中国降水区划模糊聚类软划分法  

Precipitation Regionalization Based on Fuzzy Clustering Algorithm

作　　者：何州杉月[1] 杨林[1] 

机构地区：[1]北京师范大学数学科学学院,北京100875

出　　处：《气象科技》2011年第5期582-586,共5页Meteorological Science and Technology

摘　　要：利用软划分模糊聚类分析法,对中国160个观测站的近50年的年均降水数据进行研究,按降水量的多少和年际变化规律,把中国降水区域划分n类降水带,并研究了聚类参数对于划分结果的影响。进一步研究了经典的5类降水带划分法,发现了一些被分入某一降水带的地区可能具有较大的与其他类别的相似性(如通辽,大连等),研究结论提供了更丰富的隶属度信息描述这种相似性,并能够在退化后与其他方法得到的划分吻合。可以为进一步分析地理条件影响降水提供量化信息。A soft（fuzzy） clustering algorithm is used to divide the precipitation region of China into n clusters according to the precipitation amount,based on the data from 160 meteorological stations of China in the last 50 years,with each station having a degree of belonging to clusters,as in the fuzzy logic,rather than belonging completely to just one cluster.The effect of clustering parameters on regionalization is examined.The further study of the conventional 5-category precipitation regionalization shows that some regions allocated into a specific category do not exactly belong to one certain cluster,but has some similarity with the other regions,such as Tongliao and Dalian,etc.The soft（fuzzy） clustering algorithm can provide more information on the study of the factors affecting the precipitation to better describe the similarity,and be able to obtain the agreement with the results from other classification approaches after degradation.

关 键 词：降水带 软划分 模糊聚类 

分 类 号：P425.42[天文地球—大气科学及气象学]