Está en la página 1de 12

Predicting Permeability from Porosity Using Artificial

Neural Networks1

S. J. Rogers,2 H. C. Chen,3 D. C. Kopaska-Merkel,4 and J. H. Fang2

ABSTRACT using minimal data, but other kinds of information


(e.g., log- or core-derived lithologic information)
Permeability values in a borehole are predicted are easily incorporated if available. In addition,
by an artificial neural network from the porosity compartmentalization of carbonate reservoirs may
values at the same depths. The network used in this be recognizable by this approach.
study employs an architecture called backpropaga-
tion that is good at making predictions. The tradi-
tional approach for permeability prediction is INTRODUCTION
regression analysis. In regression analysis, the rela-
tionship between porosity and permeability is Permeability is a cr itical variable for both
assumed to be known. In reality, the functional petroleum geology and petroleum engineering.
form of this relationship, i.e., the model equation, However, measuring permeability, directly or indi-
is unknown. In contrast, the neural-network rectly, is a difficult and expensive undertaking.
approach assumes no functional relationship. Over the years, numerous methods and approaches
Six wells from Big Escambia Creek (Jurassic have been proposed to predict permeability values
Smackover carbonate) field in southern Alabama using less expensive and easily obtainable data.
were used to test predicting permeability from Regression analysis is perhaps the most commonly
porosity using a neural network. Porosity and spa- and widely used technique (Wendt et al., 1986;
tial data alone were used to predict permeability Block, 1991; among others). Nelson (1994) recent-
because these data are readily available from any ly reviewed porosity-permeability relationships in
hydrocarbon field. Three scenarios were per- sedimentary rocks, but no mention was made of
formed; in each one, a subset of the six wells was the neural-network approach, although artificial
used for a training set, one well for calibration, and neural networks were used in permeability predic-
one or two wells were used for prediction. For tion by Weiner et al. (1991) and Osborne (1992).
each scenario, simple linear regression was also Predicting permeability and other variables is now
used to predict permeability from porosity. The an area of active research.
neural net predicted permeability much better than Regression analysis is popular because it is sim-
did regression in one scenario; in the other two ple to use and commercial statistical software pack-
scenarios the two methods performed equally well. ages include regression modules. However, regres-
The neural net predicted permeability accurately sion analysis predicts mean values, and thus
overestimates lower values and underestimates
higher values. In contrast, the ANN (artificial neural
©Copyright 1995. The American Association of Petroleum Geologists. All network) approach does not force predicted values
rights reserved.
1Manuscript received August 15, 1994; revised manuscript received May to lie near the mean values, thus preserving the
9, 1995; final acceptance July 28, 1995. actual variability. Also, ANN uses depth information
2Department of Geology, University of Alabama, Tuscaloosa, Alabama
by employing input windows encompassing multi-
35487.
3Department of Computer Science, University of Alabama, Tuscaloosa, ple input sampling points. This is impor tant
Alabama 35487. because regression analysis is not able to use depth
4 Geological Survey of Alabama, P.O. Box O, Tuscaloosa, Alabama
35486.
information in this way.
We thank J. M. Weiner, John H. Doveton, P. H. Nelson, and Kevin T. In view of the inadequacy of traditional methods
Biddle for reviewing an earlier version of this paper. The result was materially of predicting permeability and growing interest in
improved by their comments. Texaco owns a patent on predicting
permeability from wireline logs using neural networks (J. M. Weiner, 1994,
ANN as a predictive tool, we tested one ANN’s abil-
personal communication). ity to predict permeability from porosity. We chose
Porosity and permeability data of 31 wells are available on diskette in an architecture known as backpropagation because
DOS or Macintosh format (Excel saved in .txt format). Cost is $4.00 for North
America and $5.00 outside North America. Contact AAPG Publications it is well suited to prediction problems (discussed
Manager, P.O. Box 979, Tulsa, Oklahoma 74101-0979. more fully in a following section). Also, we used

1786 AAPG Bulletin, V. 79, No. 12 (December 1995), P. 1786–1797.


Rogers et al. 1787

Little Escambia Creek and Chunchula fields, have


similar reservoir characteristics and are found in
the same or in a nearby trend. Also, sufficient core-
derived porosity and permeability data are available
APPR
from BEC field to evaluate results of porosity-to-per-
ORMATION
O F SMACKOVER F
XI
M
AT
E U P D IP
LIM
IT O meability prediction. Details of the petrophysical
characteristics of the Smackover are given in sever-
N
eastern
400
al recent papers (Kopaska-Merkel, 1993; Kopaska-
20 100
Miss. interior
0 Merkel and Hall, 1993; Kopaska-Merkel et al.,
salt basin
500 300
1994).
400
300 Manila embayment
200

