Wellbore pressure prediction after pipe connection operation in underbalanced drilling

Abstract

In underbalanced drilling (UBD), bottom hole pressure (BHP) must be maintained in the defined limit. Maintaining underbalanced drilling conditions after pipe connection operation is required for the success of underbalanced drilling operations. Modeling and simulation of gas-liquid two phase flow in an UBD operation is very significant in order to accurately predict the wellbore pressure and other parameters of two phase flow. After pipe connection operation in UBD, upon restarting mud circulation system, frictional pressure influenced the BHP and the fluid slugs in the drill string are transferred into the annulus. Therefore, the hydrostatic pressure will increase and UBD pipe connection operations create a BHP vibration, which is a critical point. This particular time can reduce the benefits obtained to drill the well in an underbalanced environment. In this study, a mechanistic steady state gas-liquid two phase flow model was used to simulate the two phase flow after pipe connection operation in UBD. Simulation was carried out to predict the parameters; such as wellbore pressure, liquid holdup, and velocities of the two phases at different flow patterns, namely slug, bubble, churn, dispersed bubble, and annular flow. In order to predict wellbore pressure, a steady state model was developed to predict flow patterns, pressure gradient, and liquid holdup for gasliquid flow in vertical annulus and drill string. The model included flow pattern transition models and hydrodynamic models for individual flow pattern. The model equations along with appropriate constitutive relations formed a system of coupled drift flux, momentum, and energy equations, which were solved using the well known iterative Newton Raphson method. All model equations were implemented in a computer program named Fortran 95. The effect of gas and liquid flow rates, and choke pressure on the wellbore pressure, particularly in the BHP was evaluated numerically. In order to validate the results of the developed model, they were compared with actual field data and the results of the WELLFLO software using different mechanistic models. The results revealed that the two phase model developed can accurately predict wellbore pressure, particularly BHP, wellbore temperature, gas/liquid velocities, and two phase flow patterns.