Title

Power Flow Studies using Principal Component Analysis

Abstract

This paper employs Principal Component Analysis (PCA), a data mining technique, to study power flow. A simulation-based process is established to perform the study, which consists of steps such as system linearization, training data construction, PCA analysis and results interpretation. The PCA results are presented in a straightforward manner, and interpreted from power system perspective. The conclusions not only are consistent with the well-known facts such as PQ decoupling, but also discover hidden facts such as correlation pattern among input variables and state variables. The proposed power flow study method is not only a helpful tool for power system operators in practice, but also beneficial for engineering students in study.

Meeting Name

40th North American Power Symposium, NAPS2008 (2008: Sept. 28-30, Calgary, Canada)

Department(s)

Electrical and Computer Engineering

Keywords and Phrases

Correlation Patterns; Data Mining Techniques; Input Variables; PCA Analysis; Power Flow Studies; Power Flows; Power System Operators; Power Systems; PQ Decoupling; Simulation-Based; State Variables; Training Data; Data Flow Analysis; Data Mining; Electric Power Transmission Networks; Principal Component Analysis; Power Flow; Principal Component Analysis (PCA)

International Standard Book Number (ISBN)

9781424442836

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2008 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.

Share

 
COinS