200 GEOLOGIC SETTING


400 100
The Upper Jurassic Smackover Formation forms
300
an arcuate belt in the subsurface from southern
200
300
Big Escambia Creek Field
Texas to western Florida. Big Escambia Creek field
(Figures 1, 2) is located in the Conecuh embay-
y m ent ment, which also contains the giant Jay field. The
a
emb
o n ecuh diagenesis of the Smackover in this area was domi-
C
0
100
nated by the effects of several episodes of dolomiti-
200
zation and calcium-carbonate dissolution. Dolomite
300 is responsible for formation or preservation of
400
500 many permeable Smackover pore systems. Strata
assigned to the Smackover Formation in the Big
0 10 20 30 40 50 MILES Escambia Creek field area were deposited in a low-
to medium-energy protected shelf setting.
EXPLANATION 0 10 20 30 40 50 KILOMETERS
The formation in this area is characterized by
THICK SMACKOVER
microbial laminite, oncoid-pellet packstone, and
100 THICKNESS OF SMACKOVER FM coated fossil packstone in the lower Smackover;
IN FEET
fossiliferous mudstone and laminated mudstone in
C.I. 100 FEET
N
the middle Smackover; and pellet wackestone/
mudstone and oolitic grainstone in the upper
Smackover, which contains the reservoir in most
Smackover fields (Bradford, 1984) (Figure 3).
Upper Smackover strata in Big Escambia Creek field
Figure 1—Location map for Big Escambia Creek field. are arranged in shoaling-upward cycles as is charac-
Inset shows location of southern Alabama within the
United States.
teristic of the Smackover throughout southwestern
Alabama. The trap at Big Escambia Creek field is
formed from an element of the regional peripheral
only porosity and spatial information to predict per- fault system, modified by diagenetic permeability
meability because our goal was to develop a tool that barriers created by the interaction of depositional
could be used to predict permeability even where facies distribution, calcium-carbonate dissolution,
information is scanty. The purpose of this paper is to and dolomitization (Bradford, 1984; Vinet, 1984).
describe and evaluate the results of this test. In terms of original rock fabric, reservoir rock in
The test data come from the Smackover Big Escambia Creek field is dominated by dolomi-
Formation in Big Escambia Creek (BEC) field, tized oolitic grainstone and spatially associated
Escambia County, Alabama. Porosity and permeabil- dolomitized lime mudstone (Bradford, 1984). In
ity data from six cores of the Smackover Formation terms of existing rock fabric, reservoir strata in Big
in this field were used. BEC field was chosen for Escambia Creek field are dominated by nonselec-
the case study for three reasons. This field is a can- tive idiotopic dolostone, which is an important rea-
didate for future advanced recover y projects son why we chose this field as a test case for ANN
because reserves are significant (cumulative gas- permeability prediction.
condensate production exceeds 43 million bbl and Petrophysical characteristics of nonselective
cumulative gas production exceeds 630 Gcf*). dolostone are decoupled from depositional facies,
Several other large Smackover fields, including there is substantial megascopic reservoir hetero-
geneity, and porosity is the strongest known corre-
*Gcf = billion cubic feet. late of permeability (Kopaska-Merkel et al., 1994).
1788 Predicting Permeability

R5E R6E
Type Log
25 30

R6E R7E R7E R8E


L/K/G

T T
-15

-15 36 35 36
,3 5

,300 35
-15

2 33 2
0
,37

N N
8

T L/K/G -15,339 -14,85 T


-14 0
1 2 1 6 5 4 -14,9,9
50
00
1
-15,00 1
N SWD -15,050
0 N
-15
,
-15 100
,15
-15 0
12 7 ,20 12 7
EXPLANATION 8 0 10
-15,2
50
-15,300
C.I. 50 feet; contours in feet L/K/G -15,313
below sea level -15,3 15
50 14 13 18
normal fault
-15
L/K/G lowest known gas ,25
0 20
19
N
0 1 2 MILES
dry hole
SCALE
gas-condensate well 30

study area

Figure 2—Structure map on top of Smackover Formation–Big Escambia Creek field, Escambia County, southern
Alabama. Note the type well near the western limit of the field.

