DocumentCode :
1378856
Title :
Static security analysis using pipeline decomposition
Author :
Alves, A.B. ; Monticelli, A.
Author_Institution :
Univ. Estadual de Campinas, Sao Paulo, Brazil
Volume :
145
Issue :
2
fYear :
1998
fDate :
3/1/1998 12:00:00 AM
Firstpage :
105
Lastpage :
110
Abstract :
A pipeline approach to static security analysis is described. Security analysis is decomposed into data preparation, contingency selection, contingency analysis and result output and distributed in a network of workstations. Contingency selection and contingency analysis functions are further decomposed and solved in parallel by groups of processors. An important feature of the pipeline is fault tolerance (the online replacement of a faulted station). Moreover. The numbers of stations in the various stages of the pipeline may be dynamically adjusted to cope with changes in workload conditions. Excellent results have been obtained particularly regarding the way efficiency varies with the number of processors. It has been observed that saturation in speedup curves occurs later than observed with other decomposition procedures (e.g. data decomposition) under similar conditions. The new methods have been implemented and tested using a public domain software system (PVM). Tests have been carried out with the IEEE 300-bus system and the Brazilian Southern-Southeastern system with 810 buses
Keywords :
data preparation; fault tolerant computing; load flow; pipeline processing; power system analysis computing; power system security; public domain software; 810 bus system; Brazilian Southern-Southeastern system; IEEE 300-bus system; PVM software; contingency analysis; contingency selection; data decomposition; data preparation; distributed computing; fault tolerance; functional decomposition; network computing; pipeline decomposition; power flow analysis; power systems; public domain software system; speedup curves; static security analysis;
fLanguage :
English
Journal_Title :
Generation, Transmission and Distribution, IEE Proceedings-
Publisher :
iet
ISSN :
1350-2360
Type :
jour
DOI :
10.1049/ip-gtd:19981785
Filename :
675519
Link To Document :
بازگشت