自组织特征映射在鄱阳湖地区洪涝灾害研究中的应用

详细信息

Research on Flood Disaster of Poyang Lake Valley with Self- Organizing Map

1.
School of Economics & Management ,Beijing university of Aeronautics and Astronautics ,Beijing 100083 ,China
2. The French National Institute For Research In Computer Science And Control ,CEDEX 78183 , France

摘要: 自组织特征映射是一种人工神经网络方法, 可以同时实现模式识别和数据分类。它可以将高维数据间复杂、非线性的统计关系转换为低维形式中简单的几何关系。该文利用自组织特征映射方法对鄱阳湖地区月降水强度和月高水位变化模式进行分析, 对鄱阳湖地区洪水灾害形成原因进行归纳。
- 自组织特征映射 /
- 模式识别 /
- 聚类分析 /
- 洪水灾害
Abstract: Self- Organizing Map is a method of artificial neural network , whichimplements patternrecognitionand data clustering simultaneously . SOMis able to convert complex nonlinear statistical relationships of high- dimensional datainto simple geometric relationships . Inthis article , we investigate the variationpatterns of precipitationintensity and maximumwater level in Poyang Lake valley withself-organizing map . Thenthe cause of flood disaster is discussed according to the result of calculation.
- self-organizing map /
- pattern recognition /
- clustering analysis /
- flood disaster

[1]	闵骞. 鄱阳湖洪水灾变特征的分析[J]. 江西水利科技,1997,(2) :86-93.
[2]	闵骞. 关于建立洪水等级划分模式的初步构思———以鄱阳湖为例[J]. 水文,1996,(1) :43- 47.
[3]	Teuvo Kohonen. Self- Organizing Maps[M].Berlin :Springer,1995.
[4]	任若恩, 王惠文. 多元统计数据分析[M]. 北京: 国防工业出版社,1997. 12-14.