基于BP神经网络的洪涝灾害风险评价——以杭埠河流域为例  

Flood Disaster Risk Assessment Based on BP Neural Network— A Case Study of Hangbu River Basin

作　　者：程先富[1,2] 匡毅[1,2] 赵阳[1] 

机构地区：[1]安徽师范大学国土资源与旅游学院,安徽芜湖241003 [2]安徽自然灾害过程与防控研究省级实验室,安徽芜湖241003

出　　处：《皖西学院学报》2015年第2期1-4,共4页Journal of West Anhui University

基　　金：国家自然科学基金(41271516)

摘　　要：根据产生洪涝灾害风险的各种因素,从致灾因子、孕灾环境、承灾体和社会承灾能力4个方面,选取年均降雨量、坡度、河网密度、植被覆盖度、土地利用、人口密度等15个评价指标。建立BP神经网络模型,通过指标数据进行训练和检验,对杭埠河流域洪涝灾害风险进行了评价。结果表明,流域东北部洪涝灾害风险高,流域中部洪涝灾害风险属于中等水平,流域西南部洪涝灾害风险低。BP神经网络模型直观、简便具有推广和应用价值。According to the various factors resulting from flood disaster risk, 15 evaluation indexes of the average annual rainfall, slope, drainage density, vegetation coverage, land use, population density and so on were selected from hazards, disaster environment, hazard bearing body and social disaster bearing force. BP neural network model was built by training and testing by the data of index and to assess the flood disaster risk in Hangbu River basin. The results show that flood risk level is high in northeastern basin, the central region of basin belongs to the medium level and the risk level of flood disaster in southwestern basin is low. BP neural network model is intuitive, simple and has a promotion and application value.

关 键 词：洪涝灾害 风险评价 BP神经网络 杭埠河流域 

分 类 号：X43[环境科学与工程—灾害防治]