Load Curtailment Estimation in Response to Extreme Events

TL;DR

This paper introduces a machine learning approach using SVMs to predict grid component failures during hurricanes and estimate resulting load curtailments, enhancing power system resilience analysis.

Contribution

It presents a novel SVM-based model for predicting component outages due to hurricanes and integrates it into load curtailment estimation, improving accuracy over traditional methods.

Findings

Effective prediction of component outages during hurricanes

Accurate load curtailment estimation on IEEE 30-bus system

Model performance validated across various hurricane scenarios

Abstract

A machine learning model is proposed in this paper to help estimate potential nodal load curtailment in response to an extreme event. This is performed through identifying which grid components will fail as a result of an extreme event, and consequently, which parts of the power system will encounter a supply interruption. The proposed model to predict component outages is based on a Support Vector Machine (SVM) model. This model considers the category and the path of historical hurricanes, as the selected extreme event in this paper, and accordingly trains the SVM. Once trained, the model is capable of classifying the grid components into two categories of outage and operational in response to imminent hurricanes. The obtained component outages are then integrated into a load curtailment minimization model to estimate the nodal load curtailments. The merits and the effectiveness of the proposed models are demonstrated using the standard IEEE 30-bus system based on various hurricane path/intensity scenarios.