Experimental Study of Aerated Mud Flows under Horizontal Borehole Conditions

Abstract

Extensive aerated mud experiments were performed in a unique field-scale elevated pressure and elevated temperature flow loop (6" × 3.5" annular test section, 73 ft length, horizontal configuration without drillpipe rotation). A view port was installed to observe flow patterns in the test section. Two nuclear densitometers were used to measure steady state mean void fraction. During test runs, the liquid and gas phase flow rates were in the range of 50–250 gal/min and 50–150 scf/min, respectively. For all the test runs, measurements of pressure drop and average liquid holdup over the entire annular section were carried out. The two-phase flow patterns were identified by visual observations through the view port. Stratified and slug flow were the two flow patterns observed over the range of the chosen test matrix. The presence of slug flow does not justify many existing simulation practices, which assume a homogeneous gas-liquid flow. A mechanistic model has been developed for aerated mud hydraulics based on conservation equations and existing two-phase pipe flow correlations. An extensive sensitivity analysis is presented to quantify the influence of mud properties and flow parameters on the bottom-hole pressure. Comparisons between the predictions of the model and experimental measurements show a satisfactory agreement. The present model is particularly suitable for the design of underbalanced coiled tubing applications. Introduction Coiled tubing has been shown to be a technically feasible method of drilling both oil and gas wells. One of the best applications of coiled tubing drilling technology is underbalanced drilling, to counteract the reduced weight on bit and annular velocities compared to conventional drilling. In some applications, aerated mud drilling has been recognized as having many advantages over conventional mud drilling, such as a higher penetration rate, less formation damage, reduced lost circulation, and lower drilling cost. Maintaining an optimum combination of liquid and air flow rates is important in aerated mud drilling operations. A useful prediction of the optimum combination requires knowledge of the flow pattern and determination of the properties of each phase under borehole conditions. Dukler and Hubbard1 developed a two-phase flow model based on phenomenological observations. Their study contributed a great deal of understanding about the mechanisms involved in hydrodynamic of slug flows. Based on their observations, Dukler and Hubbard defined an idealized slug unit and suggested a mechanistic model. It is the basis for a detailed mathematical model, which is capable for predicting hydrodynamic behaviors of slug flows, including the length, velocity, holdup and pressure distributions. A mathematical model for slug flows was developed by Ozawa and Sakaguchi2 based on the law of conservation of mass and momentum. This model predicts the position of a slug nose and tail, and pressure drop across the slug nose. The predictions showed a satisfactory agreement with experimental data. It was suggested that their approach could be used to study transient two-phase slug flows (solid-liquid or solid-gas flows). A Lagrangian approach was used by Gilchrist and Wong3 to study slug growth and acceleration effects associated with horizontal slug flows. The slug structure comprised of three sub-volumes: a liquid slug with entrained gas bubbles; a liquid film with a zero gas void fraction; and a gas bubble with zero liquid holdup. Mass and momentum equations were developed for the liquid slug and liquid film sub-volumes. An equation of state was used to formulate the changes in pressure in the gas bubble. A system of ordinary differential equations was coupled and solved simultaneously by a 4th order Runge-Kutta method. The simulated results for a particular slug unit showed fluctuations in the average slug velocity and void fraction in the liquid slug, and growth in the slug length as it traverses in a pipeline. Gomez and Shoham4 developed a mechanistic model for steady-state two-phase flows. The model is applicable for inclination angles from horizontal to vertical. The model predicts flow patterns, pressure drop, and liquid-holdup for stratified, slug, bubble, annular and dispersed bubbly flows.

A review of technological advances and open challenges for oil and gas drilling systems engineering

“…The importance of understanding the effects of these independent parameters on the efficiency of the wellbore cleaning process has been long recognised. 41,[151][152][153][154] Previous research contributions have mostly accounted for the influence of these parameters in groups, because, considering all parameters in a single study would be very cumbersome from both experimental and modelling perspectives. [155][156][157] Although the general term 'cuttings transport efficiency' has been used thus far, it worth mentioning a set of dependent parameters exist that reflect the efficiency of the cuttings transport process.…”

Gas-liquid flow regimes in horizontal annulus

“…Exactly 1174 experimental data points integrated from literature and experiments are harnessed to develop this model. Data was obtained from studies by Sunthankar et al,(2000), Zhou et al,(2004), Omurlu et al,(2006), Ozbayoglu & Omurlu (2007), Osgouei et al,(2010), Mendes et al,(2011), Gschnaidtner (2013), Nossen et al,(2017) and integrated with experimental data from the fully eccentric annulus setup. The developed flow regime transitions in this study is based on different annulus eccentricities, hydraulic diameters, fluid properties as well as gas and liquid superficial velocity ranges.…”

“…The 62 annulus setups were small and not likely to be encountered in the 63 oil and gas industry. 64 Zhou et al, (2004) performed extensive air-water experiments 65 in a large scale annulus with Do of 0.2124 m and Di of 0.089 m.…”

Hole Cleaning During Underbalanced Drilling in Horizontal and Inclined Wellbore

“…The local velocity profile is estimated by using the dimensionless velocity profile in the inner layer. The following expression for local velocity near the particle can be used (Zhou et al 2004b): Then, the minimum shear stress for rolling the particle can be estimated by combining Eqs. 2 and 7:…”

“…This two-zone structure for slug flow can be simplified to dispersed bubbly flow if the film zone length is zero. By using this structure, the frictional pressure drop calculation method (Zhou et al 2004b) can be applied to both slug flow and dispersed bubbly flow. Therefore, the frictional pressure drop in a slug unit can be expressed as follows: To have an effectively cleaned wellbore, the wall shear stress in a slug unit should be greater than min so that cuttings are swept out while a slug unit translating along the annulus.…”