Uncovering contributing factors to interruptions in the power grid: An Arctic case

TL;DR

This study uses data analysis and machine learning to identify key factors causing power failures in an Arctic grid, achieving moderate prediction accuracy and providing insights for failure prevention.

Contribution

The paper introduces a data-driven approach combining statistical and machine learning techniques to predict and interpret causes of power failures in an Arctic region.

Findings

Wind speed and local industry activity are main factors.

Achieved 57% prediction accuracy with SVM.

Identified exposed line locations as triggers.

Abstract

Electric failures are a problem for customers and grid operators. Identifying causes and localizing the source of failures in the grid is critical. Here, we focus on a specific power grid in the Arctic region of North Norway. First, we collect data pertaining to the grid topology, the topography of the area, the historical meteorological data, and the historical energy consumption/production data. Then, we exploit statistical and machine-learning techniques to predict the occurrence of failures. We interpret the variables that mostly explain the classification results to be the main driving factors of power interruption. We are able to predict 57% (F1-score 0.53) of all failures reported over a period of 1 year with a weighted support-vector machine model. Wind speed and local industry activity are found to be the main controlling parameters where the location of exposed power lines is a likely trigger. In summary, we discuss causing factors for failures in the power grid and enable the distribution system operators to implement strategies to prevent and mitigate incoming failures.