摘要
研究工作基于分形理论,应用G-P算法,通过对常州某股份有限公司事故时间序列的分析,构建了n维相空间,并利用关联维数的基本原理,对事故时间序列进行了分形特征分析.研究结果表明当嵌入维数达到10以后,该事故动力学系统具有稳定的关联维数4.1,说明至少有4个因子在影响着事故时间序列的动态变化,并且该系统的有效自由度为10.研究对建立事故时间序列的预测模型有较大的参考价值.
The accidents time sequence of corporation in Changzhou was analyzed using G-P arithmetic , which was based on the fraetal theory, and n dimensional phase space was then constructed. Moreover, the author also had conducted the fractal features analysis combined with basic principle of correction dimensions. The results indicated that the dynamic system of accidents had a steady correction dimension(4.1 ) after the embedded dimension came to be 10, which explained that at least four factors were controlling the dynamic changes of accidents sequence, and the effective freedom was 10. The research provides much reference value for establishing predictable model of accidents sequence.
出处
《系统工程理论与实践》
EI
CSCD
北大核心
2006年第4期141-144,共4页
Systems Engineering-Theory & Practice
基金
江苏省常州市社会发展计划(CS2005004)
关键词
时间序列
重构相空间
事故分形特征
事故关联维数
time sequence
reconstruction phase space
accident fractal charaeter
accident correction dimension