Bootstrap-based confidence interval estimates for electromechanical modes from multiple output analysis of measured ambient data

Michael Anderson; Ning Zhou; John Pierre; Richard Wies

Bootstrap-based confidence interval estimates for electromechanical modes from multiple output analysis of measured ambient data

Michael Anderson
, Ning Zhou
, John Pierre
, Richard Wies

Electrical and Computer Eng

Binghamton University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Scopus citations

Abstract

Previously, variations of the Yule-Walker techniques have been applied successfully to give point estimates of electromechanical modes of a power system based on measured ambient data. This paper introduces a bootstrap method to give confidence interval estimates for the electromechanical modes. Simulation results from a 19-machine model show the validation of the bootstrap method and its consistence to Monte-Carlo methods. Actual measurement data taken from western North American Power Grid in 2000 is processed using the bootstrap method to give confidence interval estimates for inter-area mode damping ratios. The use of multiple outputs is shown to improve the performance and tighten the confidence intervals.

Original language	English
Title of host publication	2005 IEEE Power Engineering Society General Meeting
Pages	1474
Number of pages	1
State	Published - 2005
Event	2005 IEEE Power Engineering Society General Meeting - San Francisco, CA, United States Duration: Jun 12 2005 → Jun 16 2005

Publication series

Name	2005 IEEE Power Engineering Society General Meeting
Volume	2

Conference

Conference	2005 IEEE Power Engineering Society General Meeting
Country/Territory	United States
City	San Francisco, CA
Period	06/12/05 → 06/16/05

Cite this

@inproceedings{4218ca19933c4ff19e31c08de8dbcc5e,

title = "Bootstrap-based confidence interval estimates for electromechanical modes from multiple output analysis of measured ambient data",

abstract = "Previously, variations of the Yule-Walker techniques have been applied successfully to give point estimates of electromechanical modes of a power system based on measured ambient data. This paper introduces a bootstrap method to give confidence interval estimates for the electromechanical modes. Simulation results from a 19-machine model show the validation of the bootstrap method and its consistence to Monte-Carlo methods. Actual measurement data taken from western North American Power Grid in 2000 is processed using the bootstrap method to give confidence interval estimates for inter-area mode damping ratios. The use of multiple outputs is shown to improve the performance and tighten the confidence intervals.",

author = "Michael Anderson and Ning Zhou and John Pierre and Richard Wies",

year = "2005",

language = "English",

isbn = "078039156X",

series = "2005 IEEE Power Engineering Society General Meeting",

pages = "1474",

booktitle = "2005 IEEE Power Engineering Society General Meeting",

note = "2005 IEEE Power Engineering Society General Meeting ; Conference date: 12-06-2005 Through 16-06-2005",

}

Anderson, M, Zhou, N, Pierre, J & Wies, R 2005, Bootstrap-based confidence interval estimates for electromechanical modes from multiple output analysis of measured ambient data. in 2005 IEEE Power Engineering Society General Meeting. 2005 IEEE Power Engineering Society General Meeting, vol. 2, pp. 1474, 2005 IEEE Power Engineering Society General Meeting, San Francisco, CA, United States, 06/12/05.

Bootstrap-based confidence interval estimates for electromechanical modes from multiple output analysis of measured ambient data. / Anderson, Michael; Zhou, Ning; Pierre, John et al.
2005 IEEE Power Engineering Society General Meeting. 2005. p. 1474 (2005 IEEE Power Engineering Society General Meeting; Vol. 2).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Bootstrap-based confidence interval estimates for electromechanical modes from multiple output analysis of measured ambient data

AU - Anderson, Michael

AU - Zhou, Ning

AU - Pierre, John

AU - Wies, Richard

PY - 2005

Y1 - 2005

N2 - Previously, variations of the Yule-Walker techniques have been applied successfully to give point estimates of electromechanical modes of a power system based on measured ambient data. This paper introduces a bootstrap method to give confidence interval estimates for the electromechanical modes. Simulation results from a 19-machine model show the validation of the bootstrap method and its consistence to Monte-Carlo methods. Actual measurement data taken from western North American Power Grid in 2000 is processed using the bootstrap method to give confidence interval estimates for inter-area mode damping ratios. The use of multiple outputs is shown to improve the performance and tighten the confidence intervals.

AB - Previously, variations of the Yule-Walker techniques have been applied successfully to give point estimates of electromechanical modes of a power system based on measured ambient data. This paper introduces a bootstrap method to give confidence interval estimates for the electromechanical modes. Simulation results from a 19-machine model show the validation of the bootstrap method and its consistence to Monte-Carlo methods. Actual measurement data taken from western North American Power Grid in 2000 is processed using the bootstrap method to give confidence interval estimates for inter-area mode damping ratios. The use of multiple outputs is shown to improve the performance and tighten the confidence intervals.

UR - https://www.scopus.com/pages/publications/27144541001

M3 - Conference contribution

SN - 078039156X

SN - 9780780391567

T3 - 2005 IEEE Power Engineering Society General Meeting

SP - 1474

BT - 2005 IEEE Power Engineering Society General Meeting

T2 - 2005 IEEE Power Engineering Society General Meeting

Y2 - 12 June 2005 through 16 June 2005

ER -

Bootstrap-based confidence interval estimates for electromechanical modes from multiple output analysis of measured ambient data

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this