Rana Rasheed Majeed

RESERVOIR ENGINEERING
Permeability prediction of coalbed methane reservoirs during

primary depletion
A Report Submitted to the Petroleum Engineering Department of the University of Kerbala
By:- rana Rasheed majeed
"THird Stage "
Supervised
DR. Tahseen Al-Taei.
1
CONTENTS.
1 .Introduction
2. methodology
3. conclusion
4.REFERENCES
2
1 .Introduction
Permeability increase in coalbed methane (CBM) reservoirs during

primary depletion, particularly in the San Juan Basin, is a well accepted
phenomenon. It is complex since it is influenced by stress conditions and
coal matrix shrinkage associated with gas desorption. Understanding the
variations in coal permeability is critical in order to reliably project future
gas production, or consider other gas migration issues in the reservoir.
Since sorption-induced strain plays a critical role in changing the
permeability, typically observed, the theoretical strain model should be
incorporated into the permeability prediction models. An effort is made in
this paper to couple the recently developed Liu and Harpalani sorption-
induced strain model with various permeability models. The model first
calculates the theoretical coal matrix shrinkage strain and, using the
calculated strain, various commonly used permeability models are
applied to two sets of field data. The results of the coupled models show
that the agreement between the predicted permeability and that observed
in the field is very good. The merit of the coupled models is that it can
theoretically predict the permeability with less experi-mental work,
making it a more time efficient and economical technique compared to
models used in the past.
Coalbed methane (CBM) refers to gas formed as a part of the geological

processes of coal formation, and is contained in varying quantities within
all coals. With increased energy demand and signif- icant advances made
in development of CBM industry during the last three decades in the
United States, CBM has become a resource of global significance, with
emergence of active CBM plays in China, Canada, Australia and India.
CBM reserves and production in the US are both almost 10% of the total
3
values (Palmer, 2010), which is sig- nificant. The rate at which CBM
production has increased, from near zero in 1980 to more than 1.9 trillion
cu ft (TCF) in 2009, is just as striking. Scientific understanding of, and
production experience with, coalbed methane has matured significantly in
the last decade although, relatively speaking, CBM industry is still young
and further studies will continue in order to better understand the
production behavior. Compared to con-ventional natural gas reservoirs,
CBM reservoirs have several uniquecharacteristics:
(1) very low effective porosity/permeability, which often hinders the

development of CBM plays.
(2) adsorption storage mechanism since most of the methane produced in

coal is stored in the micropores in adsorbed form (Gray, 1987).
(3) desorption-induced matrix shrinkage since this results in opening up

of the cleat which, in turn, results in increased fracture permeability in the
low pressure range.
(4) the ‘negative decline curve’, that is, production increases at the
beginning and then slowly decreases over time (Harrison and Gordon,
1984), which is a distinct feature of CBM production wells.
This paper is aimed at incorporating the recently developed matrix

strain model into the existing and widely used analytical permeability
models and evaluating its performance. Following this, effort is made to
compare the modified permeability model predic-tions against published
field permeability increase data. It provides a basis for the different CBM
simulators to theoretically incorporate the sorption-induced strain into
establishing the changes in perme-ability during history matching and
projection of gas production. Italso provides a theoretical basis to
improve the performance of CBM reservoir simulators with only basic
geomechanical and adsorption parameters as input. These parameters are
either available in geologic report or easy to obtain through experiments.
Hence, the coupling effort of the strain and permeability model will avoid
to carry out expensive field tests and/or time consuming experiments, for
instance,steady-state shrinkage and swelling tests.
2. methodology
The first reported study of coal matrix volumetric response to sorption of
gas can be traced back to Moffat and Weale (1955). They observed 0.2%
4
to 1.6% volumetric increase when coal blocks were subjected to methane

