LAUSR

Abstract Landslide is a kind of geological disaster that occurs widely in the world. With the acceleration of human urbanization, landslide geological disaster appears more and more frequently in some areas, which seriously threatens the safety of human life and property. It is urgent to build a safety and effective landslide smart safety prediction model. Most of the mountainous areas are in contact with the geological conditions of soil and rock, so the mountainous areas are prone to landslides. This paper takes Soil-Rock Contact Zone of Southern Shaanxi as example to establish smart safety early warning model based on BP neural network. The model takes the contact zone lithology, void ratio, water content, liquid index, slope and slope height factors which reflect the characteristics of landslide in the contact zone of soil and rock in mountain area as input signals, and the output results are (0,1), (1,0), respectively indicating that landslide occurs or does not occur. The simulation results show that the prediction results of the model are consistent with the actual results, which can be used for the actual prediction of landslide geological hazards in the contact zone of soil and rock in mountain areas.