Analysis Of Characteristics Of Gas-Liquid Two Phase Flow Based On Complexity Entropy

Gas-liquid two phase flow is widely spread in many industry fields. It is close to human lives and producing. As there exists complex interphase interface and relative motion of two phase flow, so accurate identification of flow pattern is still difficult. Especially, the dynamic mechanism of the flow pattern change is still unclear. Entropy is a measure of complexity and regularity of the system. The research based on complexity entropy has been public developed, and a great improvement has been made on the characterization of the nonlinear system. Complexity entropy can not only dig the important message included in the nonlinear time series, but also research some complexity nonlinear dynamics system which can not be accurately described by the theoretical model. Thus complexity entropy has been a useful tool to describe the complexity of the system. Gas-liquid two phase conductance fluctuation signals is a complexity and nonlinear dynamics system. The research on the two phase flow based on the complexity entropy has some development.Wavelet multi-scale transform is mainly used to decompose the two phase flow conductance fluctuation signals in this paper, and then we can get wavelet multi-scale coefficients, which can be used to analyse the frequency of three typical flow pattern. Wavelet entropy values and multi-scale entropy values are calculated based on the wavelet coefficients, at the same time feature vector of flow patterns are got. Through the analysis of the wavelet entropy and the multi-scale entropy, we can finish the research the dynamic characteristic and nonlinear of the two phase flow. From the figures of wavelet entropy and multi-scale entropy, we can observe the fluctuation status and nonlinear characteristic, and recognize the flow pattern clearly. Besides, approximate entropy(ApEn), sample entropy(SampEn) and fuzzy entropy (FuzzyEn) these three entropies are all used in this paper. The three entropies are sensitive to the two phase flow. From the comparison of the three entropies, we can find the nonlinear characteristic of the two phase flow. And we also find fuzzy entropy is a more useful chaos complexity measure algorithm than other two,even better than wavelet entropy. Fuzzy entropy is more sensitive to the gamma space dimension (m), similar tolerance(r) and sequence length (N), better robustness and measure value continuities. After many calculation and some simulation, we find complexity measure way is a good way to performance nonlinear and dynamics, and a useful diagnose way to understand two phase flow dynamics characteristic system.