基于权重的WKFCM聚类算法在泥石流评价中的应用

摘要

泥石流的形成受地形条件、地质构造、气象水文等多种因素的综合影响，不同区域的泥石流灾害在空间分布上具有一定的差异性和相似性，聚类分析能够识别出这种相似的特性。针对 KFCM聚类算法没有考虑到不同影响因子对于泥石流灾害的贡献程度不同的缺陷，引入权重的概念，对 KFCM算法进行改进，选取沟床比降、流域面积、构造系数、冰川坡度、岩性系数、平均坡度、最大淤积、可移方量、冰川面积与流域面积比9个因素作为区域内冰川泥石流危险性的评价指标，采用层次分析法求取各个评价指标的权重值，以西藏境内30条具有良好工作基础的冰川泥石流沟为研究对象，探究改进后的 WKFCM聚类算法在泥石流危险性评价中的应用。结果表明，改进后的算法可以避免传统评价方法阈值确定时的主观不确定性，将其应用在泥石流危险性评价中是有效可行的。%The formation of debris flow include terrain condition,geological structure,weather and other fac-tors.There are some differences and similarities in debris flow of different area.Using cluster analysis can identify the similarity.General KFCM Algorithm has almost no consideration for the contribution of different factors have different weights,therefore,this article try to improve the KFCMalgorithm through introducing concept of weight, then we get WKFCMalgorithm.After analyze the present data of these glacier debris flow gullies,9 factors are cho-sen as main factors for evaluating the risk of these gullies.Analytic Hierarchy Process is used to determine the weights of the factors.The study area is s located in Tibet along the Sichuan-Tibet highway where 30 glacier debris flow gullies are selected for study.According with the reference result,it shows that it is practical to use this meth-od to evaluate the risk of debris flow.