The percentage of dolomite varies from 100% to patterns, retrieve data associatively, filter noise
locally less than 10%. In some places where the from experimental data, complete missing informa-
process of dolomitization did not go to completion, tion, and estimate sampled functions (when we do
it was strongly fabric selective. Nondolomitized not know the mathematical form of the functions).
strata in Big Escambia Creek field are impermeable. In short, ANNs can be programmed to mimic
human characteristics, such as learning, generaliza-
tion, and interpretation, where traditional comput-
NEURAL-NETWORK APPROACH ing algorithms and statistical methods have been
inadequate (Hertz et al., 1991).
A brief introduction to neural networks appeared
in Rogers et al. (1992), in which an ANN was used
to determine rock types from well logs. In the fol- Backpropagation Neural Networks
lowing sections, we give additional information rel-
evant to this paper. One of the more popular architectures of ANNs
is the backpropagation neural network (BPNN).
BPNNs have proven to be successful in a variety of
Neural Networks applications and are ver y f lexible and easy to
implement. We used the BPNN because of these
The human brain is composed of approximate- attributes and because of its suitability to predic-
ly 10 billion interconnected cells, called neurons, tion problems. BPNNs are suitable for prediction
whose interaction is responsible for the character- problems where the input-output functions need
istics attributed to intelligence. The brain is a bio- to be learned from experience. Thus, they are
logical neural network. ANNs are created within a model-free estimators.
serial or parallel computer in an attempt to simu- BPNNs are good examples of supervised learn-
late the interactions among neurons in biological ing paradigms. In supervised learning, the network
neural networks. Like brains, ANNs consist of learns from a training set consisting of input-output
numerous “neurons” or processing units and pairs of vectors. An input vector is applied to the
interconnecting synapses that we can program for network, and the output vector is used as a “teach-
computation. We can program ANNs to recognize er” to show the correct (or desired) output.
Rogers et al. 1789

through the input layer and values propagate for-


ward through the network’s hidden layers (one hid-
den layer was used in this study) until they reach
the output layer. Each connection between nodes
of the network has a weight value, which dictates
how the values propagate through the network.
The network learns by adjusting these weight val-
ues so that it responds correctly to inputs for the
problem the network is to learn. The process of
adjusting the weight values is known as training
and is performed by taking the error produced by
the network for a given input and propagating this
error backward through the layers to adjust the
weights; this process gives the backpropagation
neural network its name. Before training begins,
the weights are initialized to random values. The
training process repeatedly presents the training
set input patterns to the network, compares the
network’s results to the desired outputs, and
updates the weight values to produce better out-
puts. Training can be a lengthy process, and there
is no guarantee that the network will ever reach a
state in which all inputs produce the desired out-
puts. However, the network will usually produce
outputs within some error tolerance (see Rogers et
al., 1992, for more details).
BPNNs are good at generalization. After a BPNN
is adequately trained, it learns to identify the signif-
icant features and to ignore irrelevant data in the
input. Thus, it can produce the correct output
despite differences between training data and test
data (Wasserman, 1993).
We do not discuss some topics related to BPNNs;
these topics include normalization of input, scaling
of output, the momentum problem, and the possi-
bility of the solution converging to a local mini-
mum. These topics are beyond the scope of this
paper. The interested reader should refer to
Wasserman (1989), Hecht-Nielsen (1990), or Zurada
(1992) for further information on these topics.

Neural Networks and Prediction

BPNNs are very good at prediction because they


can learn a problem from a relatively small set of
Figure 3—Type log for Big Escambia Creek field, Smith examples and their generalizing ability allows them
25-7 No. 1 well, Alabama Oil and Gas Board Permit No.
4727. See Figure 2 for well location. Numbers on the left
to make predictions on input data that were never
of the lithology column indicate reservoir lithofacies used in their training set. BPNNs can also deal with
defined on the basis of mineralogy, rock fabric, and incomplete or noisy data better than traditional
pore types. prediction methods.
To predict permeability from porosity, a training
set is compiled using porosity values as inputs and
Learning is accomplished by adjusting the network their corresponding permeability values (known
weights until the differences between the actual from core analyses) as desired outputs. The cre-
and desired output are acceptably low. BPNNs ation of a training set for the permeability predic-
(Figure 4) are feed-forward networks because tion problem described by this paper is more com-
inputs from a training set are fed into the network plex than simply pairing a porosity value from a
1790 Predicting Permeability

