Spatial vulnerability assessment of rural settlements in hilly areas using BP neural network algorithm

Geological disasters occur frequently in hilly areas of Sichuan province, and the development of rural settlement production and life may be affected by geological disasters.This study aims to assess the spatial vulnerability of rural settlements in hilly areas of Sichuan Province through Back-propagation neural network, find the main factors that really affect the spatial vulnerability of rural settlements, analyze the distribution characteristics of spatial vulnerability of rural settlements in hilly areas of Sichuan Province, and find methods to reduce the spatial vulnerability of rural settlements. Through research, it can be found that the three main factors affecting the spatial vulnerability of rural settlements in hilly areas of Sichuan are climate factors, soil erosion degree, and geological hazard hidden danger degree. The quantitative climate factor data is the ratio of annual precipitation to population in rural settlements,the higher the annual precipitation, the higher the spatial vulnerability. The second largest influencing factor is the degree of soil erosion,the more severe the soil erosion, the higher the spatial vulnerability in the districts and counties. The third major influencing factor is the degree of geological hazards, and the higher the degree of geological hazards, the higher the spatial vulnerability. Other natural environment and rural settlement spatial factors also have a certain impact on the spatial vulnerability of rural settlements, while socio-economic factors have almost no impact on the spatial vulnerability of rural settlements.Finally, the relationship between the spatial vulnerability of rural settlements and its location are obtained from the relevant data of all sample districts and counties, and combined with the geological environment and climate of the hilly areas of Sichuan, the spatial distribution characteristics of the spatial vulnerability of rural settlements in the hilly areas of Sichuan are summarized: The closer the districts and counties are to the geological fracture zone between Longmen Mountain and Longquan Mountain, the higher the spatial vulnerability of rural settlements.The spatial vulnerability of rural settlements in the west of hilly areas is higher than that in the east, and there is little difference between north and south. Then, based on the above achievements, this study proposes strategies and suggestions to reduce the spatial vulnerability of rural settlements in hilly areas of Sichuan, providing reference materials and data support for local planning departments in rural planning work.