Research Article

Research and Application of Fire Forecasting Model for Electric Transmission Lines Incorporating Meteorological Data and Human Activities

Table 5

The results of typical models with the human index in Anhui province.

ā€‰IndexModel 1 with human indexModel 2 with human indexModel 3 with human indexModel 4 with human indexModel 5 with human index

TrainingTraditional accuracy rate0.97350.97580.97490.97900.9790
Weighted accuracy rate1.06251.07071.06521.08531.0853
Traditional accuracy rate of fire level 111111
Traditional accuracy rate of fire level 20.21670.26670.26670.30.3
Traditional accuracy rate of fire level 30.50.57140.50.78570.7857
Traditional accuracy rate of fire level 40.85180.88890.85180.96300.9630

TestTraditional accuracy rate0.98900.98900.98860.98950.9895
Weighted accuracy rate1.01461.01551.01461.01781.0178
Traditional accuracy rate of fire level 11110.99950.9995
Traditional accuracy rate of fire level 20.16670.1250.08330.16670.1667
Traditional accuracy rate of fire level 30.250.50.50.750.75
Traditional accuracy rate of fire level 40.8750.8750.8750.8750.875