Figure 4—Architecture of
neural network used in this
paper.

given depth with its corresponding permeability lookup table could be created from a large number
value. If a one-to-one correspondence existed of examples. Because this is not the case, a more
between porosity and permeability, a simple innovative approach is required. Instead of using
just one porosity value, a set of adjacent porosity
values are used to predict a single permeability
value. For this work, 25 porosity values, with 12
porosity values above and below the depth to be
predicted, as well as the porosity value at the
depth were used to predict the permeability
(Figure 5). Spatial information (geographic coordi-
nates and depth) was used as part of the input to
enhance the network’s ability to make predictions.
BPNNs can learn the significance of each input to
the predicted value so that inputs that do not influ-
ence the predicted value will not interfere with
the network’s performance. This is accomplished
by adjusting the weights of connections between
pairs of nodes during the training process, as
explained in the previous section.
The training sets described in this paper were
composed of training set porosity-permeability pairs
in the form illustrated in Figure 5. Figure 4 shows the
inputs and outputs as the BPNN sees them.

CASE STUDIES
Figure 5—Schematic diagram showing how spatial
information, porosity, and permeability data were used Six cored wells from Big Escambia Creek field in
in training the network. southern Alabama provided observed porosity and
Rogers et al. 1791

Figure 6—Porosity-
permeability data for all
wells used in this study.

permeability values in 1-ft intervals (Figure 6). These as if it were a test well. At some point, the mean
wells were drilled in a part of the field that is litho- square error, which had been falling, began to rise.
logically homogeneous laterally (cf. Bradford, 1984; The network weights that provided the best results
Figure 7). Using the cored data in various combina- (lowest mean square error) on the calibration well
tions, three scenarios were run to test the BPNN’s were used for the actual prediction of the test wells.
predictive capability. These scenarios were based on Minor reversals in the trend of the mean square error
geological considerations. In each scenario, a subset were ignored to avoid choosing a local minimum as
of the six wells was used for a training set (either the solution. The calibration well was never includ-
two, three, or four wells), one well was used to cali- ed in the training sets or the testing sets so that a
brate the network, and the remaining wells were true test of the network’s performance could be
predicted. The training set used all of the observed measured. Note that the test wells are never used in
porosity and permeability values available, the X and the training set or as a calibration well because this
Y coordinates of the well’s position in the field, and would influence the results of the network training
the depth (Z) of the observed permeability value. and might cause the network to memorize rather
The BPNN was trained for several thousand itera- than generalize. Both permeability (from cores) and
tions of the training set as follows. During the itera- porosity (either from cores or logs) are necessary in
tive training process, the current network weights the training process, but only porosity data are
periodically were checked using the calibration well needed to make predictions.
1792 Predicting Permeability

Figure 7—Comparison of depth vs. porosity and permeability in wells 1704 and 1705, indicating a strong similarity
between the Smackover sections in these two wells. Dashed line indicates correlation of strata between these two
wells.

In the first scenario, two wells (permits 1877 the network in this scenario was degraded by one
and 1928) were used for the training, and perme- or more of at least two possible factors. The area
ability values were predicted in well 1930 using covered by the wells used in this scenario was
the porosity values at the same depths. All three much greater than that for scenario one. Lateral
wells lie approximately along strike from one reservoir heterogeneity over these greater dis-
another, and this arrangement minimizes lateral tances may have adversely influenced the network.
lithologic variation. Well 1802 was the calibration In addition, a boundary between two reservoir
well. Figure 8 is the location map of the wells and compartments may lie within the larger prediction
Figure 9 gives the results of ANN prediction. The area of scenario two. As mentioned in the section
predicted values from a linear regression analysis on geologic setting, the Smackover Formation in
are also shown. ANN yielded a much more accu- Big Escambia Creek field is characterized by rela-
rate prediction than the regression analysis (Figure 9; tively high values of megascopic reservoir hetero-
Table 1). geneity, which could be expressed in part as geo-
In the second scenario, three wells were used graphically distinct reservoir compartments.
for training (wells 1802, 1877, and 1930) and the The last scenario used four training wells (1802,
calibration well was 1928. In this scenario, we pre- 1877, 1928, and 1930), also used in the previous
dicted permeability in two different wells (1704 two cases, and predicted permeability in one well
and 1705), which lie along strike with and between (1704; permeability also was predicted in this well
the training wells. We expected this scenario to in scenario two). Well 1705 was used for calibra-
perform at least as well as scenario one because the tion. Figure 8 shows the wells used and Figure 12
distribution of training wells was designed to cap- gives the results, indicating that the network per-
ture information about petrophysical variation in formed no better than the regression analysis. The
the region surrounding the predicted wells. The same factors affecting the results of scenario two
results are shown in Figures 10 and 11, respective- influenced this scenario because they differed only
ly. The BPNN predicted permeability slightly more by the addition of a fourth training well, which had
accurately than the regression model, but the dif- small effect on the distribution of data used to train
ference is not significant (Table 1). Performance of the network.
Rogers et al. 1793

