COMPLEX LITHOLOGY

ANALYSIS
[FUZZY APPROACH]
Presented by

OBOT, ITORO A.
G2011/MENG/PNG/FT/851
Department of Petroleum & Natural Gas Engineering
University of Port Harcourt
COURSE LECTURER: PROF. DULU APPAH
COURSE CODE/TITLE: PNG 613.2/ FORMATION EVALUATION

JULY 2014

OUTLINE
INTRODUCTION
LITHOLOGY ANALYSIS TOOLS

WELL LOG ANALYSIS PROCEDURE

FUZZY METHOD
CASE STUDY
CONCLUSION

INTRODUCTION
Physical Characteristics of a rock
Analysis of drill core and Geophysical logging
Improved in-situ determination of rock characteristics and
properties.

gamma ray logs, neutron logs, density logs, deep

resistivity logs, and shallow resistivity logs were
identified as lithology logs
combination of these different borehole logging techniques would
provide information related to rock characteristics and properties

Fuzzy Method of Well Log Analysis

LITHOLOGY ANALYSIS TOOLS

GAMMA RAY LOGS
Gamma ray logs measure natural radioactivity in formations.
Shale-free sandstones and carbonates give low gamma ray
readings. As shale content increases, the gamma ray log response
also increases.
[units and representation]

LITHOLOGY ANALYSIS TOOLS

THE RESISTIVITY LOG
The resistivity log is a measure of a formations
resistivity. Most rock materials are essentially insulators,
while their enclosed fluids are conductors. When a
formation is porous and contains salty water, the
resistivity will be low

LITHOLOGY ANALYSIS TOOLS

THE RESISTIVITY LOG
Deep resistivity is the resistivity recorded farther
away from the inversion core created by the
drilling mud.
Shallow resistivity log is the resistivity recorded
close to the oil well bore.
Shales show low resistivity values with high
gamma ray values.

LITHOLOGY ANALYSIS TOOLS

THE RESISTIVITY LOG MECHANISM
Compton scattering.
The measurements recorded by the GR logging tool, after appropriate
corrections for the borehole size etc. are proportional to the weight
concentrations of the radioactive material in the formation:

. . . . . . . . . eqn (1)
where,
i = densities of the radioactive minerals,
Vi = bulk volume factors of the minerals
Ai = proportionality factors corresponding to the radioactivity of the mineral,
b = the bulk density of the formation.

LITHOLOGY ANALYSIS TOOLS

THE DENSITY LOG
continuous record of a formations bulk
density.
determination of porosity, hydrocarbon density, oil-shale yield, the
differentiation between liquids and gases (when used in combination
with neutron log), evaluation of shaly sands and complex lithologies,
calculation of overburden pressure and rock mechanical properties.
Variation of density indicates porosity changes.
organic content = low density, low density = high porosity and viceversa.

LITHOLOGY ANALYSIS TOOLS

THE NEUTRON LOG
lithology identification, delineation of porous formations and
determination of porosity, and the differentiation between
liquids and gases when used in combination with density log.
On crossplot of neutron and density logs, pure shale can be
recognized by the high neutron value relative to the density
value which gives a large positive separation to the logs
while gas stands out distinctly giving a large negative
separation.
They respond primarily to the amount of hydrogen in the
formation.

LITHOLOGY ANALYSIS TOOLS

WELL LOG ANALYSIS PROCEDURE:

FUZZY METHOD

A Fuzzy Inference System (FIS) is a system that uses fuzzy set

theory to map inputs (features in the case of fuzzy
classification) to outputs (classes in the case of fuzzy
classification)
Fuzzy rules are a collection of linguistic statements that
describe how the FIS should make a decision regarding
classifying an input or controlling an output.
Fuzzy rules are always written in the following form:
if (input1 is membership function1) and/or (input2 is
membership function2) and/or (?) then (outputn is output
membership functionn).

WELL LOG ANALYSIS PROCEDURE:

FUZZY METHOD

A.
B.
C.
D.
E.
F.

DATA IDENTIFICATION
DATA SANITIZATION
CORRELATION TEST
NORMALIZATION
GENERATING CLUSTERS
FUZZIFICATION

WELL LOG ANALYSIS PROCEDURE:

FUZZY METHOD

A. DATA IDENTIFICATION
Mainly from open-hole wireline sub-surface logging data report.

WELL LOG ANALYSIS PROCEDURE:

FUZZY METHOD
B. DATA SANITIZATION
The raw well logs were inspected and sanitized to eliminate
negative and null values were removed from the log data. This
includes readings gotten when the tool is not moving or unrelated
spike that might have been recorded.

WELL LOG ANALYSIS PROCEDURE:

FUZZY METHOD
C. CORRELATION TEST
The raw well logs are subjected to a statistical correlation test, to
determine the relationships among the data elements for the
purpose of clustering them. The correlation test was carried out on
the log values to determine if there was any relationship between
the log data values.

WELL LOG ANALYSIS PROCEDURE:

FUZZY METHOD

D. NORMALIZATION
The raw well log values were normalized (in the range 0 to 1) for
the purpose of rendering the data dimensionless, and removing the
effect of scaling on the values.

WELL LOG ANALYSIS PROCEDURE:

FUZZY METHOD

E. GENERATING CLUSTERS

Self Organizing Map (SOM) of neural networks for the determination of

oil well lithology and fluid contents is used here. It consist of tools for
building human-like intelligence system.
The normalized log values were subjected to the clustering algorithm of
the SOM neural network for the purpose of generating meaningful clusters
so that the data sample can be assigned into their respective cluster group.
The mean of the log values and their standard deviation are then
computed. The computed mean of the log values are then used to infer the
lithology and fluid content of the rock species that characterize the
geological formation of the oil well being investigated by determining
their fuzzy value

WELL LOG ANALYSIS PROCEDURE:

FUZZY METHOD
F. FUZZIFICATION
The mean and standard deviation values of the log data by clusters
were computed to determine their fuzzy value which can be
considered High, Moderate, or Low.
These are important variables desirable for identifying the
physical property of each cluster and subsequently generating a
chart which shows the distribution of the rocks in the well and
location of hydrocarbon and fluid content.

WELL LOG ANALYSIS PROCEDURE:

FUZZY METHOD
F. FUZZIFICATION

CASE STUDY
Based on a study carried out by Akinyokun in
2002 on geophysical well log data from the
Niger-Delta region of Nigeria. Twelve cluster
groups were identified in the well log data.
The cluster groups (denoted by their cluster
numbers), their mean values and computed
standard deviations are presented in Table

CASE STUDY

CASE STUDY
Identified Clusters.

CASE STUDY

Well Stratigraphy Chart

CONCLUSION

The fuzzy inference methodology adopted in the

interpretation of the clusters was derived from the
methods used in the interpretation of traditional
graphical cross-plots by log analysts
The result not only gives the oil well lithology, it also
gives an indication of the fluid content of the oil well
and rock materials identified.
The SOM based clustering and the fuzzy inference rules
developed in this paper can form the basis for the
development of a neuro-fuzzy expert system that can be
used for the detection of fluid content in oil wells.

THANK
YOU