Leveraging Accuracy-uncertainty Tradeoff in SVM to Achieve Highly Accurate Outage Predictions

Authors

Rozhin Eskandarpour, University of Denver
Amin Khodaei, University of Denver

Publication Date

1-2018

Document Type

Article

Organizational Units

Daniel Felix Ritchie School of Engineering and Computer Science, Electrical and Computer Engineering

Keywords

Support vector machines, Hurricanes, Power grids, Logistics, Training, Learning systems

Abstract

This letter proposes a three-dimensional Support Vector Machine (SVM) for power grid component outage prediction, and furthermore leverages its accuracy-uncertainty tradeoff to achieve highly accurate results. The model is developed based on three distinct features of component deterioration, distance from the extreme event, and the intensity of the extreme event, and is analytically investigated to exhibit its acceptable performance.

Publication Statement

Copyright held by author or publisher. User is responsible for all copyright compliance.

Recommended Citation

Eskandarpour, Rozhin, and Khodaei, Amin. “Leveraging Accuracy-Uncertainty Tradeoff in SVM to Achieve Highly Accurate Outage Predictions.” IEEE Transactions on Power Systems, vol. 33, no. 1, 2018, pp. 1139–1141. doi: 10.1109/tpwrs.2017.2759061.