Page 333 - Standard Handbook Petroleum Natural Gas Engineering VOLUME2
P. 333

300   Reservoir Engineering


                      Future  oil  production  and ultimate  recovery  are  often  extrapolated  from
                    graphical methods. One of  the more popular methods is a plot  of  the WOR
                    on a log scale vs.  cumulative oil production on a linear scale or a linear plot of
                    the fractional water cut (or percent water produced) vs. cumulative oil produced.
                    Alternatively, the oil-water ratio can be plotted on a log scale vs.  the cumulative
                    production on a linear scale. One of  the purposes of  these plots is to predict
                    the ultimate oil recovery by  extrapolating the curve to some economic limit at
                    which  time  it becomes  no  longer  profitable  to  continue  the  flood.  If  the
                    operating methods  remain  relatively unchanged,  a  method  [317]  has  been
                    proposed for a fully developed waterflood that permits an easy  extrapolation
                    of  recovery to a given water cut. This latter method consists of  a linear plot of
                    ql fractional recovery of  oil in place, vs.  the term - {[(l/fJ  - I]  - (l/fJ}.
                    This method also provides an estimate of water-oil relative permeabiliries.

                    Waterflood Parameters. Important parameters in waterflood operations are the
                    water residual oil saturation, Sm, and the relative permeability to -water, k,.   A
                    statistical study of  these parameters, as well  as peak oil rates, was provided by
                    Felsenthal [SlS].  Data on Sm and k,   from core data are listed in Table 545.
                    Endpoint krw values were  higher in  carbonates than in sandstones; for a given
                    lithology, k,  decreased as the absolute permeability decreased.



                                                 Table 5-45
                                           Watetflood Parameters

                          Waterflood  residual oll saturations measured in core test samples
                                                                               Carbonate
                                                           Sandstone             Rock
                    Mean average S,,   % pore space          27.7                 26.2
                    Median average  So,, % pore space        26.6                 25.2
                    Number ot core samples tested           316                  108
                    Number of  source reservoirs             75                   20
                    Standard deviation,  % pore space         8.76                 8.84
                                    Effect of lithology and K,  on the end points
                                      of oluwater relative permeabllity curves
                    Permeablllty Group          Low           Medium            High
                    range of 6, md            (1 to IO)      (I1 to loo)    (101 to 2,000)
                    Sandstone
                    Median kk at S,             0.065           0.133            0.256
                    Median kc at S,             0.033           0.095            0.210
                    Number of  core samples
                      tested                   30             21 3             143
                    Carbonate rock
                    Median k*,  at  So,         0.211           0.357            0.492
                    Median k;   at S,           0.179           0.503            0.428
                    Number of  core samples
                      tested                   33              4s               24
                    From Reference 318.
                      Expressed as fracton of   at connate water aaturabon.
                    ** Expressed as fractlon of  q.
   328   329   330   331   332   333   334   335   336   337   338