Figure 8—Location of wells


mentioned in this paper.

Porosity and permeability data for Big Escambia “window.” However, in practical terms this short-
Creek field are available on diskette from AAPG. coming may not be severe for two reasons. First, a
1-ft-thick high-permeability streak may not consti-
tute a reservoir interval. If the high-permeability
DISCUSSION streak is several feet thick, then the method does
a much better job of prediction (e.g., thicker per-
All of the case studies have a few features in meable zones in Figures 9 and 10). Second,
common that illustrate both the strengths and although it is true that thin high-permeability
weaknesses of the neural-network approach. The zones strongly inf luence the success of water-
neural network yielded more accurate predictions flood operations, the neural network does predict
than did linear regression (Figure 9; Table 1). Both the stratigraphic location and thickness of such
methods, however, are unsuccessful at predicting zones. Intervals characterized by locally high- and
very high permeability values that are surrounded low-permeability values in the “real” data exhibit
by substantially lower permeability values. The the same relative relationships to their neighbors
neural network does a poor job in these situations in the predicted data (Figures 10–12). Another
because of the averaging effect of the 25-point implication of this result is the observation that an

Training: 1877,1928 (1802) Figure 9—Comparison of


Prediction: 1930 predicted permeability
35 values with those estimated
by regression method and
Observed those measured from the
30 Neural Network cores of well 1930 for
Regression
scenario one.
25
Permeability (md)

20

15

10

0
15453 –



15456 –



15459 –



15462 –



15465 –



15468 –




15471 –



15474 –



15477 –



15480 –



15483 –



15486 –



15489 –



15492 –



15495 –



15498 –



15501 –



15504 –



15507 –



15510 –




15513 –



15517 –



15520 –



15523 –



15526 –

Depth (ft)
1794 Predicting Permeability

Table 1. Mean Squared Error (in md) for BPNN and for Lithologic information was not used to predict
Linear Regression for the Three Scenarios* permeability for four reasons. First, and most
important, this project was specifically designed to
Scenario test a network that employed only spatial and
2 2 porosity information to predict permeability
Method 1 (Well 1704) (Well 1705) 3 because our goal was to develop a tool that could
BPNN 20.540 0.05400 0.08316 0.123035 be applied to any field, no matter how scanty the
Regression 101.978 0.07185 0.10127 0.110998 available data. (However, the architecture of the
network used makes it very simple to add more
Regression/
BPNN 4.964849 1.330555 1.217773 0.902166
kinds of input data if they are available.) Second,
lithologic units in the Smackover Formation, identi-
*BPNN = Backpropagation neural network. fied on the basis of mineralogy, rock fabric, pore
types, and sedimentary structures, do differ consis-
tently in porosity and per meability values
ANN is not used straight off the shelf. An ANN (Kopaska-Merkel et al., 1994). However, individual
ought to be designed for the problem at hand. In a lithofacies tend to exhibit a wide range in perme-
permeability-prediction study for evaluating a ability values, coupled with low correlation coeffi-
proposed or actual waterf lood operation, for cients for porosity and permeability within a single
instance, a narrow window should be used, or lithofacies. Thus, lithologic information is a crude
other characteristics incorporated into the model, predictor of permeability. Third, the effect of X–Y
such as recognition of the shape of the permeabil- variation in rock type on porosity and permeability
ity distribution vs. depth where permeability was minimized in this study by using a relatively lat-
changes dramatically over a short stratigraphic erally homogeneous part of the field (Bradford,
interval. This could enhance prediction of thin 1984). Fourth, porous and permeable strata in BEC
very high permeability zones. field are dominated by intercrystalline and mixed

Training: 1802,1877,1930 (1928) Figure 10—Comparison


Prediction: 1704 of predicted permeability
4 values with those
measured from the cores of
well 1704 for scenario two.
3.5
Observed
Neural Network
Regression
3

