With the construction of a new type of power system with new energy as the main body under the "double carbon" target, the challenges of natural disasters such as wildfires, icing and typhoons in the production, transmission and distribution of electric energy are more severe. Therefore, a fire point identification algorithm based on the world's first operational dawn-dusk orbit meteorological satellite Fengyun-3E satellite (FY-3E) was proposed, which was suitable for mountain fire monitoring in transmission line corridors, the adverse effects of large solar zenith angle observation conditions and satellite perspective differences on the accurate acquisition of infrared channel detection data were eliminated. Cloud information extraction and cloud pixel fire point extraction under complex atmospheric observation conditions were realized, which reduced fire point false alarms and missed alarms. The analysis of mixed pixel linear spectrum separation method shows that the fire point detection sensitivity of FY-3E mid-infrared channel is 4 times higher than that of geostationary meteorological satellite. The effectiveness of the proposed algorithm and the superiority of FY-3E in fire detection sensitivity, spatial range accuracy and positioning accuracy were verified. Compared with geostationary meteorological satellites, the fire location accuracy can be increased by more than one time, and it can effectively detect the fire time in advance by the transmission line operation and maintenance department, and guide the relevant departments to take timely measures to reduce the impact of wildfires on power grid operation.