Basic HTML Version
OH – ST - ET: Analysis of Dynamic Data in Shale Gas Reservoirs – Part 2
p 12/18
Comparison of 10-year forecasts
4.2 - Discussion
For the same parameters and geometry, the productivity is better in the numerical model
compared to the analytical solution, thus requiring smaller fracture sizes and permeability to
match the same set of data. The same is observed when comparing 10-year forecasts
Explanation: Even though we use pseudopressures instead of pressure in the case of gas, we
still have to consider constant PVT properties at a reference pressure (generally the initial
pressure) for the other terms of the linearized diffusivity equation. Because of the low
permeability the pressure gradients are very high and the pressure substantially drops around
the wells. This creates a substantial increase of the gas compressibility which helps the
diffusion. This increase is not taken into account when using pseudopressures based analytical
models, and therefore the productivity is under-estimated compared to the “real” productivity
simulated by the numerical model.
While the straight line extrapolation was optimistic, the analytical model is pessimistic for
production forecasting.
However the analytical MFHW model was useful, as it captures the correct flow geometry and it
was useful to get a first parameter estimate for the numerical model, which should be
preferred for the 10-year forecast.
0
2000
4000
6000
8000
10000
12000
14000
16000
Gas rate [Mscf/D]
0
1E+9
2E+9
3E+9
4E+9
5E+9
6E+9
7E+9
Gas volume [scf]
Analytical MFHW
Equiv single fracture
Non linear numericalMFHW
0
10000 20000 30000 40000 50000 60000 70000 80000 90000 1E+5
Time [hr]
2000
4000
6000
8000
10000
Pressure [psia]
Production history plot (Gas rate [Mscf/D], Pressure [psia] vs Time [hr])