Horizontal oil–water two-phase flow measurement with information fusion of conductance ring sensor and cone meter

• Fusion of a conductance-ring sensor and a cone meter for two-phase flow measurement is presented.
• Information is fused for flow pattern identification and phase fraction estimation.
• Individual flow rate is estimated based on the combination of the two sensors.

Oil–water two-phase flow is a commonly seen flow phenomenon in petroleum and chemical industries. Online measurement of oil–water two-phase flow is a difficult problem due to the rheological complexity of its flow structure. This work presents a method of solving this measurement problem by combining a conductance-ring sensor and a cone meter to measure phase fraction, overall and individual flow rate of oil–water two-phase flow. This proposed method is based on the idea of information fusion between the conductance-ring sensor and the cone meter. Experiments were conducted on a multiphase flow loop, with the flow rate of water and oil ranges 3.9m3/h–10.2m3/h and 0.7m3/h–7.3m3/h, respectively. Phase fraction estimations with different correlations of the conductance-ring sensor calibrated under different typical phase distribution are fused at data level to deliver an estimation on volumetric phase fraction of water with an average of relative error εave=1.7%εave=1.7%. Flow pattern is found to affect the discharge coefficient Cd of the cone meter and therefore is identified with a physically meaningful feature obtained by fusing features from the conductance-ring sensor and the cone meter. Based on the accurate flow pattern identification, the overall flow rate of oil–water two-phase flow is estimated with an associate error εave=1.6%εave=1.6%, and the error of both the individual flow rate of water phase and oil phase is εave=2.8%εave=2.8%. This work provides a solution of how to combine the information from heterogeneous sensors for two-phase flow measurement, and can be extended to gas–liquid two-phase flow and also in vertical flows.

This present manuscript is about combining the information provided by a conductance-ring sensor and a cone meter for horizontal oil–water two-phase flow measurement. The information from the two heterogeneous sensors is fused at data level for phase fraction estimation. The discharge coefficient of the cone meter is found affected by the flow pattern of oil–water two-phase flow, and therefore the information from the conductance ring sensor and the cone meter is fused at feature level for flow pattern identification. The individual flow rate is then determined from the phase fraction estimated by the conductance ring sensor and the overall flow rate estimated by the cone meter, based on the flow pattern identification results.Figure optionsDownload as PowerPoint slide