Page 313 - Intelligent Digital Oil And Gas Fields
P. 313

Smart Wells and Techniques for Reservoir Monitoring          261


                               w
              iteratively samples em and minimizes the misfit between the dynamic smart
                               s,i
              well response and the observed production profile (e.g., water cut, pressure,
              GOR) that renders, for example, the modeled liquid rate profile that
              matches the well completion profile from the ICV segmentation. By further
              calculating the water flow rates per ICV segment, any type of water flooding
              optimization can be facilitated. To quantify the uncertainty in the smart well
              model response (i.e., the virtual PLT profile), the optimization method can
              further be integrated into multiple stochastic realizations of reservoir
              properties.



                   7.6 SMART WELL MODELING AND CONTROL
                   For wellbore modeling, the oil and gas well deliverability per segment
              can be calculated using inflow and outflow well performance analysis. Most
              of the commercial simulators can reproduce the effect of an ICV in a
              wellbore. Konopczynski and Ajayi (2004) have described the fundamental
              concepts to evaluate the well performance for single- or multiple-zone res-
              ervoirs using ICVs. The IPR and Vogel expressions can be easily used under
              the same assumptions that are used under radial and Darcy’s flow conditions.
              We show a methodology to estimate ICV performance as follows:
              •  Estimate maximum well rate potential from an IPR/Vogel model.
              •  Productivity index, PI¼Q/(P res   P wf ).
              •  Calculate the pressure drop across the choke (P out   P in ) required deliv-
                 ering the expected rate reduction/increase.
              •  Determine the flow trim characteristic by lab testing the flow coefficient
                 at each choke position of pressure drop.
              •  Carry out in the lab the mechanical design of the flow trim for the desired
                 well behavior.
              The Perkins model (1993) is the most applicable equation to predict the
              subcritical pressure drop across a valve using Eq. (7.3):

                                    Q   Q  ρ                   r  ffiffiffiffiffiffi
                                                                  1
                                         jj
                        P out  P in ¼           ¼ C  γ mix   Q tot        (7.3)
                                  Cv  Cv  ρ                       Cv
                                             wstd
              where
              P out is the flowing BHP at the outlet (psi) of the ICV,
              P in is the flowing BHP at the inlet of the valve (psi),
              C is the conversion factor,
              Q is the total fluids in USG/min,
   308   309   310   311   312   313   314   315   316   317   318