| | |
Summary: 448 IEEETransactionson Powa Systems,Vol. 10. No. 1, February 1995
ALGORITHMSFOR EFFICIENTVECTORIZATION OF
REPEATED SPARSEPOWER SYSTEMNETWORK COMPUTATIONS
Cevdet Aykanat, Member Ozlem Ozgu
Computer EngineeringDepartment
Bilkent University
Ankara, Turkey
Abstract-Standardsparsity-basedalgorithms used in power sys-
tem appllcations need to be restructuredfor efficientvectorization
due to the extremely short vectors processed. Further, intrinsic
architectural features of vector computers such as chaining and
sectioningshould also be exploited for utmost performance. This
paper presentsnovel data storageschemes and vectorizationalso-
r i m that resolve the recurrenceproblem, exploit chaining and
minimizethe number of indirectelementselectionsin the repeated
solution of sparse linear system of equations widely encountered
in various power system problems. The proposed schemes are
alsoapplied and experimentedfor the vectorizationof power mis-
matchcalculationsarisingin the solution phaseofFDLF which in-
volves typical repeated sparse power networkcomputations. The
|
