Combining belief networks to generate expected outcomes

• US 7,861,800 B2
• Filed: 10/08/2008
• Issued: 01/04/2011
• Est. Priority Date: 10/08/2008
• Status: Active Grant

• Current Assignee: Schlumberger Technology Corporation (Schlumberger NV)
• Original Assignee: Schlumberger Technology Corporation (Schlumberger NV)
• Inventors: Chen, Charles, Chapman, Clinton D.
• Primary Examiner(s): Neuder; William P
• Application Number:
  US12/247,352
• Publication Number:
  US 20100084191A1
• Time in Patent Office: 818 Days
• Field of Search: 175/24, 175/40
• US Class Current: 175/24
• CPC Class Codes: E21B 2200/22 Fuzzy logic, artificial intelligence, neural networks or the like
• CPC Class Codes: E21B 44/00 Automatic control systems specially adapted for drilling operations, i.e. self-operating systems which function to carry out or modify a drilling operation without intervention of a human operator, e.g. computer-controlled drilling systems; Systems specially adapted for monitoring a plurality of drilling variables or conditions
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A method is described for determining an oilfield parameter for a drilling operation in an oilfield, the oilfield having a well site with a drilling tool advanced into a subterranean formation with geological structures and reservoirs therein. A first set of decision factors and a second set of decision factors about the drilling operation are identified. Each of the first set of decision factors is contained within a first set of nodes, and each of the second set of decision factors is contained within a second set of nodes. Both the first set of nodes and the second set of nodes contain a set of common nodes, which are common to both the first and second sets. The first set of nodes is associated to create a first belief network and the second set of nodes is associated to create a second belief network. Then, the set of common nodes of the first belief network is associated with the set of common nodes of the second belief network to form a multinet belief network. Oilfield parameter can be generated from the multinet belief network.

Claims

1. A method for determining an oilfield parameter for a drilling operation in an oilfield, the oilfield having a well site with a drilling tool advanced into a subterranean formation with geological structures and reservoirs therein, the method comprising:

identifying a first set of decision factors about the drilling operation and a second set of decision factors about the drilling operation, wherein each of the first set of decision factors is contained within a first set of nodes, and wherein each of the second set of decision factors is contained within a second set of nodes, wherein both the first set of nodes and the second set of nodes contain a set of common nodes;

associating the first set of nodes to create a first belief network and associating the second set of nodes to create a second belief network;

associating the set of common nodes of the first belief network with the set of common nodes of the second belief network to form a multinet belief network; and

generating at least one oilfield parameter from the multinet belief network.

2. The method of claim 1 further comprising:

assigning a first degree of certainty to the first belief network; and

assigning a second degree of certainty to the second belief network.

3. The method of claim 2, the step of generating the at least one oilfield parameter from the multinet belief network comprising:

preferring the first belief network over the second belief network in generating the at least one oilfield parameter to form a first preference, wherein the first preference is based at least in part on a relative difference between the first degree of certainty and the second degree of certainty.

4. The method of claim 2, wherein the first belief network comprises a first node and a second node, and wherein the second belief network comprises a third node and a fourth node, wherein the first node and the third node are first common nodes, and wherein the second node and the fourth node are second common nodes, the step of associating the set of common nodes of the first belief network with the set of common nodes of the second belief network to form a multinet belief network comprising:

associating the first node with the third node; and

associating the second node with the fourth node.

5. The method of claim 1, wherein the first set of decision factors about the drilling operation consists of information selected from the list of:

equipment for use in the well site;

completion techniques used in the well site;

production data from the well site;

well log data from the well site;

mud log data from the well site;

expert information; and

combinations thereof.

6. The method of claim 1 further comprising:

directing the drilling operation in the oilfield at the well site based on the at least one oilfield parameter generated from the multinet belief network.

7. The method of claim 6 further comprising:

collecting a set of information; and

entering the set of information into at least one of the first set of nodes or the second set of nodes in the multinet belief network, wherein the step of generating the at least one oilfield parameter from the multinet belief network is responsive to entering the set of information; and

adjusting the drilling operation based on the at least one oilfield parameter generated from the multinet belief network.

8. The method of claim 7, wherein the drilling operation in the oilfield at the well site is adjusted in real time based on the at least one oilfield parameter from the multinet belief network.

9. The method of claim 1, wherein the at least one oilfield parameter is one of a potential problem in an oil well, measured and controlled values, downhole tool settings, signal event detection, and validity of entered or measured oilfield parameters, wherein the potential problem in the oil well is selected from the group consisting of:

hole cleaning, sticking, borehole analysis design, trajectory design, whirl, vibration, shock, stability, tendency, bit design, bit selection, offset correlation, kick detection, swab/surge, motor efficiency, drilling automation, positional uncertainty, and a survey program detection;

wherein the measured and controlled values is selected from the group consisting of;

a weight on a bit, a hookload, a surface weight on the bit, a pump flow rate, and a mud rheology;

wherein the downhole tool setting is selected from the group consisting of a preferred tool face, a rotary steerable system idle mode, and a logging while drilling control parameter; and

wherein the signal event detection is selected from one of the group consisting of telemetry, logging while drilling, formation determination, signal filtering, and change point determination.

