Parallel harmonic-oriented method for large-scale electric systems based on MPI programming and the limit cycle method

An efficient harmonic-oriented method based on parallel processing and the limit cycle method to compute the periodic, steady-state solution of large scale electric systems is presented in this paper. The time domain numeric solution is based on Newton methods and the Poincare map to accelerate the...

Full description

Saved in:
Bibliographic Details
Published in:2005 IEEE Russia Power Tech pp. 1 - 6
Main Author: Garcia, N.
Format: Conference Proceeding
Language:English
Published: IEEE 01.06.2005
Subjects:
Acceleration
Concurrent computing
Convergence of numerical methods
FACTS
Harmonics
Large-scale systems
limit cycle
Limit-cycles
MPI programming
Newton method
Newton methods
Numerical Differentiation
Parallel processing
Parallel programming
Poincaré map
Power Quality
Steady-state
Transient response
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:An efficient harmonic-oriented method based on parallel processing and the limit cycle method to compute the periodic, steady-state solution of large scale electric systems is presented in this paper. The time domain numeric solution is based on Newton methods and the Poincare map to accelerate the convergence to the limit cycle. Furthermore, the implementation of parallel processing using MPI programming makes possible to speed up more the computations in the time domain. A three-phase version of the IEEE 118-node system is used to locate the limit cycle and study the transient response during a single-phase fault including a DVR-STATCOM. Simulations are implemented using a PC-Linux cluster with 856 processors. This approach allows the implementation of harmonic and power quality analyses, which help to understand the real impact of the operation of FACTS controllers on the power network.
DOI:10.1109/PTC.2005.4524460