Landslide Risk Assessment Based on Artificial Neural Network Model

Abstract

According to the public data, the overall trend of economic loss caused by landslide disaster is growing year by year, and improving the prediction level of landslide disaster can effectively reduce the loss caused by landslide disaster. Therefore, the study of landslide prediction and prediction has important practical significance, and landslide displacement prediction is an important content of landslide prediction and prediction. This paper mainly studies landslide risk assessment based on artificial neural network model. In this paper, based on the results of field investigation, literature summary and expert experience, collected samples of 100 landslides, analyzes the distribution and the main influence factors of landslide, and establishes the surface elevation, and vegetation index, cutting slope, annual average rainfall, surface density, overlying soil types for the impact factor of regional landslide hazard assessment system. According to the requirement of BP neural network (BPNN) for input data, the landslide sample data processing method which is suitable for Sichuan landslide risk assessment system is established.