Evaluation of Drifting Snow Susceptibility Based on GIS and GA-BP Algorithms

Drifting snow, the flow of dispersed snow particles near ground level under action wind, is a major form damage. When drifting occurs on railways, highways, and other transportation lines, it seriously affects their operational safety results in disasters. disasters frequently occur high latitudes northwest China. At present, most scholars are committed to studying prevention control measures but prerequisite for effectively evaluate susceptibility along railways highways identify areas with risk occurrence. Taking Xinjiang Afukuzhun Railway as an example, this study uses geographic information system (GIS) combined on-site monitoring surveys establish evaluation index system. The (DSSI) calculated through weight evidence (WOE) model, genetic algorithm backpropagation (GA-BP) used obtain optimised weights improve accuracy model evaluation. show that accuracies WOE (WOE-BP) (WOE-GA-BP) 0.747, 0.748, 0.785, respectively, indicating method can be applied susceptibility.