10. A method for determining an oilfield parameter for a drilling operation in an oilfield, the oilfield having a well site with a drilling tool advanced into a subterranean formation with geological structures and reservoirs therein, the method comprising:

assigning a first degree of certainty to a first belief network in a multinet belief network comprising a plurality of belief networks;

assigning a second degree of certainty to a second belief network within the multinet belief network; and

generating at least one oilfield parameter from the first belief network and the second belief network.

11. The method of claim 10, further comprising:

identifying a first set of decision factors about the drilling operation and a second set of decision factors about the drilling operation, wherein each of the first set of decision factors is contained within a first set of nodes, and wherein each of the second set of decision factors is contained within a second set of nodes, wherein both the first set of nodes and the second set of nodes contain a set of common nodes;

associating the first set of nodes to create the first belief network and associating the second set of nodes to create the second belief network.

12. The method of claim 11 further comprising:

associating the set of common nodes of the first belief network with the set of common nodes of the second belief network to form the multinet belief network, wherein the step of generating the at least one oilfield parameter comprises generating the at least one oilfield parameter from the multinet belief network.

13. The method of claim 11, the step of generating the at least one oilfield parameter from the first belief network and the second belief network further comprising:

preferring the first belief network over the second belief network in generating the at least one oilfield parameter to form a first preference, wherein the first preference is based at least in part on a relative difference between the first degree of certainty and the second degree of certainty.

14. The method of claim 12, wherein the first belief network comprises a first node and a second node, and wherein the second belief network comprises a third node and a fourth node, wherein the first node and the third node are first common nodes, and wherein the second node and the fourth node are second common nodes, the step of associating the set of common nodes of the first belief network with the set of common nodes of the second belief network to form the multinet belief network comprising:

associating the first node with the third node; and

associating the second node with the fourth node.

15. The method of claim 11, wherein the first set of decision factors about the drilling operation consists of information selected from the list of:

equipment for use in the well site;

completion techniques used in the well site;

production data from the well site;

well log data from the well site;

mud log data from the well site;

expert information; and

combinations thereof.

16. The method of claim 11 further comprising:

directing the drilling operation in the oilfield at the well site based on the at least one oilfield parameter generated from the first belief network and the second belief network.

17. The method of claim 11 further comprising:

collecting a set of information; and

entering the set of information into at least one of the first set of nodes of the first belief network or the second set of nodes of the second belief network, wherein the step of generating at least one oilfield parameter from the first belief network and the second belief network is responsive to entering the set of information; and

adjusting the drilling operation based on the at least one oilfield parameter generated from the first belief network and the second belief network.

18. The method of claim 17, wherein the drilling operation in the oilfield at the well site is adjusted in real time based on the at least one oilfield parameter.

19. The method of claim 11, wherein the at least one oilfield parameter is one of a potential problem in an oil well, measured and controlled values, downhole tool settings, signal event detection, and validity of entered or measured oilfield parameters, wherein the potential problem in the oil well is selected from the group consisting of:

hole cleaning, sticking, borehole analysis design, trajectory design, whirl, vibration, shock, stability, tendency, bit design, bit selection, offset correlation, kick detection, swab/surge, motor efficiency, drilling automation, positional uncertainty, and a survey program detection;

wherein the measured and controlled values is selected from the group consisting of;

a weight on a bit, a hookload, a surface weight on the bit, a pump flow rate, and a mud rheology;

wherein the downhole tool setting is selected from the group consisting of a preferred tool face, a rotary steerable system idle mode, and a logging while drilling control parameter; and

wherein the signal event detection is selected from one of the group consisting of telemetry, logging while drilling, formation determination, signal filtering, and change point determination.

20. A method for determining an oilfield parameter for an oilfield operation in an oilfield, the oilfield having a well site with a drilling tool advanced into a subterranean formation with geological structures and reservoirs therein, the method comprising:

identifying a first set of decision factors about the drilling operation and a second set of decision factors about the drilling operation, wherein each of the first set of decision factors is contained within a first set of nodes, and wherein each of the second set of decision factors is contained within a second set of nodes, wherein both the first set of nodes and the second set of nodes contain a set of common nodes;

associating the first set of nodes to create a first belief network and associating the second set of nodes to create a second belief network;

associating the set of common nodes of the first belief network with the set of common nodes of the second belief network to form a multinet belief network; and

generating at least one oilfield parameter from the multinet belief network.

21. The method of claim 20 further comprising:

assigning a first degree of certainty to the first belief network; and

assigning a second degree of certainty to the second belief network.

22. The method of claim 21, the step of generating the at least one oilfield parameter from the multinet belief network comprising:

preferring the first belief network over the second belief network in generating the at least one oilfield parameter to form a first preference, wherein the first preference is based at least in part on a relative difference between the first degree of certainty and the second degree of certainty.

23. The method of claim 21, wherein the first belief network comprises a first node and a second node, and wherein the second belief network comprises a third node and a fourth node, wherein the first node and the third node are first common nodes, and wherein the second node and the fourth node are second common nodes, the step of associating the set of common nodes of the first belief network with the set of common nodes of the second belief network to form a multinet belief network comprising:

associating the first node with the third node; and

associating the second node with the fourth node.