pressure of up to ~15 MPa. However, for methane pressure between 15
and 71 MPa, the volume of coal block either decreased or remained
constant, which was attributed to the changes in the solid volume of coal,
effectively the grain compressibil-ity effect.
After initiation of the CBM industry in the early 1980's in the US, and
availability of production data for a few years, coal matrix shrinkage
research gained a great deal of attention due to its impact on changes in
the cleat/fracture aperture, inducing a considerable increase in the cleat
permeability during later part of life of CBM wells. The coal matrix
shrinkage and swelling strain has been quantified in the laboratory by
several researchers (George and Barakat, 2001; Harpalani and Chen,
1997; Harpalani and Mitra, 2010; Harpalani and Schraufnagel, 1990;
Levine, 1996; Robertson, 2005; Seidle and Huitt, 1995; Wang et al.,
2011). In each one of these laboratory studies, matrix volumetric in-
crease/decrease with methane adsorption/desorption was reported.
However, no theoretical explanation to unlock the relationship between

volumetric strain and sorption of gas was presented, making it difficult to
integrate the actual effect of the phenomenon into CBM reservoir
permeability prediction models.
There is a lot of talk about the mathematical equations for this topic, but
we will try in this part to explain the mathematical equations for the most
famous models to help in calculating this position, as I found after
reading that there are four mathematical models as following:-
2.1 Sorption-induced strain and permeability models
2..1.1 Sorption-induced strain models.
In order to quantitatively model the sorption-induced volumetric

strain, some empirical approaches have been developed to establish a link
between the measured strain and reservoir pressure.
5
Model the volumetric changes in coal matrix during gas de/adsorption
using elastic properties, sorption parameters and physical properties of
coal. The proposed model is based on the principles of physics and chem-
istry of a surface and the interface theory. The volumetric strain of coal
matrix for sorbing gas includes two components, sorption-induced and
mechanical-induced strain. The sorption-induced strain is directly pro-
portional to the decrease in surface energy and mechanical-induced strain
is calculated by the Hooke's law. In this model, these two strainswere
assumed to be purely additive. It highlights the relationships be-tween
sorption and coal matrix strain. The volumetric strain is calculated as
follows:
where, PL and VL are Langmuir constants, ρs is the density of the solid adsorbent
(coal), EA is the modulus of the solid expansion, E is the Young's modulus, ν is the
Poisson's ratio, Vo is the gas molar volume (22.4 m3/kmol), R is the universal gas
constant and T is the absolute temperatur.
2.1.2 Coal permeability modeling with sorption-induced strain.

The volumetric strain is calculated as follows:
Instead of laying heavy emphasis on the pore volume/cleat com-

pressibility, which is still a topic of controversy for CBM reservoirs,the
model focuses on changes in the grain volume and then converts this to
change in cleat volume, assuming that the total volume re- mains
constant. Moreover, the authors pointed out that this model is more
“friendly” due to its transparency, validation with field data using
reasonable input parameters, simplicity and ability to measure every
parameter required in the model.
6
Siriwardane et al. (2012) carried out a numerical simulation to model

CBM production and CO2 injection by using PSU-COALCOMP
simulator. The shrinkage and swelling strain was incorporated into the
deformation behavior of porous medium (coal) for pressure depletion.
The linear relationship between the shrinkage/swelling strain and
adsorption volume was applied in their investigation, which maybe not
true for coal since the fracture geometry plays a role on the coal de-
formation (Wang et al., 2011). Similar to P&M model, the permeability
variation was calculated by cubic law as shown in Eq. (10). Localized
coal matrix swelling was found around the injection well, inducing a
rapid injection permeability drop. In addition, tracer data indicated a sig-
nificant level of reservoir anisotropy and heterogenei
2.1.3. Brief discussion of other permeability models
quantitative comparison of the S&D and C&B models is presented below.

According to Cui and Bustin (2005), the two models can be rearranged as
follows:
There are two competing terms governing the permeability variation
during depletion. The first term is the increased effective stress with
reduction in gas pressure, denoted as the stress term. The second term is
the matrix shrinkage term due to gas desorption, denoted as the strain
term. For the stress term, the coefficients are ν 1−ν and 1þν 3 1ð Þ −ν for
S&D and C&B, respectively. For coal, the Poisson's ratio is always less
than 0.5. Thus, the term 1þν 3 1ð Þ −ν is larger than ν 1−ν. Obviously,
7
the permeability predicted by the C&B model decreases more than that
by S&D model for the same pressure drawdown. For the strain term,
coefficient of C&B model is smaller than that in the S&D model. Hence,
the permeability increase predicted by the C&B model is lower compared
to S&D model for the same pressure drawdown. For these two reasons,
the permeability predicted by C&B model always lies below the S&D
model, as illustrated in Fig. 1. It is, therefore believed by the authors that
these two models provide the upper and lower limiting responses of th
real CBM permeability changes as a function of reservoir pressure.
The actual permeability behavior should lie withes these two
limits.
2.2Coupling Liu and Harpalani sorption-induced strain and

