Page 251 - Fundamentals of Enhanced Oil and Gas Recovery
P. 251

Enhanced Gas Recovery Techniques From Coalbed Methane Reservoirs                    239


                   Coal permeability, similar to other reservoirs, is dependent on the effective stress onto
                   the reservoir that is a function of reservoir depth and pressure differential. However,
                   unique to coals, the gas desorption process is also influential in determination of the
                   permeability value of the reservoir at any given time. In fact, these two effects function
                   competitively. The effective stress increase due to the reservoir pressure drawdown obvi-
                   ously imposes an adverse effect on the cleat permeability through narrowing the fissures.
                   On the other hand, the gas desorption as a response to reservoir pressure drawdown
                   causes the reservoir matrix to shrink, thereby increasing the cleat permeability.
                   Consequently, the interaction of these two phenomena governs the shape of the cleat
                   system and determines permeability value at any given condition [32].
                      The outcome of this interaction might increase the permeability value up to 100
                   times, being the case in San Juan basin, USA. This occurrence is in a sharp contrast
                   with conventional reservoirs in which the reservoir depletion leads to a decrease in
                   absolute permeability [33]. In bituminous coals, the typical porosity is around 1%, and
                   99% of the reservoir volume is accounted for by matrix [2]. Under such condition,
                   given the cubic relationship between permeability and porosity, an increase in cleat
                   porosity from 1% to 2% in response to reservoir depletion and corresponding matrix
                   shrinkage would result in an eightfold rise in permeability value of the reservoir. It
                   goes without saying that for lower rank coals, in which the initial porosity is far more
                   than 1%, the same increase in porosity (1%) results in smaller increase in permeability
                   measure. Therefore, matrix shrinkage has a more significant effect on permeability
                   measure in high-rank coals. Moreover, apart from the positive matrix-shrinkage effect
                   on cleat absolute permeability, such phenomenon favors the relative permeability to
                   gas, because for the same amount of water inside the cleats, an increase in porosity
                   lowers the water saturation percentage.
                      There has been some models presented to describe the behavior of coal permeabil-
                   ity with respect to the effect of mentioned influential factors on this parameter within
                   the reservoir depletion. One of the most widely used permeability models, which is
                   based on the cubic relationship of permeability and porosity, was suggested by Palmer
                   and Mansoori and is shown below after solving for the axial modulus and the relation-
                   ship between axial and bulk modulus [34]:
                         φ       ð1 1 νÞð1 2 2νÞ         c o 2ð1 2 2νÞ     p i  p
                         φ i  5 1 1  ð1 2 νÞEφ i  ðp 2 p i Þ 1  φ 3ð1 2 νÞ  p i 1 p l  2  p 1 p l  (8.3)
                                                          i
                   where ν is the Poisson ratio, E is the Young modulus, p is the pressure at any given
                   time, p i is the initial pressure, c o is the volumetric strain coefficient, and p l is the
                   Langmuir pressure.
                      The second term on the right-hand side of the equation represents the effects of
                   stress on the porosity, and the third term illustrates how porosity is affected by matrix
                   shrinkage. While during natural depletion, the third term of this equation is always
   246   247   248   249   250   251   252   253   254   255   256