中国民航大学学报 ›› 2026, Vol. 44 ›› Issue (1): 91-96.

• 基础研究 • 上一篇    

基于顺序熵的时间序列复杂性分析

  

  1. 中国民航大学 a. 民航航空器适航审定技术重点实验室;b. 理学院,天津 300300
  • 收稿日期:2024-03-04 修回日期:2024-06-19 出版日期:2026-02-28 发布日期:2026-03-07
  • 作者简介:董科强(1979— ),男,河北邯郸人,教授,博士,研究方向为复杂系统建模与分析.
  • 基金资助:
    天津市科技计划项目(23JCZDJC00070);民航航空器适航审定技术重点实验室开放基金项目(SH2020112701)

Complexity analysis of time series based on order entropy

  1. a. Key Laboratory of Airworthiness Certification Technology for Civil Aircraft; b. College of Science, CAUC, Tianjin300300, China
  • Received:2024-03-04 Revised:2024-06-19 Online:2026-02-28 Published:2026-03-07

摘要: 由于传统信息熵在分析时间序列复杂性时未考虑到先后元素的顺序关系,其分析结果存在一定局限性针对该问题,本文提出了一种改进的熵。首先,定义顺序熵来衡量先后元素顺序的不确定性;进一步引入联合顺序熵和平均顺序熵,用于研究两个时间序列之间的复杂依赖关系;然后,将这些熵度量方法应用于服从不同分布的随机序列及航空发动机性能参数序列,以验证改进方法的有效性。结果表明,该改进熵能够有效度量时间序列中的不确定性,且时间延迟对顺序熵和平均顺序熵的影响较小。

关键词: 顺序熵, 复杂性分析, 时间序列, 时间延迟, 航空发动机

Abstract: Traditional information entropy has certain limitations in analyzing the complexity of time series, as it does nottake into account the sequential relationship among successive elements. To address this issue, this paper pro-poses an improved entropy. First, order entropy is defined to measure the uncertainty of the sequence of succes-sive elements, and joint order entropy and average order entropy are further introduced to study the complex de-pendency between two time series. Then, these entropy -based measurement methods are applied to random se-quences following different distributions and to performance parameter sequences of aircraft engines to verify theeffectiveness of the improved method. The results show that the improved entropy can effectively measure the un-certainty in time series, and the influence of time delay on order entropy and average order entropy is relativelysmall.

Key words: order entropy, complexity analysis, time series, time delay, aircraft engine

中图分类号: