Gas lift optimization and flow instability

Elmabrok, KMO 2017, Gas lift optimization and flow instability , PhD thesis, University of Salford.

Download (3MB) | Preview
[img] Microsoft Word - Submitted Version
Restricted to Repository staff only

Download (7MB)


Gas lift is an artificial lift method used in the oil industry to lift reservoir fluid to the surface, by supplementing the reservoir pressure when it is depleted or insufficient. During oil production, this method can be affected by two-phase (gas-liquid) flow instability within the production tubing, which results in a reduction in the total oil recovered. There are three main flow instabilities caused by, the density wave oscillation, the casing heading pressure and the flow perturbation within the two-phase flow regime.

Within this investigation of the flow structure, behaviours and instability of two-phase flow have been investigated experimentally using a high-speed motion imaging with a dedicated processing package “Dynamic studio 2015a” in a vertical transparent pipe (ID: 66 mm, Length: 2 m) thus simulating the prototype sizing of the common artificial gas lift. Numerically a Computational Fluid Dynamics (CFD) models were used with air and water as the working fluids for various cases.

The experimental results demonstrated that initial bubble size plays a major role in the development and instability of the upward two-phase flow in the vertical pipe. A new Multiple Nozzle Injection Technique (MNIT) with the aim of reducing initial bubble size and distribution across the simulated vertical column was also utilised, thereby stabilising the gas lift system. Thus the present findings are compared with the current Single Nozzle Injection Technique (SNIT) (or so- called sharp-edge) that are utilised in normal gas lift operation. It has thus been manifested that the new method has the potential to increase the total oil production rate from gas lifted wells. It was found that this new injection technique reduced the overall average bubble size from 7.01 to 5.47 mm and the average overall minimum bubble size from 1.23 to 1.03 mm. The average large bubble size of the Taylor bubble was also reduced from 44.07 to 39.95 mm in the simulated pipe. This perceived to increase in production rate from 40 to 43.05 l/min, which give overall increment of 7.5% at different operating conditions. This is in comparison with the single orifice injection technique at the same operating conditions. Throughout this investigation, water was used as working fluid since the column of corresponding water in the petroleum production tubing has the highest hydrostatic pressure 0.20 bar compared with crude oil. Hence, during the gas lift process crude oil will be less cumbersome to produce than water.

Moreover, it was found that when using the Multiple Nozzle Injection Technique the distribution of gas bubbles could changed from the middle of the vertical pipe (core peaking) to across the entire pipe area (wall peaking). This minimised the two-phase flow development and flow instability, even when the mixture velocity was increased. This was due to a reduction in the coalescence process of the gas bubbles as a result of improved bubble distribution when compared with the Single Nozzle Injection Technique with the same dimensions.

The numerical three-dimensional CFD model using the multi-fluid volume of fluid (VOF) gas-lift with the same dimensions and operating conditions compared qualitatively with bubble distribution similar to those found by experimental trials. In addition, the pressure drop long the simulated test section was calculated numerically. It was also found that the pressure drop was reduced from 0.18 bar to 0.11 bar when the new MNIT was used as compared with the SNIT that are normally used in gas lift operation practise.

Item Type: Thesis (PhD)
Schools: Schools > School of Computing, Science and Engineering
Depositing User: KMO Elmabrok
Date Deposited: 22 Feb 2018 15:50
Last Modified: 27 Aug 2021 20:43

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)


Downloads per month over past year