Quantitative flow visualization system for gas-liquid two phase flows

The main objective of this research was to develop a quantitative flow visualization technique to measure gas bubble size and velocities of vertical-up gas-liquid flows. A system to acquire high-speed digital images was designed and integrated with a hot-film anemometry system in the test section of the 76 mm flow loop at Memorial University of Newfoundland. Digital image processing algorithms were developed to obtain the gas bubble size and velocity information from the high-speed flow images. The gas slug and bubble velocities were estimated using two separate image processing algorithms: a supervised motion tracking algorithm and an edge detection cross-correlation algorithm. The supervised motion tracking algorithm allows the user to identify and track the movement of distinguishable gas bubbles and slugs. The edge detection cross-correlation algorithm uses standard edge detection routines to identify the boundaries of the slugs and bubbles. The bubble size information is obtained through morphological operations on the edge-detected images. The displacement of the bubbles between two frames is obtained through a cross-correlation analysis between the frames. Measurements were performed in the slug flow regime at several gas and liquid superficial velocities. Results obtained from the quantitative noninvasive flow visualization and image analysis techniques are in good agreement with results from simultaneous hot film anemometry measurements