Powder Technology | 2019
Vibration multisensor fusion method for the identification and characterization of sand particles in dispersions of oil in water flow
Abstract
Abstract Transport of sand particles in dispersions of oil in water pipe flow is common in chemical and petroleum engineering. To overcome the existing limitations in the identification and characterization of sand particles with different concentration and size in multiphase flows, a vibration multisensor fusion method with three specially selected high-frequency vibration sensors for the detection of minuscule sand particles was evaluated. A series of detection methods, namely, the elementary sensor configuration and sand-wall impingement analysis, the time-frequency joint analysis, and the advantage sand multiband analysis methods, were developed to obtain the microvibration sand features submerged in strong background noise. Verification experiments were conducted to record vibration signals from the sand-carrying flow, thereby revealing different sand characteristic frequency bands in dispersions of oil in water flow combined with the advantage of the frequency response of each sensor: 20.8–21.9\u202fkHz and 23.55–24.85\u202fkHz for sensor-1, 30–31.25\u202fkHz for sensor-2, and 42.05–44.1\u202fkHz for sensor-3. Additionally, compared with the sand energy detection results of individual sensors, better identification results and good agreement for the detected sand energy were found under different conditions: sand concentration (0.05–0.25\u202fwt%), sand size (75–212\u202fμm), and oil content (5–20\u202fvol%). Finally, this method was verified and lays the foundation for the future detection of solids in more complex multiphase flows.