Page 258 - Standard Handbook Petroleum Natural Gas Engineering VOLUME2

Mechanisms & Recovery of Hydrocarbons by Natural Means 227

Empirical Estimates of Primary Oil Recovery

Several attempts have been made to correlate primary oil recovery with
reservoir parameters [239-2421. Based on field data [241] from water-drive
reservoirs, a statistical study [242] yielded the following empirical relationship
for primary oil recovery:

N, = (0.271910g k+0.25569Sw +0.135510gp, -15,380$
B, ’I
- 0.00035h+ 0.11403) 7,758Ah$A (5-141)

where NP is oil production in STB, k is permeability in md, Sw is fractional water
saturation, po is oil viscosity in cp, $ is fractional porosity, h is pay thickness in
ft, A is a real extent in ft5, and Boi is the initial formation volume factor of oil
in reservoir barrels per STB. Based on the first API study [240], correlations
were developed for recoverable oil. For solution gas drive reservoirs, the
recoverable oil (RO) in stock tank barrels per net acre-ft was:
[Sw]o’37a[ E]
0.174
1.1611
0.0979
RO = 3,2M[ MI- S,) ] [ 41
(5-142)
Bob p ob
For water drive reservoirs, the correlation was:
RO = 4,259[ $(I- s, ’ ] [ ”1 0.0770 [Sw]4”go3[ E]
1.0422
4.2159
(5143)
B oi CI oi

In the second API study [239], analysis of 116 solution gas drive reservoirs gave
the following equation:
RO = 6,533[ @(I- S,) ] [ r] [Sw]o.46s[ y]
0.0816
0.249
1.312
Bob Pch (5-144)
However, the second study concluded that none of the equatirms developed in either
study was statistically appropriate to provide a valid correlation. Furthermore, no
statistically valid correlation was found between oil recovery and definable
reservoir parameters. The second study found that when reservoirs were separated
by lithology, geographical province, and producing mechanism, the only reason-
able correlations that could be developed were between recoverable oil and
original oil in place. Even then, the correlations were of poor quality as indicated
by Figure 5-139 which presents the best correlation for Texas sandstone natural-
water-drive reservoirs. The average primary recovery for various groups of
reservoirs at the average value of OOIP for each group is listed by production
mechanism in Table 5-29 [239,243].
In view of the lack of suitable correlations, primary oil recovery for an
individual reservoir must be estimated by one of three methods: (1) material
balance equations in conjunction with equations for gas-oil ratio and fluid