Linguistic goal oriented decision making

• US 10,354,191 B2
• Filed: 09/11/2015
• Issued: 07/16/2019
• Est. Priority Date: 09/12/2014
• Status: Active Grant

• Current Assignee: Chevron USA Incorporated (Chevron Corp.), University of Southern California
• Original Assignee: University of Southern California
• Inventors: Rajati, Mohammad Reza, Mendel, Jerry M., Popa, Andrei, Brenskelle, Lisa A.
• Primary Examiner(s): Vincent, David R
• Application Number:
  US14/851,643
• Publication Number:
  US 20160078345A1
• Time in Patent Office: 1,404 Days
• Field of Search: 706 15, 706 45
• CPC Class Codes: G06N 5/04 Inference or reasoning models

Abstract

A method of attaining a linguistic goal output of a process using a linguistic goal oriented decision making system, comprising setting the output value to a linguistic goal, and establishing a set of descriptive rules using historical data, wherein each descriptive rule has at least one variable, a decision variable, and at least one linguistic output value. The method further comprises grouping the set of descriptive rules into subsets of descriptive rules according to the linguistic output value of each descriptive rule and establishing decision rules based on comparing the grouped subsets of descriptive rules, the one or more decision rules describing a modification of a decision variable associated with at least one of the descriptive rules to affect the process.

Claims

1. A method of attaining a linguistic goal output of a process using a linguistic goal oriented decision making system, the method comprising:

setting a linguistic goal for an output value from among a plurality of linguistic goals;

grouping a set of descriptive rules into at least one subset of descriptive rules, wherein each descriptive rule is established using historical data, each descriptive rule having at least one decision variable, and at least one linguistic output value, the decision variable representing an input to the process, according to the linguistic output value of each descriptive rule;

establishing one or more decision rules based on the grouped subsets of descriptive rules, the one or more decision rules describing a modification of a decision variable associated with at least one of the descriptive rules to affect the process; and

modifying the decision variable based on the one or more decision rules thereby changing the input to the process to arrive at the linguistic goal,

wherein the method is executed by a computer.

2. The method of claim 1, wherein establishing one or more decision rules comprises:

grouping at least one set of consecutive, intermediate goals together, wherein the set of consecutive, intermediate goals comprises a primary goal and a target goal;

establishing pairs of descriptive rules, the pairs of descriptive rules each comprising a first descriptive rule and a second descriptive rule, wherein the first descriptive rule has a primary goal in its consequent and the second descriptive rule has a target goal in its consequent; and

comparing each of the pairs of descriptive rules to establish pairs of decision rules, the pairs of decision rules each comprising a first decision rule and a second decision rule, wherein the pairs of decision rules define a modification of the decision variable of the first descriptive rule based on a comparison to the second descriptive rule.

3. The method of claim 2, further comprising calculating a weight for each of the pairs of decision rules.

4. The method of claim 2, wherein comparing the first descriptive rule to the second descriptive rule further comprises:

computing a difference between a first decision variable associated with the first descriptive rule with a second decision variable associated with the second descriptive rule.

5. The method of claim 1, wherein the historical data is a combination of real data and expert data.

6. The method of claim 1, wherein the plurality of linguistic goals includes low, medium, and high or equivalently named operational states.

7. The method of claim 1, wherein the process is selected from among a group of processes comprising:

a steamflood wellhead system;

a waterflood wellhead system;

a steam injection distribution system;

an acid stimulation system;

a drilling prediction system;

a financial system;

an economic system;

a pharmacological system;

a medical system;

a biological system; and

a social system.

8. The method of claim 1, further comprising a pruning process, the pruning process comprising:

establishing at least one group, wherein each group comprises one or more decision rules having identical one or more antecedents and one or more consequents, wherein each decision rule has an associated weight; and

for each established group, retaining a decision rule having a highest weight.

9. The method of claim 8, further comprising modifying the one or more consequents of each of the retained rules, wherein modifying comprises taking a triangular norm between membership functions and the associated weight of the retained decision rule.

10. A decision-making system comprising:

a computing device including a processor and a memory communicatively coupled to the processor, wherein the computing device is communicatively connected to one or more data streams associated with a process and at least one input to the process; and

a descriptive rules engine stored in the memory, the descriptive rules engine including one or more descriptive rules;

wherein the memory stores instructions executable by the processor to perform a method of attaining a linguistic goal output of the process, the method including;

changing the numerical value of input data into fuzzy sets;

generating one or more descriptive rules with a historical data set defining historical data for a set of variables associated with the process;

aggregating decision rules to produce a decision set representing actions to take place on a decision variable based on a comparison of one or more descriptive rules defining the process, the descriptive rules each including at least a decision variable and a linguistic output value;

converting the one or more fuzzy sets to corresponding numerical output values; and

adjusting the at least one input to the process based on a numerical output value.

11. The decision-making system of claim 10, wherein the process comprises a steamflood system or wellflood system, and wherein an adjustment engine adjusts a choke setting of an input to the process.

12. The decision-making system of claim 11, wherein the choke setting is adjusted to achieve a linguistic goal of the process.

13. The decision-making system of claim 10, wherein the instructions further cause the computing system to perform establishing membership functions for the historical data and establishing descriptive rules using the membership functions as well as historical data.

14. A non-transitory computer-readable storage medium comprising computer-executable instructions which, when executed by a computing system, cause the computing system to perform a method of attaining a linguistic goal output of a hydrocarbon well control process using a linguistic goal oriented decision making system, the method comprising:

setting at least one linguistic goal for at least one output value from among a plurality of linguistic goals;

establishing a set of descriptive rules using historical data, each descriptive rule having at least one decision variable, and a linguistic output value, the decision variable representing an input to the hydrocarbon well control process;

grouping the set of descriptive rules into at least one subset of descriptive rules according to the linguistic output value of each descriptive rule;

establishing one or more decision rules based on the grouped subsets of descriptive rules, the one or more decision rules describing a modification of a decision variable associated with at least one of the descriptive rules to affect the hydrocarbon well control process; and

adjusting at least one physical input to the hydrocarbon well control process based on the one or more decision rules to arrive at the linguistic goal.

15. The computer-readable storage medium of claim 14, wherein establishing one or more decision rules comprises:

grouping at least one set of consecutive, intermediate goals together, wherein the set of consecutive, intermediate goals comprises a primary goal and a target goal;

establishing pairs of descriptive rules, the pairs of descriptive rules each comprising a first descriptive rule and a second descriptive rule, wherein the first descriptive rule has a primary goal in its consequent and the second descriptive rule has a target goal in its consequent; and

comparing each of the pairs of descriptive rules to establish pairs of decision rules, the pairs of decision rules each comprising a first decision rule and a second decision rule, wherein the pairs of decision rules define a modification of the decision variable of the first descriptive rule based on a comparison to the second descriptive rule.

16. The computer-readable storage medium of claim 15, further comprising calculating a weight for each of the pairs of decision rules.

17. The computer-readable storage medium of claim 15, wherein comparing the first descriptive rule to the second descriptive rule further comprises:

computing a difference between a first decision variable associated with the first descriptive rule with a second decision variable associated with the second descriptive rule.

18. The computer-readable storage medium of claim 15, wherein the hydrocarbon well control process is selected from among a group of processes comprising:

a steamflood process;

a waterflood process;

a steam injection distribution process;

an acid stimulation process; and

a drilling prediction process.