Title :
Real-time power flow algorithm for shared memory multiprocessors for European distribution network types
Author :
Dzafic, I. ; Neisius, H.-T.
Author_Institution :
Siemens AG, Nuremberg, Germany
Abstract :
Size of the distribution and transmission networks to be planned or to be monitored in online systems is increasing significantly. The number of nodes of a complete distribution system exceeds up to several hundred thousand. As a consequence, efficient methods for the calculation of such a system are required. On the other hand, architecture and computation power of modern computers is not always used on efficient way. In this paper, an approach using Single Instruction Multiple Data (SIMD) instructions and multiprocessing is proposed, taking a symmetrically constructed but unbalanced loaded distribution network as an use case for a power flow calculation.
Keywords :
distributed shared memory systems; load flow; power distribution planning; power engineering computing; power transmission planning; distribution network; online systems; real-time power flow algorithm; shared memory multiprocessors; single instruction multiple data; transmission network planning; Computer architecture; Computers; Equations; Instruction sets; Load flow; Mathematical model; Symmetric matrices; Basic Linear Algebra Subprograms; C++; Multiprocessing; OpenMP; SIMD; Symmetrical Components; parallel programming; real-time;
Conference_Titel :
IPEC, 2010 Conference Proceedings
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-7399-1
DOI :
10.1109/IPECON.2010.5697097