Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Earth science
Reexamination Certificate
1999-02-19
2001-05-01
McElheny, Jr., Donald E. (Department: 2862)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Earth science
C702S007000
Reexamination Certificate
active
06226595
ABSTRACT:
BACKGROUND OF THE INVENTION
Borehole imaging tools provide different types of borehole images, for example, electrical tools like the Formation Microlmager (FMI) tool, Resistivity At Bit (RAB) and Azimuthal Laterolog (AzLL) deliver electrical images of the borehole wall, and acoustical tools like the Ultrasonic Imaging Tool (USIT) deliver acoustic images of the borehole wall. Electrical and acoustic borehole images show variations on the borehole wall caused, for example, by geological bed boundaries and fractures. One objective of borehole image processing is to geometrically characterize bed boundaries and fractures. When bed boundaries and fractures are inclined or “dipping” at some angle relative to the axis of the borehole, they sweep out a sinusoidal pattern azimuthally around the borehole image. Each bed boundary or fracture is a “dip event” in the borehole image.
The subject matter of the present invention relates to a workstation and software based method and apparatus for analyzing input data representing images of an internal wall of a borehole, containing a plurality of “dip events” and producing a first plurality of output data comprising a first plurality of “tracks”, each of which is a set of connected track points lying along a dip event, the first plurality of tracks representing the respective first plurality of dip events inherent in the input data, and a second plurality of output data comprising a plurality of “dip data” d which are the parameters for a dip event that best fits each first plurality of “tracks” representing the first plurality of “dip events”, the software-based method and apparatus of this invention utilizing a “Multi-Sine Tracking” (MST) algorithm derived by modifying a Multi-Target Tracking (MTT) algorithm.
There is a software application, known as “BorView”, which receives borehole image data from a downhole borehole imaging tool and which allows an operator sitting at a workstation to manually define, by using a mouse in conjunction with the workstation display, a plurality of “dip events” which are inherent in the borehole image data being displayed on the workstation display. Hereafter, a “dip event” is defined as a formation fracture or a bed boundary or any other type of approximately sinusoidally varying feature which appears in an image of the internal wall of a borehole penetrating an earth formation. In fact, the workstation operator previously had to manually define each such “dip event”. However, since there are a multitude of such dip events in the borehole image data, the aforementioned manual operation being performed by the operator is very tedious and can be very time consuming. Consequently, a method and apparatus is needed for automatically defining and generating, as an output, a geometrical characterization and a set of dip data corresponding to all, or nearly all, such dip events in the borehole image data in response to the touch of a key on the workstation keyboard.
In the domain of oilfield data interpretation, there is a problem called “dip estimation” of fitting sine curves to dip events in borehole images. In this specification, an algorithm and associated method known as “Multi-Target Tracking” (MTT) is adapted from outside the oilfield domain. It has the potential to be an extremely efficient means for processing borehole image data in the dip estimation problem mentioned above and disclosed in this specification.
MTT algorithms have been developed, mainly for military applications, over the past 20 years [see the “Bar-Shalom” and “Kurien” references which are cited along with other references in the “reference” section located at the end of this specification]. MTT algorithms combine many different, intermittent sources of information into a self-consistent and complete representation of military and civilian vehicle (e.g. aircraft) identity and location in a space such as a heavily traveled airspace. In military defense applications, MTT algorithms quickly integrate large amounts of data from a diverse set of sensor types (e.g. radar, infrared, imaging sensors, human reports, etc). As described therein, MTT algorithms integrate all available data (prior information, sensor data), with the aid of “models”, to form many “hypotheses” about what and where each aircraft or other vehicle is, and how it relates to each data item. One component of MTT, and in particular, the Track Hypotheses Management algorithm, ranks competing hypotheses and ultimately determines which one best represents the data in a manner consistent with the models.
The novelty of the method and associated apparatus disclosed in this specification in accordance with the present invention lies in the recognition that fundamental mathematical similarities exist between the MTT problem and the oilfield dip estimation problem addressed in this specification. The principle challenge of the novel method in accordance with the present invention is to reflect the constraints and characteristics of the oilfield dip estimation problem into the mathematical framework that underlies the MTT algorithm.
The present-day use of the MTT algorithm for military and civilian vehicle surveillance and tracking is reported in the public literature [refer to the “Bar-Shalom” and “Kurien” references in the reference section of this specification]. Although the approach is general, and applies for many different types of sensors and target vehicle types, consider a representative problem of processing radar measurements to track one or more aircraft. As defined in the Kurien reference, the input data consist of a set of “scans”. At each time, a scan is a set of measurements generated by the sensor (e.g. a radar scan) from a single look over the entire surveillance volume. Typically, the raw scan measurement data are processed by some type of radar detection pre-processor to create a set of “returns” or “reports” at each time. In the Kurien reference, a “report” is defined to be a set of measurements originating from a single source in a single scan. The output of the detection pre-processor is thus a set of reports, that are the input to the next stage of processing called the Multi-Target Tracking (MTT) algorithm. The functional blocks of the MTT algorithm are shown in
FIG. 3.5
of Kurien, reproduced as
FIG. 13
a
of this specification (Kurien assigns the name “Multi-Tracker” to his particular implementation). As shown in
FIG. 13
a
, the MTT algorithm is a global approach to process the input sequence of radar reports from multiple targets, output from the pre-processing Detector to form an output of a set of confirmed tracks, one track per aircraft in the scanned space.
During the MTT processing, a large number of potential “hypotheses” must be considered regarding which of the many input reports is associated with which of the multiple targets being tracked. The algorithmic logic to manage these multiple hypotheses is included in the functional module in
FIG. 13
a
labeled “Track Hypotheses Management”. The “Track Hypotheses Management” module itself consists of a collection of functional modules, which are shown in more detail in
FIG. 3.6
of Kurien, reproduced as
FIG. 13
b
of this specification. Two of the functional modules within the “Track Hypotheses Management” function shown in
FIG. 13
b
are the “Predict Tracks” and “Update Existing Tracks” modules. These two modules have embedded within them a mathematical physical model describing the way aircraft move and accelerate in space, each model being called a “dynamics model”. A representative dynamics model for an aircraft is shown in equation 3.1 of Kurien, rewritten in simpler form here:
X
t
(
k+
1)=&THgr;
X
t
(
k
)+
v
t
(
k
) (A)
where
X
t
=
[
x
t
y
t
x
.
t
y
.
t
]
and
Θ
=
[
1
0
Δ
T
0
0
1
0
Δ
T
0
0
1
0
0
0
0
1
]
In these equations, each target is tracked in a Cartesian frame with the origin located at the sensor position. The target “state” X
t
is represented with four variables constituting the p
Rossi David J.
Willsky Alan S.
Bouchard John H
McElheny Jr. Donald E.
Schlumberger Technology Corporation
LandOfFree
Method and apparatus using multi-target tracking to analyze... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus using multi-target tracking to analyze..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus using multi-target tracking to analyze... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2519695