permeability models
2.2.1. Sorption-induced strain
8
In order to couple this model with different permeability

models, the stress-dependent- strain component must be dropped
for two reasons. First, the mechanical compression of coal is
not analogous to thermal shrinkage since the two mechanisms
are entirely different. Second, the permeability models already
take this component into account and, by including this, it would
effectively be counted twice. Therefore, according to Eq.the
sorption-induced strain alone for different pressures can be
given as:
2.2.2 Coupling of Liu & Harpalani and P&M models
Clarkson et al. (2010) gave a general form of the P&M model
equa-tion as:
2.2.3. Coupling of Liu & Harpalani and S&D models

S&D model can be rewritten as the fol-lowing equation using
incremental matrix shrinkage strain (Δε)
2.2.4. Coupling of Liu & Harpalani and C&B models.
9
2.2.5. Coupling of Liu & Harpalani and Ma et al. models
2.3. Validation of coupled models

2.3.1. Field data validation — Case 1
In the present work, Liu and Harpalani model, combined with the P&M
permeability model, was used first to match the field data for one well in
the San Juan basin. The required input parameters for Liu and Harpalani
(in press) model are consistent with the values measured/estimated by the
modelers, as listed in Table 1. For P&M model, the required input
parameters are presented in Table 2 for the San Juan basin coal. The
Young's modulus (E) and Poisson's ratio (ν) are fixed at 2200 Mpa
(330,000 psi) and 0.3, based on the values recommended by Palmer et al.
(2007).
10
Next, the combined Liu & Harpalani model and Ma et al. model was used
to match the same set of field data.There is excellent agreement between
the field data and modeled results. The required input parameters for this
combined model are listed in Table 3. The Young's modulus, Poisson's
ratio and adsorption data are the same as those used for the P&M model.
The only parameter that is different is the initial cleat porosity, a value of
0.5%, which is also reasonable. The advantage of this model is that the
controversial difficult parameters, ‘f’, ‘g’ and Cf are not required to get a
good match. In other word, in this combined model, every parameter can
be either estimated or measured easily and has a physical meaning that is
easy to relate to. Moreover, Liu & Harpalani and Ma et al. combined
model is more transparent and easy to understand.
11
3. conclusion
Coal formation permeability is complex in that it is influenced by the
stress and is affected by coal shrinkage with gas desorption and swelling
with adsorption. Understanding coal permeability is critical in order to
reliably predict gas production or consider other reservoir gas migration
issues. Since the sorption-induced strain plays an important role in the
variation of permeability, the strain theoretical model should be
incorporated into the permeability prediction models.
An important aspect of this paper is that theoretical sorption-induced

strain model (Liu and Harpalani) is coupled with different analytical
permeability models and the coupled models are validated by matching
two different sets of observed field data. This coupling process signifi-
cantly reduces the experimental work necessary to estimate the volu-
metric strain due to sorption. Based on the work discussed in this paper, a
few important conclusions can be made. These are summarized below:
(1) Liu and Harpalani (in press) sorption-induced strain model has been
successfully and easily incorporated into the different permeability
models.
(2) P&M and Ma et al. models, coupled with Liu and Harpalani
model, are effective methods to predict permeability increase in the San

Juan basin CBM wells operated using primary recovery. Moreover, all
input parameters are within reasonable ranges.
(3) S&D and C&B models, combined with Liu model, can be validated
by varying the cleat compressibility for a set of field data reported by Shi
and Durucan (2005).
(4) The coupled model of Liu and Harpalani and Ma et al. model is valid
for both field cases with fewer input parameters and rea-sonable initial
porosities.
(5) The most important implication is the combining of the sorption-

induced strain and different permeability models. This can be-come a
powerful tool to predict the permeability in the early stage of any new
CBM plays. It only requires the basic adsorption data and mechanical
properties, which are either available or relatively easy to estimate from
laboratory during the early stages of CBM development.
12
4.REFERENCES
Department of Energy and Mineral Engineering, Pennsylvania State University,
University Park, PA 16802, USA
Department of Mining and Mineral Resources Engineering, Southern Illinois

