An integrative image measurement technique for dense bubbly flows with a wide size distribution. (27th January 2015)

• Record Type: Journal Article
• Authors: Karn, Ashish; Ellis, Christopher; Arndt, Roger; Hong, Jiarong
• Is Part Of: Chemical engineering science. Volume 122(2015)
• Page Start: 240
• Page End: 249
• Publication Date: 2015-01-27
• Subjects: Image analysis -- Bubbly flow -- Bubble size distribution -- Shadow image velocimetry -- Ventilated hydrofoil wake -- Synthetic bubble images; Chemical engineering -- Periodicals; Génie chimique -- Périodiques; Chemical engineering; Periodicals; Electronic journals; 660
• DOI: 10.1016/j.ces.2014.09.036
• Languages: English
• ISSNs: 0009-2509

Abstract

Abstract: The measurements of bubble size distribution are ubiquitous in many industrial applications in chemical engineering. The conventional methods using image analysis to measure bubble size are limited in their robustness and applicability in highly turbulent bubbly flows. These flows usually impose significant challenges for image processing such as a wide range of bubble size distribution, spatial and temporal inhomogeneity of image background including in-focus and out-of-focus bubbles, as well as the excessive presence of bubble clusters. This article introduces a multi-level image analysis approach to detect a wide size range of bubbles and resolve bubble clusters from images obtained in a turbulent bubbly wake of a ventilated hydrofoil. The proposed approach was implemented to derive bubble size and air ventilation rate from the synthetic images and the experiments, respectively. The results show a great promise in its applicability for online monitoring of bubbly flows in a number of industrial applications. Highlights: A robust image analysis approach for highly turbulent bubbly flows is proposed. It can resolve both in-focus and out-of focus bubbles over a wide size range. It can segment individual bubble from large clusters in high void fraction images. The approach was validated using both synthetic bubble images and experimental data. It allows real time analysis of two-phase flows in many industrial applications.