2.5
Permeability (md)

1.5

0.5

0
15248 –



15252 –



15256 –



15260 –



15277 –



15281 –



15285 –



15289 –



15293 –



15297 –


15301 –



15305 –



15309 –



15313 –



15317 –



15321 –



15325 –



15329 –



15333 –



15337 –



15341 –



15345 –



15349 –



15353 –



15357 –



Depth (ft)
Rogers et al. 1795

Training: 1802,1877,1930 (1928) Figure 11—Comparison


Prediction: 1705 of predicted permeability
4 values with those
measured from the cores
Observed of well 1705 for scenario
Neural Network
3.5 two.
Regression

2.5
Permeability (md)

1.5

0.5

0
15136 –


15140 –



15144 –


15148 –



15152 –



15156 –


15160 –



15164 –


15168 –



15172 –


15176 –



15180 –



15184 –


15188 –



15192 –


15196 –



15200 –



15204 –


15210 –



15214 –


15218 –



15222 –


15226 –



15230 –



15234 –


15238 –



15242 –


15246 –



Depth (ft)

pore systems (Kopaska-Merkel and Mann, 1991). of crystalline dolostone, which exhibit subtle differ-
This kind of pore system is controlled by deposi- ences in cr ystal size, shape, and distribution.
tional fabric only in the grossest sense (e.g., lime Hence, petrophysical properties, especially perme-
mudstone is commonly not dolomitized and there- ability, are relatively independent of rock type. Also,
fore is neither porous nor permeable). It is com- intercrystalline reservoirs exhibit consistently small
monly impractical to discriminate among subtypes aspect ratios (pore to throat size ratio), consistent

Training: 1802,1877,1928,1930 (1705) Figure 12—Comparison


Prediction: 1704 of predicted permeability
4 values with those
measured from the cores
3.5 Observed of well 1704 for scenario
Neural Network three.
3 Regression

