基于时间序列分析的制造质量水平变化的混沌特性

在定义了制造企业生产制造时间序列的基础上,使用G-P算法计算时间序列的关联维数。通过关联维数的计算得到相应的嵌入维数后,使用基于相空间重构的小数据量法计算混沌时间序列的Lyapunov指数。采集HZ近三年的日生产产品合格率作为研究制造质量水平变化混沌特性的原始数据。在以上技术路线及数据的基础上,得到的关联维为分数,而Lyapunov指数为正值,说明日生产产品合格率变化时间序列呈现出混沌特性。另外将以上数据分为8个时间序列,每个时间序列同样得到分数关联维数与正Lyapunov指数,说明制造质量水平的变化是一直是混沌的,为制造质量水平的预测在理论上提供了可能性。

The Chaotic Characteristic of the Variation of Manufacturing Quality Level Based on Time Series Analysis

At firstly,Manufacturing Enterprise Producing Time Series is defined.Then G-P algorithm is used for the calculation of correlated dimension.The embedded dimension is obtained throught the G-P calculation.After that little data method is used for calculating the Lyapunov exponent of the chaotic time series based on phase space reconstruction.The daily product qualified rate of Huananzhixin is collected as data source.The calculation result is that the correlation dimension is a fraction and Lyapunov exponent is positive,suggesting there exists a chaotic characteristic in the manufacturing quality level.Furthermore,8 time series are constructed based on the above daily product qualified rate time series,and fractional correlation dimension and positive Lyapunov exponent are also gotten for each time series,that is to say,manufacturing quality level is always chaotic.When the factors influencing the manufacturing quality level are concentrated by the reducing correlation dimension,product quality is out of control quickly by increasing Lyapunov exponent.