Wildfire Risk Assessment of Transmission-Line Corridors Based on Logistic Regression

Wildfires, occurred in transmission line corridors, seriously affect the security and stability of power grids. In order to reduce the effects of wildfire on power grids, a logistic regression (LR) model was employed to assess the risk distribution of wildfire occurrences in transmission line corridors. Guizhou Province which locates at south-west of China was taken as the study region. An inventory dataset of Guizhou Province including 14 wildfire-related variables and historical wildfire occurrences were assembled. After the multicollinearity analysis of wildfire-related variables, the occurrence probability of wildfires in the study region was calculated by the LR model. And a wildfire risk distribution map was drawn in a resolution of 1 km × 1 km for visualization. The wildfires happened from January to May 2020 was applied to the validity of wildfire risk distribution map. The results shows that 82.65% of fire points were happened in the area with the third and fourth risk level. It indicates that the proposed method can be used to provide technical support for the maintenance staffs to carry out differentiation technology and strategy for wildfire protections.