Fine-Grained Fully Parallel Power Flow Calculation by Incorporating BBDF Method Into a Multistep NR Algorithm

In recognizing urgent needs in fast calculation of AC power flow (PF) problems, PF computation has been explored under different parallel computing platforms. Specifically, a block-bordered-diagonal form (BBDF) method has been widely studied to permute linear equations in PF calculations into a BBDF...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on power systems Jg. 33; H. 6; S. 7204 - 7214
Hauptverfasser: Su, Xueneng, Liu, Tianqi, Wu, Lei
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 01.11.2018
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Block bordered diagonal form
Computation
Iterative methods
Jacobian matrices
Linear equations
Load flow
Mathematical analysis
Mathematical model
Newton method
Newton-Raphson algorithm
parallel computation
Parallel processing
Power flow
Power system stability
Segmentation
ISSN:0885-8950, 1558-0679
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In recognizing urgent needs in fast calculation of AC power flow (PF) problems, PF computation has been explored under different parallel computing platforms. Specifically, a block-bordered-diagonal form (BBDF) method has been widely studied to permute linear equations in PF calculations into a BBDF form for facilitating parallel computation. However, determining an optimal network segmentation scheme that leads to the best speedup ratio of BBDF-based parallel PF is challenging. As a first contribution, this paper proposes a node-tearing-based approach to determine the optimal network segmentation scheme, which leverages sizes of subnetworks and the coordination network to achieve the best speedup ratio of BBDF-based parallel PF calculation. In addition, a fine-grained fully parallel PF approach is proposed to further enhance parallel performance, in which all three key steps of the Newton-Raphson based PF calculation are implemented in parallel. Studies illustrate effectiveness of the proposed network segmentation method and the fully parallel PF approach.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0885-8950
1558-0679
DOI:10.1109/TPWRS.2018.2834734