Predictive model of the outage of transmission lines exposed to wildfires

Abstract

Transmission lines are essential for access to clean and affordable energy sources, Sustainable Development Goal 7. Wildfires are an important factor in the degradation of the quality of public transmission service provision. This work sought to build a model to predict the outage of a transmission line when exposed to a wildfire. The characteristics analysed of the spans exposed to fires of twelve transmission lines at a voltage level of 500 kV in Brazil totalled 3,998 km. The logistic regression technique was used for the study. It was possible to reach a model with a hit rate higher than 73% for the occurrence of transmission line outages. The quantity of fire outbreaks, the climatic variables, and the type of biome of the spans were observed to be the best predictive variables available. The temperature rise can potentially increase the number of outages caused by wildfires.