University, Carbondale, IL 62901, USA
Palmer, I., 2010. Coalbed methane completions: a world view. International Journal
of Coal Geology 82, 184–195
Gray, I., 1987. Reservoir engineering in coal seams: part 1 — The physical process of
gas storage and movement in coal seams. SPE Reservoir Engineering 28–34.
Harrison, G.C., Gordon, R.B., 1984. Negative decline curves of coalbed

degasification.
wells. SPE Unconventional Gas Recovery Symposium, 13–15 May, 1984, Pittsburgh,
Pennsylvania, USA.
Moffat, D.H., Weale, K.E., 1955. Sorption by coal of methane at high-pressures. Fuel
34, 449–462.
Levine, J.R., 1996. Model study of the influence of matrix shrinkage on absolute
perme-ability of coal bed reservoirs. In: Gayer, R., Harris, I. (Eds.), Coalbed Methane
and
Coal Geology. Geol Soc Special Pub., London, pp. 197–212.
Wang, S., Elsworth, D., Liu, J., 2011. Permeability evolution in fractured coal: the
roles of fracture geometry and water-content. International Journal of Coal Geology
87,
13–25. Liu, S., Harpalani, S., 2013. A new theoretical approach to model sorption
induced coal shrinkage/swelling. AAPG Bulletin.
http://dx.doi.org/10.1306/12181212061 (in press; preliminary version published

online Ahead of Print 21 January 2013). Shi, J.Q., Durucan, S., 2005. A model for
changes in coalbed permeability during primary and enhanced methane recovery. SPE
Reservoir Engineering 8, 291–299.
13

Rana Rasheed Majeed

Uploaded by

Copyright:

Available Formats

Rana Rasheed Majeed

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rana Rasheed Majeed

Uploaded by

Copyright:

Available Formats

RESERVOIR ENGINEERING

Permeability prediction of coalbed methane reservoirs during

A Report Submitted to the Petroleum Engineering Department of the University of Kerbala

By:- rana Rasheed majeed

"THird Stage "

DR. Tahseen Al-Taei.

Permeability increase in coalbed methane (CBM) reservoirs during

Coalbed methane (CBM) refers to gas formed as a part of the geological

(1) very low effective porosity/permeability, which often hinders the

(2) adsorption storage mechanism since most of the methane produced in

(3) desorption-induced matrix shrinkage since this results in opening up

This paper is aimed at incorporating the recently developed matrix

to 1.6% volumetric increase when coal blocks were subjected to methane

However, no theoretical explanation to unlock the relationship be- tween

2.1 Sorption-induced strain and permeability models

2..1.1 Sorption-induced strain models.

In order to quantitatively model the sorption-induced volumetric

2.1.2 Coal permeability modeling with sorption-induced strain.

Instead of laying heavy emphasis on the pore volume/cleat com-

Siriwardane et al. (2012) carried out a numerical simulation to model

2.1.3. Brief discussion of other permeability models

quantitative comparison of the S&D and C&B models is presented below.

2.2Coupling Liu and Harpalani sorption-induced strain and

In order to couple this model with different permeability

2.2.3. Coupling of Liu & Harpalani and S&D models

2.2.4. Coupling of Liu & Harpalani and C&B models.

2.2.5. Coupling of Liu & Harpalani and Ma et al. models

2.3. Validation of coupled models

An important aspect of this paper is that theoretical sorption-induced

model, are effective methods to predict permeability increase in the San

(5) The most important implication is the combining of the sorption-

Department of Mining and Mineral Resources Engineering, Southern Illinois

Harrison, G.C., Gordon, R.B., 1984. Negative decline curves of coalbed

Coal Geology. Geol Soc Special Pub., London, pp. 197–212.

http://dx.doi.org/10.1306/12181212061 (in press; preliminary version published

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.