Flow regime identification of gas-liquid two-phase flow based on higher-order statistics

Based on higher-order statistics(HOS), a new method for flow regime identification of gas-liquid two-phase was presented. Differential pressure fluctuation signals obtained from the differential pressure transducer in the horizontal pipe were analyzed by using bispectral. method and the non-Gaussian character of different flow regimes could be extracted exactly. The average bispectrum amplitude based on higher-order statistics of different pressure fluctuation signals was regarded as an eigenvalue. Meanwhile, the features of bispectrum of the differential pressure fluctuation signals of different flow regimes were discussed. The experimental results show the extent of different pressure fluctuation signals deviation from Gaussian distribution is different for different flow regimes and the proposed method is effective for flow regime identification of the gas-liquid two-phase in horizontal pipe.