聚类分析法在贵州省雷电灾害易损性区划中的应用

采用R型聚类分析法对尽可能多的影响雷灾易损性的因子进行了筛选,确定100 k A以上平均地闪次数、地闪密度、雷电灾害频度、生命易损模数、区域生命易损模数5个因子为依据变量,用欧氏距离计算各样本的相似性测度,应用Q型聚类中的Ward’s法进行层次聚类分析,采用迭代聚类分析的方法进行相互验证,最后对贵州省雷电灾害易损性区划结果进行解释和证实、绘制雷灾风险区划图,结果表明:贵阳市雷电灾害的易损性最高,属高风险区,安顺市、六盘水市属中等风险区,遵义市、黔南州、黔东南州、铜仁市、毕节市、黔西南州属弱风险区。

Cluster Analysis Applied in Lightning Disaster Vulnerability Zoning in Guizhou Province

Based on the method of R￣Cluster Analysis, this article screens the impact of the thunder disaster vulnerability factors as much as possible, and finally 5 factors—average lightning number above 100kA, lightning density, accident number of lightning disaster per year, vulnerability casualties modulus, and region vulnerability casualties modulus are confirmed. Comparability measures of each sample are calculated by using Euclidean distance, and then the Ward＇s method of Q-Cluster is used to analyze and calculate the data. By using the method of iterative cluster analysis into the mutual authentication, the vulnerability of lightning disasters in Guizhou province is divided. The article also explains the results for zoning and draws the thunder disaster risk zoning map. The results show Guiyang is in high risk area for its lightning disaster vulnerability is the highest; Anshun and Liupanshui are in moderate risk area;Zunyi, Qiannan, Qiandongnan, Tongren, Bijie and Qianxinan are in the weak risk area.