Forest Fire Tripping Probability Prediction System Based on Partial Mutual Information Method

摘要

High voltage transmission lines are mainly built in deep forests. However, forest fires have occurred frequently in recent years. At present, the power grids mainly monitor and warn forest fire, but they lack the longer time scale for the prediction of forest fire. Based on the partial mutual information method, the local fire risk level model is established, and then the forest fire tripping probability prediction system is established for the transmission lines of the power grid. On September 1, 2017 wildfires trip accident happened in some regions of Guangdong Province, and then the forest fire tripping probability in Guangdong Province is forecasted. The results show that the system can predict the accident area with a high probability of fire tripping about 1 day in advance, and the application of the system can provide longer time prediction information of forest fire tripping for the safe operation of power grid transmission lines.