标题:
基于统计极值的流程对象时间序列时序计算算法A Novel Timing Calculation Algorithm Based on Statistical Extremum for the Time Series of Process Object
作者:
朱桐霖, 杜涛, 曲守宁, 朱连江
关键字:
流程对象, 数据挖掘, 时间序列, 统计, 极值, 延迟Process Object, Data Mining, Time Series, Statistics, Extreme Value, Delay
期刊名称:
《Hans Journal of Data Mining》, Vol.6 No.4, 2016-10-27
摘要:
本文针对流程对象采样数据集，提出了一种基于统计极值的流程对象环节间时序计算算法，同时通过理论分析证明了该算法的正确性。该算法通过取数据的特征点，计算环节间特征点的时间距，并通过统计方法，计算出流程对象任意两环节间的延迟时间，进而得到多环节间的时序关系。通过实际流程工业采样数据集测试，可基本准确的求得任意环节数据之间的延迟时间距以及各环节间的时序关系。
In this paper, an algorithm for computing timing relationship among each link of the process object is proposed, and the validity of the algorithm is proved through the theoretical analysis. The algorithm is designed based on statistical time distance among extremum points of sampling data set of the process industry, can calculate the delay time between any two time series, and then get timing relationship between any two links. At the same time, experiments with sampling data set of the process industry demonstrates that the algorithm can obtain the delay time interval among time series and the timing relationship between each link of process object.