摘要
灰关联分析要求数据序列个数必须大于2才能构成灰关联空间,难以解决两个数据序列之间属性检验问题。为此,以灰色系统的基本原理为依据,提出灰关系概念。从生成来自这两个数据序列均值的定值参考序列入手,定义最大灰差与最小相似系数以及系统属性权重系数,给出有关的性质与定理,从而构建出一个特殊的灰关联空间即灰相似空间,允许概率分布未知和数据个数很少。在灰相似空间中,给出系统属性检验的一个准则和相应的灰置信水平。对捷联惯组(SIMU)的误差系数和武器系统效能的两个试验研究表明,灰关系概念在系统分析中具有显著性,灰置信水平可以得到95%以上。
It is difficult to tackle the problem about testing of attributes between two data sequences for grey relational analysis because the number of data sequences must be more than 2 in order tO structure a grey relational space. For this reason, a concept, grey relation, was advanced using rationale of grey system. Started with generation of the fixed value reference sequence from mean of these two data sequences, maximum grey difference, minimum coefficient of similarity and weighting coefficient of sys- tem attributes were defined, relational properties and theorems were given, and accordingly an especial grey relational space, viz. grey similarity space which permits the probability distribution unknown and the number of data small, was formed. On grey similarity space, a testing criterion of system attributes was proposed, carried with its corresponding grey confidence level. Experiments on the error coefficient of strapdown inertial measurement unit （SIMU） and the effectiveness of weapon systems show that grey relation is provided with significance at the above 95 % grey confidence level.
出处
《兵工学报》
EI
CAS
CSCD
北大核心
2008年第1期72-77,共6页
Acta Armamentarii
基金
国家自然科学基金资助项目(5037501150675011)
国防科工委项目子项项目(MKPT-2004-51ZD/7)
关键词
系统学
关系
数据序列
乏信息
非统计原理
systematology
relation
data sequence
poor information
non-statistical theory
