摘要
介绍了一种新的度量序列复杂性的统计方法——近似熵(ApEn),并将其应用于气固流化床流型识别中。通过分析和研究不同操作条件下气固流化床压力脉动时间序列,证明了近似熵与气泡行为有着良好的对应关系,能够较好反映床内复杂性的变化,可以作为有效辨识气固流化床起始鼓泡态、充分鼓泡态和湍动流化态等流型的特征参数。实验数据的计算和分析也表明了近似熵随数据长度的变化小,只需要较短的时间序列就能得到较为稳定的ApEn值;受数据中的瞬态噪声影响小,抗干扰能力强。是一种适合数据变化快、干扰大的工业现场测量的复杂性参数。
Approximate entropy (ApEn) was introduced as a new statistical method of complexity measurement for the research of pressure fluctuation time series complexity in gas-solid fluidized beds. It was proved that ApEn was consistent with the action of bubble by analyzing pressure fluctuation time series in gas-solid fluidized beds, and it could reflect the complexity of gas-solid fluidized beds, so it could be used as the characteristic parameter to identify different flow regimes of start-bubble state, fully-bubble state and turblent-bubble state. The calculation and analysis of the experimental data also conclude that ApEn changes little with the variety of the date length; shorter time series can get a stable ApEn; the influence caused by instant noise is smaller; the anti-interference ability is stronger. It is concluded that complexity parameter can be used in industrial implementation with quick various data and strong interference.
出处
《高校化学工程学报》
EI
CAS
CSCD
北大核心
2004年第3期281-286,共6页
Journal of Chemical Engineering of Chinese Universities
基金
国家自然科学基金(60075003)
