Open Access

ARTICLE

An Experimental Study on the Void Fraction for Gas-Liquid Two-Phase Flows in a Horizontal Pipe

by Li Lei¹, Jun An¹, Fushun Liang¹, Cheng Cheng¹, Naixiang Zhou², Yanhong Ning³, Jingzhi Zhang^1,4,*

1 School of Energy and Power Engineering, Shandong University, Jinan, 250061, China
2 Shandong Urban and Rural Planning and Design Institute, Jinan, China
3 China National Petroleum Corporation Jichai Power Co., Ltd., Jinan, China
4 Shenzhen Research Institute of Shandong University, Shenzhen, China

* Corresponding Author: Jingzhi Zhang. Email:

(This article belongs to the Special Issue: Recent Advances in Fluid Mechanics and Thermal Sciences)

Fluid Dynamics & Materials Processing 2021, 17(6), 1037-1048. https://doi.org/10.32604/fdmp.2021.016081

Received 04 February 2021; Accepted 26 May 2021; Issue published 08 September 2021

Abstract

The flow patterns and the void fraction related to a gas-liquid two-phase flow in a small channel are experimentally studied. The test channel is a transparent quartz glass circular channel with an inner diameter of 6.68 mm. The working fluids are air and water and their superficial velocities range from 0.014 to 8.127 m/s and from 0.0238 to 0.556 m/s, respectively. The void fraction is determined using the flow pattern images captured by a high-speed camera, while quick closing valves are used for verification. Four flow patterns are analyzed in experiments: slug flow, bubbly flow, annular flow and stratified flow. For intermittent flows (bubbly flow and slug flow), the cross-sectional void fraction is in a borderline condition while its probability distribution function (PDF) image displays a bimodal structure. For continuous flows (annular flow and stratified flow) the cross-sectional void fraction behaves as a fluctuating continuous curve while the (PDF) image displays a single peak structure. The volumetric void fraction data are also compared with available predictive formulas, and the results show that the agreement is very good. An effort is also provided to improve the so-called Gregory and Scott model using the available data.

---

Keywords

---