Page 326 - Standard Handbook Petroleum Natural Gas Engineering VOLUME2
P. 326

Fluid Movement in Waterflooded Reservoirs   293


                  where N   = oil potentially recoverable by  waterflooding, STB
                         Pj = initial oil in place, STB
                         NP = oil produced during primary operations, STB
                         Boi = initial FVF
                  These equations can be altered to include a residual gas saturation, if  present.
                  The volumetric  sweep efficiency can be  estimated from  one  of  the  correla-
                  tions given previously or can be obtained from an analogy from similar water-
                  flood projects.

                  Prediction Methods

                    An  extensive survey on prediction of  waterflood performance was  provided
                  by  Craig  [133].  Of  the  methods  reviewed,  three  appeared  most  promising:
                  (1) the  Higgins-Leighton streamtube model  [309],  (2)  the  Craig, Geffen, and
                  Morse  model  [282],  and  (3) the  Prats  et al.  method [283].  Discussion of  the
                  various prediction methods is beyond the scope of  this text, and only two very
                  simple methods will  be  presented for  illustrative purposes.  Both  the  Dykstra-
                  Parsons [294] and Stiles [310] methods are very cursory and, if  used, they are
                  normally followed by more extensive evaluations, usually by  computer simulation.
                    For  either  the  Dykstra-Parsons or Stiles methods,  the  permeabilities  are
                  arranged in descending order. For the Dykstra-Parsons method, the permeability
                  variation is  determined as described earlier. Two  options are then possible:  a
                  progrim [311] for hand-held calculators can be used, or the graphical technique
                  presented by Johnson  [296] can be used. The fractional recovery, R, (see Figure
                  5-164 for example) expressed as a fraction of  the oil in place when the waterf-
                  lood is  started, muii be multiplied by  the areal sweep efficiency (for example
                  from Figure 5-161) to  obtain the waterflood recovery.
                    For the Stiles technique, a program [312] for hand-held calculators is available
                  or the  procedure  summarized in  Table  5-43 can be  used.  A  straightforward
                  presentation of  the Stiles method is in the text by  Craft and Hawkins [17]. The
                  fractional recovery obtained with the Stiles method is a fraction of  the recover-
                  able oil (ST  - Sm) that has been recovered at a given reservoir water cut. Since
                  a water-oil ratio (WOR) is measured at surface conditions, the fractional water
                  cut at reservoir conditions, fw, is obtained (assuming B,  =  1.0) from:


                           WOR
                    f,  =                                                       (5-228)
                         WOR + Bo
                  where Bo is  the oil formation volume factor.

                  Performance Evaluation
                    Monitoring waterflood performance is  crucial to  the  success of  the  flood.
                  From  a  reservoir  engineering  standpoint,  the  primary  concerns  are  water
                  injectivity and oil productivity. A few important factors related to these concerns
                  will be summarized.

                  Injectivlty and lnjectivlty Index.  Whereas productivity index was  the  ability
                  of a well to produce hydrocarbons, injectivity index, I, in B/D/psi,  is a measure
                  of  the ability of  a well  to accept fluids [17]:
   321   322   323   324   325   326   327   328   329   330   331