论文部分内容阅读
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。
Based on the definition of manufacturing manufacturing time series, G-P algorithm is used to calculate the correlation dimension of time series. After the corresponding embedding dimension is obtained through the correlation dimension calculation, the Lyapunov exponent of chaotic time series is calculated using the small data volume method based on phase space reconstruction. Acquisition HZ nearly three years of production on the pass rate as the quality of manufacturing changes in the quality of the chaos characteristics of the original data. On the basis of the above technical routes and data, the correlation dimension obtained is a fraction, while the Lyapunov exponent is positive, indicating that the time series of the pass rate of production products presents chaotic characteristics. In addition, the above data is divided into eight time series. Each time series also obtains the fractional correlation dimension and the positive Lyapunov exponent, which shows that the manufacturing quality level has always been chaotic, and it is theoretically possible to predict the manufacturing quality level Sex.