Page 197 - Petroleum Geology
P. 197

174

            where p*  is the reservoir pressure estimated from infinite closed-in time, and
            p  is  the pressure  measurement made at time  AT after  closing-in. When p  is
            plotted  against the logarithm  of  the dimensionless time  term, a straight line
            is  found.  This  method  is  due to Horner  (1951) and  such  plots  are  called
            Horner plots.
               If  the well  being  tested  is a new discovery well, the Horner plot gives two
            pieces  of  valuable information  (Fig. 8-16). The linear part of the plot extra-
            polates to infinite  time where the logarithm  of  the dimensionless time term
            equals zero; and this gives the initial reservoir pressure  at the level at which
            the  pressures  were  measured.  The slope of  this line is equal to Q7?0/4nh,t.
            After the properties  of the oil have been found from tests, only the effective
            permeability  to  oil  is  unknown,  so  it can be determined.  This value is the
            mean value for the reservoir in the drainage area of  the well.
               There is another point of interest to geologists. If the reservoir is bounded
            by a fault  or a marked change of permeability near the well within its radius
            of  influence, the plot shows two straight lines (Fig. 8-1 7), the second, at larger
















             Fig.  8-16. Horner  plot  of  pressure build-up in a well.  The linear  part extrapolates top*,
             which is the original pressure at depth of  measurement; reservoir parameters can be deter-
             mined from slope.

                        \           I                      \
                                                            P*

             i







             4          3           2          1           0
             Fig. 8-17. If  there  is a permeability barrier (such as a fault) near the well, the Horner plot
             shows two linear trends - the second theoretically with twice the slope of  the first.
   192   193   194   195   196   197   198   199   200   201   202