Summary For ultradeep gas wells within narrow pressure windows, conventional shut-in well control may lead to excessive casing pressure, delayed gas removal, and increased lost-circulation risk. Existing predictive methods are also limited because gas dissolution, transient annular flow, and managed pressure drilling (MPD) pressure control are often treated separately. To address these deficiencies, we develop an integrated modeling framework for MPD-assisted dynamic well killing in ultradeep wells. The framework couples (1) equation-of-state (EOS)-based methane (CH4)/carbon dioxide (CO2) dissolution and breakout calculations for water-based and oil-based drilling fluids, (2) a transient annular drift-flux multiphase-flow model, (3) local dissolved/free-gas partitioning based on pressure-dependent solubility with a prescribed quasisteady wellbore temperature profile, and (4) continuous-circulation well killing simulation with explicit surface-backpressure (SBP) control. Four thermodynamic models are first evaluated against experimental solubility measurements for actual drilling fluids over 293.15–423.15 K and 0.101–70 MPa. The Peng-Robinson (PR) EOS gives the best overall performance for water-based fluids, whereas the Redlich-Kwong (RK)-Aspen model performs better for oil-based fluids. The coupled multiphase-flow framework is further validated against field SBP data from an ultradeep well, showing good agreement at the engineering level. The results show that gas dissolution can substantially alter early kick signatures and well-killing dynamics. In oil-based drilling fluids, strong dissolution delays gas breakout, weakens early surface responses, prolongs circulation, and changes the evolution of bottomhole pressure (BHP) and SBP. Sensitivity results indicate that, within the tested ranges, formation pressure, permeability, and exposed pay length exert the strongest influence on influx severity and pressure-control demand, whereas higher mud density and circulation rate improve controllability. Two field applications show that, relative to conventional shut-in well control, MPD-assisted dynamic well killing reduces average handling time by approximately 56–66%, decreases gas influx volume by about 65–69%, and lowers peak surface/casing pressure demand by about 3–8 MPa. The proposed framework provides a practical basis for kick interpretation, pressure prediction, and operational design in deep and ultradeep well control.
Peng et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: