The dynamics Shannon entropy Hk of the signal of time series is applied to nonlinear time series analysis because of less data demand and simpler calculation. It was used to analyze the signals of two-phase flow conductance. Firstly, the selection criteria of the parameters were analyzed through the application in white Gaussian noise. Secondly, the effect of different parameters in the two-phase flow pattern identification was discussed, and the Hk-up of flow pattern were compared with three typical simulated signals. Finally, the signals of two-phase flow conductance were analyzed by calculating the segmented Hk-up, and the evolutions of various flow patterns were analyzed. The dynamics Shannon entropy was an effective method for the analysis of two-phase flow conductance signals, different flow patterns could be distinguished and the evolution characteristics of different flow patterns could be identified.%时间序列的符号动力学信息熵Hk因其计算简单快速,对数据量要求小,而被应用于非线性时间序列的分析中.将其应用于气液两相流型电导信号的分析中,首先通过其在高斯白噪声中的应用,来分析参数的选择标准.接下来讨论不同参数选择对气液两相流型识别的影响,并将流型信号的Hk-up同3种典型仿真信号进行对比.最后采用分段计算Hk-up的方法对气液两相流型电导信号进行分析,剖析了各种流型的演化规律.分析结果表明:符号动力学信息熵是一种有效的分析气液两相流电导信号的方法,能够有效区分不同流型,并能够反映不同流型的演化规律特性.
展开▼
