GPU, while the linear network equations are solved in CPU using sparse direct solver. Simulation on IEEE 22-bus power system with 6 generators is reported to validate the feasibility of the proposed method.-
dc.languageengen_US
dc.publisherEngineering and Technology Publishing. The Journal's web site is located at http://www.ijsgce.com/-
dc.relation.ispartofInternational Journal of Smart Grid and Clean Energyen_US
dc.rightsCreative Commons: Attribution 3.0 Hong Kong License-
dc.subjectTransient stability simulation-
dc.subjectRunge-Kutta algorithm-
dc.subjectGraphics processing units (GPU)-
dc.subjectParallel computing-
dc.titleA GPU-based Transient Stability Simulation using Runge-Kutta Integration Algorithmen_US
dc.typeArticleen_US
dc.identifier.emailHou, Y: yhhou@eee.hku.hken_US
dc.identifier.authorityHou, Y=rp00069en_US
dc.description.naturepublished_or_final_version-
dc.identifier.doi10.12720/sgce.2.1.32-39-
dc.identifier.hkuros222526en_US
dc.identifier.volume2en_US
dc.identifier.issue1en_US
dc.identifier.spage32en_US
dc.identifier.epage39en_US
dc.publisher.placeUnited States-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats