Oil-gas-water Three-Phase Flow Characterization and Velocity Measurement Based on Time-Frequency Decomposition
Xuewei Shi,Chao Tan,Feng Dong,Yuichi Murai
DOI: https://doi.org/10.1016/j.ijmultiphaseflow.2018.11.006
IF: 4.044
2018-01-01
International Journal of Multiphase Flow
Abstract:As a highly complex and random process, oil-gas-water three-phase flow is commonly encountered in petroleum production and transportation. Flow velocity of each individual phase is an important parameter for production estimation and flow assurance. To measure the phase velocity of three-phase flow, a one-side ultrasonic transducer working in continuous wave Doppler mode and a conductance sensor were combined to obtain the Doppler shift signal and the holdup signal. Firstly, the water holdup fluctuations were decomposed by the continuous wavelet transform to get the local wavelet energy (LWE) coefficients map on the time-frequency diagram to characterize the local flow structures of each typical flow regime. Since the ultrasound was reflected and scattered by both liquid droplets and gas bubbles, the Doppler shift signal contains fluctuations in different scales. The Doppler shift signal was then processed by the short-time Fourier transform to reveal the correlation between the time-varying velocity characteristics and the flow structure. Thirdly, the ensemble empirical mode decomposition was used to decompose the Doppler shift signal into different intrinsic mode functions (IMFs), and the first three IMFs were selected as the main IMFs. Finally, combined with the LWE coefficient map to jointly analyze the flow dynamic characteristics, the specific IMFs were found well-correlated with the flow information of the specific phases. The models for phase velocities extraction were then established. The extraction results were validated through dynamic experiments. The root mean square error of the liquid superficial velocity, the gas superficial and the overall superficial velocity are 0.06 m/s, 0.06 m/s and 0.07 m/s respectively. (C) 2018 Elsevier Ltd. All rights reserved.