2.5
Permeability (md)

1.5

0.5

0
15246 –



15251 –



15256 –



15261 –


15279 –



15284 –



15289 –



15294 –


15299 –



15304 –



15309 –



15314 –


15319 –



15324 –



15329 –



15334 –


15339 –



15344 –



15349 –



15354 –



15359 –


15364 –



15369 –



15374 –



15379 –



15384 –


15389 –

Depth (ft)
1796 Predicting Permeability

pore and throat shapes, and are relatively homoge- Addison-Wesley, 433 p.
neous on a microscopic scale (Kopaska-Merkel and Hertz, J., A. Krogh, and R. G. Palmer, 1991, Introduction to the
theory of neural computation: Redwood City, California,
Mann, 1992). Consequently, porosity is the most Addison Wesley, 327 p.
useful predictor of permeability in intercrystalline Kopaska-Merkel, D. C., 1993, Capillary-pressure characteristics of
reservoirs. Smackover reservoirs in Alabama: Geological Survey of
Alabama Circular 170, 38 p.
Kopaska-Merkel, D. C., and D. R. Hall, 1993, Reservoir characteri-
zation of the Smackover Formation in southwest Alabama:
CONCLUSIONS Geological Survey of Alabama Bulletin 153, 111 p.
Kopaska-Merkel, D. C., and S. D. Mann, 1991, Pore facies of
We have demonstrated by way of three different Smackover carbonate reservoirs in southwest Alabama: Gulf
scenarios the utility and superiority of neural net- Coast Association of Geological Societies Transactions, v. 41,
p. 374–382.
works in predicting permeability values from Kopaska-Merkel, D. C., and S. D. Mann, 1992, Regional varia-
porosity values. A properly trained network would tion in microscopic and megascopic reservoir heterogeneity
predict permeability values in noncored wells from in the Smackover Formation, southwest Alabama: Gulf Coast
log-derived porosity values more accurately than Association of Geological Societies Transactions, v. 42,
p. 189–212.
would regression analysis. The ANN approach is Kopaska-Merkel, D. C., S. D. Mann, and J. W. Schmoker, 1994,
model free; that is, no functional form (such as lin- Controls on reservoir development in a shelf carbonate: Upper
ear, polynomial, or multiple) between permeability Jurassic Smackover Formation of Alabama: AAPG Bulletin,
and porosity values need be assumed; however, the v. 78, p. 938–959.
neural-network approach is but one of several pre- Nelson, P. H., 1994, Permeability-porosity relationships in sedi-
mentary rocks: The Log Analyst, v. 35, p. 38–62.
dictive tools available in the geologist’s toolbox. Osborne, D. A., 1992, Neural networks provide more accurate
Also, more field studies should be conducted to fur- reservoir permeability: Oil & Gas Journal, (Sept. 28),
ther explore the scope and power of this new p. 80–87.
approach. Rogers, S. J., J. H. Fang, C. L. Karr, and D. A. Stanley, 1992,
Determination of lithology from well logs using a neural net-
The network used in this study was specifically work: AAPG Bulletin, v. 76, p. 731–739.
designed to use only spatial and porosity informa- Vinet, M. J., 1984, Geochemistry and origin of Smackover and
tion to predict permeability. These kinds of informa- Buckner dolomites (Upper Jurassic), Jay Field area,
tion are available for nearly any hydrocarbon field Alabama–Florida, in W. P. S. Ventress, D. G. Bebout, B. F.
and are inexpensive to collect; however, the neural- Perkins, and C. H. Moore, eds., The Jurassic of the Gulf Rim:
Proceedings of the Third Annual Research Conference, Gulf
network architecture readily permits use of other Coast Section, SEPM, p. 365–374.
kinds of information to improve the performance of Wasserman, P. D., 1989, Neural computing: theory and practice:
the network, if such information is available. New York, Van Nostrand Reinhold, 230 p.
Wasserman, P. D., 1993, Advanced methods in neural computing:
New York, Van Nostrand Reinhold, 255 p.
REFERENCES CITED Weiner, J. M., J. A. Rogers, J. R. Rogers, and R. F. Moll, 1991,
Predicting carbonate permeabilities from wireline logs using a
Block, S., 1991, Empirical prediction of porosity and permeability back-propagation neural network, CM1.1: Society of
in sandstones: AAPG Bulletin, v. 75, p. 1145–1160. Exploration Geophysicists, Expanded Abstracts with
Bradford, C. A., 1984, Transgressive-regressive carbonate facies Biographies, Technical Program, v. 1, p. 285–289.
of the Smackover Formation, Escambia County, Alabama, in Wendt, W. A., S. Sakurai, and P. H. Nelson, 1986, Permeability pre-
W. P. S. Ventress, D. G. Bebout, B. F. Perkins, and C. H. diction from well logs using multiple regression, in L.W. Lake,
Moore, eds., The Jurassic of the Gulf Rim: Proceedings of the ed., Reservoir characterization: Orlando, Florida, Academic
Third Annual Research Conference, Gulf Coast Section, Press, p. 181–221.
SEPM, p. 27–39. Zurada, J. M., 1992, Introduction to artificial neural systems: New
Hecht-Nielsen, R., 1990, Neurocomputing: Reading, Massachusetts, York, West Publishing Co., 683 p.
Rogers et al. 1797

ABOUT THE AUTHORS

S. J. Rogers D. C. Kopaska-Merkel
Samuel J. Rogers is an application engineer for the David Kopaska-Merkel is ground water section head
Intergraph Corporation in Huntsville, Alabama. He at the Geological Survey of Alabama. He earned a Ph.D.
received his M.A. degree in computer science from the in geology from the University of Kansas in 1983 and has
University of Alabama. He is currently working on a book worked as a petroleum geolgist, carbonate sedimentolo-
adressing object-oriented neural-network architectures. gist, and hydrogeologist for Shell Oil Company, the
Northeastern Science Foundation, and the Geological
H. C. Chen Survey of Alabama.
Hiu-Chuan Chen is a professor of computer science at J. H. Fang
the University of Alabama. She received a Ph.D. from the
State University of New York at Buffalo in 1972, and was Jen-Ho Fang is a professor of geology at the University
a visiting scholar in the Department of Computer of Alabama. He received his B.S. degree from the
Science at Stanford University from 1979 to 1980. Her National Taiwan University, and his Ph.D. from
research interests include algorithm design, AI tech- Pennsylvania State University in 1961. He is a fellow of
niques, expert systems, neural networks, and fuzzy the Mineralogical Society of America and the Geological
logic. She has conducted research for JPL, the Society of America. He was once a mineralogist, and a
Department of Energy, Marine Terminal Computer mineral, “Fangite” (TI3ASS4), was named after him. He
Systems, and several oil companies. served as president of the Mathematical Geologists of the
United States.

También podría gustarte