A Symbolic Aggregate Approximation-Based Data Mining Tool for the Detection and Classification of Power Grid Voltage Events

Publication Name

Advances in Control Techniques for Smart Grid Applications

Abstract

Precise detection and classification of voltage events would require very detailed investigation from a very large pool of data. There exist several data mining and parametric analysis techniques to detect voltage events, however, they are computationally burdensome. In this chapter, a symbolic aggregate approximation (SAX)-based data mining tool is developed that can not only detect and classify voltage events accurately but also with considerably less computational effort. Instead of cycle-by-cycle analysis, a cluster-based analysis is proposed to classify the voltage events where the SAX algorithm is used for reducing the dimensionality of the raw time series. The proposed algorithm has been tested on a practical test network and results have been presented.

Open Access Status

This publication is not available as open access

First Page

123

Last Page

140

Share

COinS
 

Link to publisher version (DOI)

http://dx.doi.org/10.1007/978-981-16-9856-9_5