Pneumatic Transport Modeling of Air/Gas Drilling Hydraulics in Horizontal and Deviated Wellbores

Open Access
Graduate Program:
Petroleum and Mineral Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
April 29, 2015
Committee Members:
  • Michael Adebola Adewumi, Dissertation Advisor
  • Mku Thaddeus Ityokumbul, Dissertation Advisor
  • Michael Adebola Adewumi, Committee Chair
  • Turgay Ertekin, Committee Member
  • Russell Taylor Johns, Committee Member
  • Kenneth Steven Brentner, Committee Member
  • Yilin Wang, Committee Member
  • Robello Samuel, Special Member
  • Air Drilling
  • Drilling Hydraulics
  • Hole Cleaning
  • Optimum Rate
  • Circulation Time
  • Pneumatic Transport
  • Modeling
  • Deviated Wells
Optimal design of any drilling program requires a strong knowledge of drilling hydraulics. Field drilling technicians and drilling engineers designing air/gas drilling projects both typically need to know (1) the time it takes to circulate out a pill of formation material at any given depth in order to reduce drilling time (thereby reducing the cost of drilling); and (2) the optimum rate at which to flow the circulation fluid to ensure the hole is maintained free of cuttings at the minimum capacity (and thereby cost) of surface equipments. While there have been substantial studies and some established techniques for wellbore hydraulics design/analysis for conventional mud drilling, this is not the case for air/gas drilling. To this end, a transient numerical pneumatic transport model has been developed to study air/gas drilling hydraulics. This model is based on a mono-dispersed particulate two-fluid model (TFM). The mathematical representation of the problem involves the solution to coupled non-linear hyperbolic PDE’s and therefore a detailed review of monotone and higher resolution schemes for hyperbolic systems is presented. Two hyperbolic model equations: (1) The benchmark Euler equation (2) The mono-dispersed particulate Two Fluid Model (TFM) equation, are presented along with analytical expressions for their decomposed characteristics. The Euler equation is used to demonstrate the efficacy of the numerical schemes discussed. The most efficient numerical scheme is then applied to solve the TFM equation. Numerical results compared to experimental data were satisfactory. The experimental work of Temple was used to evaluate the model for annular vertical wellbore air drilling flow conditions. Flow characteristic curves (FCC) that cover both dilute and dense phase flow conditions were generated. Simulation results showed a good match with experiments. While matching the data points in the dilute phase flow region was less precise towards the extreme end of the larger particles, the optimum flow velocity was still accurately predicted at 9.1 %AAD. Unlike vertical flow, air drilling experiments for horizontal flow pneumatic transport are difficult to come by. This applies to industry field data as well as they are usually incomplete and typically labeled as confidential information. Therefore horizontal pneumatic transport studies for food processing applications were used to evaluate the performance of the current model for horizontal and highly-deviated well systems. The area ratio correlations were incorporated into the TFM to account for the lower-side deposition and re-entrainment phenomenon peculiar to horizontal wells. While we were able to show a match in the FCC plots in the dilute flow region, it was difficult to establish a match for the dense phase region of the FCC plot. Predicted pressure drops in dense flow region of the FCC plot for horizontal systems were significantly lower than experiment. This was due to the inability of the current model formulation to account for slug flow regime typical of horizontal pneumatic transport systems transitioning from dilute to dense flow conditions. Overall, it was observed that the model is capable of predicting volumetric rate requirements and circulation time for hole cleaning in vertical wells. The model is also capable of estimating cutting circulation time in horizontal and deviated well sections while in dilute flow conditions only. However, the model was unable to accurately simulate the inherent transient behaviors expected under dense horizontal pneumatic transport. More studies need to be conducted in the area of constitutive relations specific to horizontal pneumatic transport flow regimes and empirical relations for the prediction of critical velocities. These will provide the capability of upgrading the current model for the prediction of fluid volumetric rate requirements in horizontal/highly-deviated well sections particularly under dense flow conditions. The transient model was also used to predict circulation time for dispersed slugs of cuttings from the bottom of a well with both vertical and horizontal sections. The model results can only be trusted to predict accurately when the flow in the horizontal section is maintained within the dilute flow region. In order to ensure that the flow in the horizontal section is maintained above flow condition, we developed correlations for optimum circulation velocity using three distinct data sets for both vertical and horizontal pneumatic transport systems in the literature. Our results showed a relative error less than 10 % for most data points and RMSRD less that 5.1 %.