摘要
为了解决传统模糊聚类算法对单一隶属度依赖较重的问题,设计了一种直觉模糊聚类算法,运用于导弹打击目标规划领域;该算法首先根据目标的原始数据矩阵,利用隶属函数和非隶属函数,计算得出直觉模糊特征量矩阵;然后利用直觉模糊相似度和相异度,计算得出相似矩阵,再进行合成运算求得等价矩阵;最后根据设定的置信度,进行聚类分析。经实例仿真计算,该算法能够较好地按照规划标准对目标规划分类,规划结果的准确性有了大幅的提高。
A kind of intuitionistic fuzzy clustering algorithm is designed to be applied to missile attac- king target planning the single membersh to resolve the problem of the traditional fuzzy clustering algorithm more depending on ip. Firstly, according to the original data matrix of the targets, the intuitionistic fuzzy feature matrix is calculated by membership function and non-membership function. Then, the intuitionis- tic fuzzy similarity matrix is calculated by the degrees of the similarity and dissimilarity. The equivalent matrix is obtained from compositional operations. Finally, according to the set degree of confidence, the equivalent matrix is clustered. Simulation shows that the algorithm can better classify the targets by the planning standards. The planning accuracy is greatly improved
出处
《现代防御技术》
北大核心
2014年第2期50-55,71,共7页
Modern Defence Technology
关键词
目标规划
直觉模糊
聚类
target planning
intuitionistic fuzzy
clustering
作者简介
贺小亮（1986-）,男,河北肃宁人.博士生,从事导弹作战建模与仿真的研究.通信地址：100091 北京市海淀区红山口甲三号国防大学研究生院学员13队 E-mail：xiaoliang415222902@163.com
