IDENTIFICATION OF WAVY FLOW SUB-REGIMES USING POWER SPECTRUM OF LASER DOPPLER VELOCIMETRY SIGNALS

Abstract

The formation of a slug happens when the roll waves on the interface of two-phase flow undergo constructive interference. Slug flowing with high momentum results in flow-accelerated corrosion (FAC) leading to cracks in pipes. The leakages from such cracks are hazardous to industry. In order to avoid such hazards, a detailed analysis is required for the transition of smooth stratified flow to slug flow. This transition happens through different wavy flow sub-regimes. Previous studies in literature used visual observation and image processing to determine sub-regimes of wavy flow. However, visual observation alone is not sufficient to differentiate between the different sub-regimes of wavy flow. In the present study the power spectrum of the velocity signals measured using laser Doppler velocimetry (LDV) is used in addition to visual observations to develop a precise wavy-flow sub-regime map. Experiments are conducted in a two-phase flow test rig by varying liquid superficial Reynolds number in the range of 1685 and 6830 and gas superficial Reynolds number between 393 and 9824. The transition to wavy flow is observed much earlier in the present study compared to that reported in literature based solely on visualization methods. The present paper successfully utilized power spectrum analysis to accurately and precisely categorize wavy flow sub-regimes, including the creation of a flow subregime map demarcating "3D large-amplitude waves (3DLA)," "2D small-amplitude waves (2DSA)," "3D small-amplitude waves (3DSA)," and "2D large-amplitude waves (2DLA)." The modified flow-pattern map developed based on power-spectrum analysis of velocity signals and the flow visualization is compared with the flow-pattern maps available in literature.