Study On Flow Patterns Identification Method Of Two Phase Flow Based On Time Series Analysis

The conductance fluctuating signals of oil/water two phase flow in vertical upward pipes have been characterized using conventional and nonlinear time series analysis methods to discuss flow patterns identification of oil/water two phase flow. The main research results are summarized as follows:The conventional analyses of the conductance fluctuating signals include average value, standard deviation, kurtosis and so on. It shows that there are no correlation between these statistical feature parameters and the total flow rate and water cut. It is difficult to identify flow patterns using these parameters.Based on theory of Chaos and Fractal, K2 information entropy and Kolmogorov entropy are discussed. The calculation of K2 information entropy has been optimized by analysis of G-P algorithm. The optimal delay times have been obtained from the first infinitesimal value of mutual information.The K2 information entropy and D2 correlation dimension of fluctuating signals have been extracted. It is shown that the K2 information entropy and D2 correlation dimension are sensitive to flow patterns variations of oil/water two phase flow and they are sensitive 'indicators' of flow patterns variations.