Fire Risk Assessment of Transmission Line Based on BP Neural Network

With global warming, the increasing forest fires have caused trips and outages frequently along transmission lines, which is a serious threat to the operation stability of power grids. Impact factors on transmission line fires are numerous and the current risk assessment methods could hardly handle the complex nonlinear relationship between risks and factors, therefore, a method based on BP neural network is presented to assess fire risk of transmission line in this paper. Firstly, risk assessment system would be established according to impact factors on transmission line fire. Then, based on neural network model, the complex nonlinear relationships between fire risk grade and evaluation factors could be built. Finally, combined with GIS technology, risk assessment on transmission line fire would be done. The applicability and accuracy of this method have been explored by a fire risk assessment of transmission line in Shanxi province. The result shows that the BP neural network based model has good recognition effect, credibility, and consistent with the survey result on transmission line fire risk in the region over the years.