划分水团的模糊密度聚类法  

A CLUSTERING METHOD WITH THE FUZZY DENSITY IN DISTINGUISHING WATER MASSES

作　　者：李凤岐[1] 王凤钦 苏育嵩[1] 

机构地区：[1]青岛海洋大学物理海洋与海洋气象系

出　　处：《青岛海洋大学学报（自然科学版）》1992年第1期12-18,共7页Journal of Ocean University of Qingdao

基　　金：自然科学基金

摘　　要：提出不受资料输入顺序影响的模糊密度聚类法。给出模糊点密度的定义与计算方法,并用渤、黄、东海的温盐资料,进行模糊密度聚类计算,对聚类结果及水团划分进行了讨论。The t-s diagram is a weighty tool for distinguishing water masses.But the increase of investigation data causes the limits of point-sets on the diagram to be more fuzzy.Particularly,it is difficult to divide water masses in,shallow water areas intuitively from the t-s diagram.The clustering method is applied widly because it can partly overcome the above mentioned difficulties.However,the order of data input may sometimesaffect merging results.We suggest a clustering method with the fuzzy density,which isnot affected by the order of data input.The definition of fuzzy density for points andits calculating method are given in this paper The fuzzy density clustering is made,the merging results and the distinction of water masses are discussed.

关 键 词：水团 模糊数学 聚类分析 

分 类 号：P731.16[天文地球—海洋科学]