Transient stability assessment using artificial neural networks

A. R. Edwards; K. W. Chan; R. W. Dunn; A. R. Daniels

Transient stability assessment using artificial neural networks

A. R. Edwards, K. W. Chan, R. W. Dunn, A. R. Daniels

The Hong Kong Polytechnic University

Research output: Unpublished conference presentation (presented paper, abstract, poster) › Conference presentation (not published in journal/proceeding/book) › Academic research › peer-review

Abstract

This paper discusses a method for fast transient stability assessment of large interconnected power systems. Composite indices, such as the minimum post-contingency busbar voltage magnitude and sum of the changes in rotor angles, provide an effective method for reducing the dimensionality of feature vectors for transient stability classification. Typically, less than 20 composite indices are required to construct a feature vector for a contingency, which can then be classified using an artificial neural network into a transiently stable or unstable contingency. Simulation results are presented for an IEEE test network as well as a reduced model of the UK National Grid System, and the application of this technique to contingency screening in an Energy Management System is discussed.

Original language	English
Pages	252-255
Number of pages	4
Publication status	Published - Sept 1994
Event	Proceedings of the 29th Universities Power Engineering Conference. Part 2 (of 2) - Galway, Irel Duration: 14 Sept 1994 → 16 Sept 1994

Conference

Conference	Proceedings of the 29th Universities Power Engineering Conference. Part 2 (of 2)
City	Galway, Irel
Period	14/09/94 → 16/09/94

ASJC Scopus subject areas

General Engineering

Cite this

@conference{55e739394b874bd0af803f14985c44e3,

title = "Transient stability assessment using artificial neural networks",

abstract = "This paper discusses a method for fast transient stability assessment of large interconnected power systems. Composite indices, such as the minimum post-contingency busbar voltage magnitude and sum of the changes in rotor angles, provide an effective method for reducing the dimensionality of feature vectors for transient stability classification. Typically, less than 20 composite indices are required to construct a feature vector for a contingency, which can then be classified using an artificial neural network into a transiently stable or unstable contingency. Simulation results are presented for an IEEE test network as well as a reduced model of the UK National Grid System, and the application of this technique to contingency screening in an Energy Management System is discussed.",

author = "Edwards, {A. R.} and Chan, {K. W.} and Dunn, {R. W.} and Daniels, {A. R.}",

year = "1994",

month = sep,

language = "English",

pages = "252--255",

note = "Proceedings of the 29th Universities Power Engineering Conference. Part 2 (of 2) ; Conference date: 14-09-1994 Through 16-09-1994",

}

Transient stability assessment using artificial neural networks. / Edwards, A. R.; Chan, K. W.; Dunn, R. W. et al.
1994. 252-255 Paper presented at Proceedings of the 29th Universities Power Engineering Conference. Part 2 (of 2), Galway, Irel.

Research output: Unpublished conference presentation (presented paper, abstract, poster) › Conference presentation (not published in journal/proceeding/book) › Academic research › peer-review

TY - CONF

T1 - Transient stability assessment using artificial neural networks

AU - Edwards, A. R.

AU - Chan, K. W.

AU - Dunn, R. W.

AU - Daniels, A. R.

PY - 1994/9

Y1 - 1994/9

N2 - This paper discusses a method for fast transient stability assessment of large interconnected power systems. Composite indices, such as the minimum post-contingency busbar voltage magnitude and sum of the changes in rotor angles, provide an effective method for reducing the dimensionality of feature vectors for transient stability classification. Typically, less than 20 composite indices are required to construct a feature vector for a contingency, which can then be classified using an artificial neural network into a transiently stable or unstable contingency. Simulation results are presented for an IEEE test network as well as a reduced model of the UK National Grid System, and the application of this technique to contingency screening in an Energy Management System is discussed.

AB - This paper discusses a method for fast transient stability assessment of large interconnected power systems. Composite indices, such as the minimum post-contingency busbar voltage magnitude and sum of the changes in rotor angles, provide an effective method for reducing the dimensionality of feature vectors for transient stability classification. Typically, less than 20 composite indices are required to construct a feature vector for a contingency, which can then be classified using an artificial neural network into a transiently stable or unstable contingency. Simulation results are presented for an IEEE test network as well as a reduced model of the UK National Grid System, and the application of this technique to contingency screening in an Energy Management System is discussed.

UR - http://www.scopus.com/inward/record.url?scp=0028733816&partnerID=8YFLogxK

M3 - Conference presentation (not published in journal/proceeding/book)

AN - SCOPUS:0028733816

SP - 252

EP - 255

T2 - Proceedings of the 29th Universities Power Engineering Conference. Part 2 (of 2)

Y2 - 14 September 1994 through 16 September 1994

ER -

Transient stability assessment using artificial neural networks

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this