基于径向基函数神经网络（RBFN）的内蒙古土壤风蚀危险度评价

土壤风蚀是北方干旱和半干旱地区土地沙化和沙尘暴灾害的首要环节和主要动力过程之一. 选取影响内蒙古自治区土壤风蚀演化的相关指标，运用GIS技术提取各指标数据，构建径向基函数神经网络（Radial Basis Function Network，RBFN）；根据不同风蚀危险程度标准，选取12个市、县（旗）相关数据进行训练，确定网络模型参数，进而对内蒙古自治区88个市、县（旗）的土壤风蚀危险度进行了评价. 结果表明:内蒙古自治区西部为土壤风蚀发生的极强危险区，西北为强危险区，中部为中度危险区，而东部为轻度危险区；利用其他研究对该评价结论进行对比验证，结果较为理想. 

Using RBFN Model and GIS Technique to Assess Wind Erosion Hazard in Inner Mongolia

Soil wind erosion is the primary process and the main driving force for land desertification and sand-dust storms in arid and semi-arid areas of Northern China. Related indicators affecting the soil wind erosion of Inner Mongolia Autonomous Region were selected, the GIS technology was used to extract the spatial data, and the Radial Basis Function Network （RBFN） model was construct for assessment of wind erosion hazard. After training the sample data of the different levels of wind erosion hazard of 12 cities （counties）, the parameters of the model were obtained, and then the wind erosion hazard assessed for 88 cities （counties） in Inner Mongolia. The result shows that in the western parts of Inner Mongolia wind erosion hazard are very severe, counties in the middle regions of Inner Mongolia vary from moderate to severe, and in eastern are slight. The comparison of the result with other researches shows that the result is in conformitywith actual conditions, proving the reasonability and applicability of the RBFNmodel.