Adaptive Kalman Estimation of Phase Holdup of Water-Continuous Oil-Water Two-Phase Flow

Phase holdup measurement of the two-phase flow is of significance in industrial process. Accurate and real-time measurement of phase holdup is a critical problem requiring urgent solutions. Conductance sensors determine the flow parameters by detecting the change of electrical parameters. Kalman filter is widely applied in state estimation, and oil-water two-phase flow holdup measurement is considered as a state estimation problem in this paper. By using upstream and downstream measuring fluctuation information, an adaptive Kalman estimation model based on discrete cross-correlation with adaptive state transition matrix, and a serial estimation model based on unscented Kalman filter carrying out linear and nonlinear estimations successively are proposed combining with specific flow state. The proposed methods have been validated by online experiments, and the average error of phase holdup estimate is 1.5%.