Well-1: PNGE 411 - Petroleum Property Valuation
Well-1: PNGE 411 - Petroleum Property Valuation
2016/2017 Fall
Project - Part 1: Engineering Analysis (Solution)
qi
q= 1
(1 + btDi ) b
Adjusting the decline curve parameters b, qi , and Di resulted in following curves (note that logarithmic scale
on the y-axis was used for better visualization, but trends could have been visualized using a Cartesian scale,
as well):
1000
Well-1
qi = 325 STB/d
b = 0.9
Di = 0.4 1/month
100
10
27.12.1415.7.15 31.1.16 18.8.16 6.3.17 22.9.17 10.4.1827.10.1815.5.19 1.12.19 18.6.20
1000
Well-2
qi = 50 STB/d
b = 0.5
Di = 0.3 1/month
100
10
1
27.12.1415.7.15 31.1.16 18.8.16 6.3.17 22.9.17 10.4.1827.10.1815.5.19 1.12.19 18.6.20
1000 Well-3
qi = 550 STB/d
b = 0.2
Di = 0.2 1/month
100
10
1
27.12.1415.7.15 31.1.16 18.8.16 6.3.17 22.9.17 10.4.1827.10.1815.5.19 1.12.19 18.6.20
1
2. Probabilistic Volumetric Reserves Estimation:
The original-oil-in-place can be calculated from the following equation (assuming Vb is in ft3 , Bo is in rb/STB, N
is in STB):
Vb φ(1 − Swi )
N=
5.615Bo
Since the initial pressure is given as 1,500 psi, from fluid properties Boi is 1.215. To calculate the bulk vol-
ume using Area vs Thickness plot, first contour areas at real scale were calculated using the Planimeter unit
conversion ( 1 unit = 1,500,000 ft2):
30,000,000
AT1
25,000,000
20,000,000
Area, ft2
15,000,000 AT2
AR1
10,000,000
AT3
5,000,000
AR2R
AR3R AT44
-
0 5 10 15 20 25 30 35 40
Thickness, ft
2
Summing all areas would give us the total bulk volume:
3
X 4
X
Vb = ARn + AT n = 246, 000, 000+88, 500, 000+40, 500, 000+40, 500, 000+78, 750, 000+24, 000, 000+16, 200, 000
n=1 n=1
phi and Sw are given as uncertain parameters with specified distribution functions. φ has a triangular distribu-
tion and can be input to the Monte Carlo Simulation using the following distribution:
1. Calculate xr :
xm − xl
xr =
xh − xl
2. Generate 1000 rows of random numbers between 0 and 1 (can be done by using =RAND() formula in
Excel)
3. For each random number, rn , calculate xn which will result in 1000 values of each variable with a trian-
gular distribution: p
xl + (xm − xl )(xh − xl )(rn ) if rn < xr
xn =
x − p(x − x )(x − x )(1 − r ) if r >= x
h h m h l n n r
Sw has a normal distribution and 1,000 numbers with a normal distribution can be obtained from Excel’s random
number generator using the given mean and standard deviation values. The resulting distributions are given
below:
250 250
200 200
Number of Occurances
Number of Occurances
150 150
100 100
50 50
0 0
0.275
0.280
0.285
0.290
0.295
0.300
0.305
0.310
0.315
0.320
0.325
0.330
0.335
0.340
By inputting the calculated Vb and Bo as constants and inputting the generated Sw and φ values for 1,000
simulations, the resulting distribution for OOIP is the following:
3
Expectation Curve for OOIP
100%
90% P10 = 3,429,268 STB
80%
70%
Probability
60%
50% P50 = 4,310,478 STB
40%
30%
20%
P90 = 4,927,324 STB
10%
0%
3,000,000 3,500,000 4,000,000 4,500,000 5,000,000 5,500,000 6,000,000
OOIP
Comparison with observed production: We can sum observed daily averaged rates for each month, n, with
*30.4 days/month to obtain the cumulative recovery from the wells:
3 X
X 22
Np = qoin = 142,436 STB
i=1 n=1
Less than 2% error compared with the material balance calculation indicates consistent results.
At the bubble-point pressure, the recovery would be: