Dimensionality Reduction and Early Event Detection Using Online Synchrophasor Data Conference Paper

abstract

• This paper proposes a novel approach to utilizing large online synchrophasor data for early event detection in power systems. Based on principal component analysis (PCA), a linear basis of the massive online phasor measurement unit (PMU) data can be extracted to reduce the dimensionality. Using the linear basis with much reduced dimensionality, an early event detection algorithm is proposed. This algorithm is capable of predicting the changes of system operating conditions by comparing the error between PCA-projected and actual values from a few selected locations. Numerical case studies based on both PSS/E simulation and actual PMU data from Electric Reliability Council of Texas are conducted to demonstrate the efficacy of this algorithm. 2013 IEEE.

name of conference

• 2013 IEEE Power & Energy Society General Meeting

authors

• Kumar, Panganamala
• Xie, Le

published proceedings

• 2013 IEEE POWER AND ENERGY SOCIETY GENERAL MEETING (PES)

author list (cited authors)

• Chen, Y., Xie, L. e., & Kumar, P. R.

citation count

• 26

complete list of authors

• Chen, Yang||Xie, Le||Kumar, PR

publication date

• January 2013

publisher

• Institute of Electrical and Electronics Engineers (IEEE)  Publisher

published in

• IEEE Power and Energy Society General Meeting  Journal

keywords

• Dimensionality Reduction
• Early Event Detection
• Phasor Measurement Unit
• Principal Component Analysis

Digital Object Identifier (DOI)

• 10.1109/pesmg.2013.6672974

International Standard Book Number (ISBN) 13

• 9781479913039

start page

• 1

end page

• 5

URL

• http://dx.doi.org/10.1109/pesmg.2013.6672974