System, method and computer program product for wellbore event modeling using rimlier data

Patent metadata

Field	Value
Publication number	US-10242130-B2
Application number	US-201214435404-A
Country	US
Kind code	B2
Filing date	Nov 5, 2012
Priority date	Nov 5, 2012
Publication date	Mar 26, 2019
Grant date	Mar 26, 2019

Abstract

A data mining and analysis system which analyzes clusters of outlier data (i.e., rimliers) to detect and/or predict downhole events.

First claim

What is claimed is:

1. A computer-implemented method to model downhole events, the method comprising:
extracting a dataset from a database, the dataset comprising normal wellbore data and outlier wellbore data;
clustering a plurality of the outlier data into a plurality of clusters;
segregating the plurality of clusters into a high density cluster and a low density cluster, wherein the high density clusters are utilized as a rimlier;
analyzing the rimlier to determine data variables within the rimlier that indicate a downhole event, wherein the analyzing comprises:
determining a Head Rimlier Factor as defined by:

E
h

E
r

=

Σ
-

p
h

⁡

(
x
)

⁢

log
⁡

(
x
)

Σ
-

p
r

⁡

(
x
)

⁢

log
⁡

(
x
)

;

 and
determining a Tail Rimlier Factor as defined by:

Et

E
r

=

Σ
-

p
t

⁡

(
x
)

⁢

log
⁡

(
x
)

Σ
-

p
r

⁡

(
x
)

⁢

log
⁡

(
x
)

,

wherein E h is entropy of head data, p h is probability of the head data, E r is entropy of rimlier data, p r is probability of the rimlier data, E t is entropy of tail data, and p t is probability of the tail data,
wherein the Head Rimlier Factor and the Tail Rimlier Factor are utilized to determine the data variables indicating the downhole event; and
modeling the downhole event based upon the analysis of the rimlier,
wherein a wellbore is drilled, completed or stimulated in accordance to the modeled downhole events.

2. A computer-implemented method as defined in claim 1 , further comprising removing corrupted data from the extracted dataset.

3. A computer-implemented method as defined in claim 1 , wherein analyzing the rimlier further comprises:
segregating the rimlier into a normal high density rimlier and an outlier high density rimlier; and
analyzing the outlier high density rimlier to determine the data variables that indicate the downhole event.

4. A computer-implemented method as defined in claim 1 , wherein clustering the plurality of the outlier data further comprises forming a plurality of rimliers.

5. A computer-implemented method as defined in claim 4 , wherein modeling the downhole event further comprises modeling an energy efficiency of a downhole assembly.

6. A computer-implemented method as defined in claim 1 , further comprising determining whether the modeled downhole event can be avoided.

7. A computer-implemented method as defined in claim 1 , further comprising producing an alert signal corresponding to the modeled downhole event.

8. A computer-implemented method as defined in claim 1 , further comprising displaying the modeled downhole event in the form of a tree or earth model.

9. A computer-implemented method as defined in claim 4 , wherein analyzing the plurality of rimliers further comprises determining a pattern across the plurality of rimliers, wherein the downhole events are modeled based upon the determined patterns.

10. A system comprising processing circuitry to perform the method comprising:
extracting a dataset from a database, the dataset comprising normal wellbore data and outlier wellbore data;
clustering a plurality of the outlier data into a plurality of clusters;
segregating the plurality of clusters into a high density cluster and a low density cluster, wherein the high density clusters are utilized as a rimlier;
analyzing the rimlier to determine data variables within the rimlier that indicate a downhole event, wherein the analyzing comprises:
determining a Head Rimlier Factor as defined by:

E
h

E
r

=

Σ
-

p
h

⁡

(
x
)

Assignees

• Landmark Graphics Corp
• Samuel Robello
• Germain Olivier Roger

Inventors

• Samuel Robello
• Germain Olivier Roger

Classifications

• G06F30/20

  Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) · CPC title

• G06F18/2433

  Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection · CPC title

• G06F18/2321

  using statistics or function optimisation, e.g. modelling of probability density functions · CPC title

• G06F16/24565

  Triggers; Constraints · CPC title

• G06F16/355

  Creation or modification of classes or clusters · CPC title