Automated material balance system for hydrocarbon reservoirs using a genetic procedure

• US 5,924,048 A
• Filed: 03/14/1997
• Issued: 07/13/1999
• Est. Priority Date: 03/14/1997
• Status: Expired due to Term

• Current Assignee: Atlantic Richfield Company Incorporated (BP PLC)
• Original Assignee: Atlantic Richfield Company Incorporated (BP PLC)
• Inventors: McCormack, Michael D., Stoisits, Richard F., MacAllister, Donald J., Ma, Tuan D., Scherer, Perry W.
• Primary Examiner(s): McElheny, Jr., Donald E.
• Application Number:
  US08/818,733
• Time in Patent Office: 851 Days
• Field of Search: 702/12, 702/13, 706/929
• US Class Current: 702/13
• CPC Class Codes: E21B 49/00 Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells
• CPC Class Codes: G05B 13/0265 the criterion being a learning criterion

Abstract

A system for producing a material balance solution for well patterns in a hydrocarbon reservoir is described that automatically optimizes the fluid allocation factors for each well used in determining the solution. The system automatically optimizes estimates for the allocation factors to be used in the material balance solution by randomly generating a first generation of allocation factor strings, each string in the generation assigning allocation factors to each of the wells in the reservoir. A fitness function value is determined for each of the strings by evaluating a fitness function, wherein the fitness function comprises the sum of the differences between computed and measured field pressures for each pattern, and the sum of the differences between target allocation factors and the allocation factors specified within the string for each well. A succeeding generation of allocation factor strings is produced according to a genetic algorithm. The process of determining a fitness function value for each of the strings is then repeated for the succeeding generation. The string having a fitness function value meeting a specified criteria is identified, wherein the identified string represents the optimized estimates of the allocation factors for the reservoir for use in determining the material balance solution.

Claims

1. A method of producing a material balance solution for well patterns in a hydrocarbon reservoir that automatically optimizes the fluid allocation factors for each well used in determining the solution, comprising the steps of:

randomly generating a first generation of allocation factor strings, each string in the generation assigning allocation factors to each of the wells in the reservoir representing the fraction of fluid injected into or produced from a well into or out of a well pattern;

determining a fitness function value for each of the strings by evaluating a fitness function, wherein the fitness function comprises the sum of the differences between computed and measured field pressures for each pattern, and the sum of the differences between target allocation factors and the allocation factors specified within the string for each well;

producing a succeeding generation of allocation factor strings according to a genetic algorithm;

repeating the step of determining a fitness function value for each of the strings in the succeeding generation; and

identifying the string having fitness function value meeting a specified criteria, wherein the identified string includes optimized estimates of the allocation factors for the reservoir for use in determining the material balance solution.

2. The method of claim 1 wherein the identifying step further comprises determining whether the predetermined criteria has been satisfied, whereupon the string identification occurs when the criteria has been satisfied;

and further comprising;

responsive to determining that the predetermined criteria is not satisfied, repeating the steps of producing a succeeding generation of strings; and

determining a fitness function value for each of the strings in the succeeding generation.

3. The method of claim 1 wherein the step of producing a succeeding generation of allocation factor strings comprises:

creating a probability value for each of the allocation factor strings in first generation by determining the fractional contribution of each string in the generation to the sum of the fitness function values for the strings in the generation;

selecting parent sequences from the first generation according to the probability values;

randomly selecting pairs of parent strings; and

crossing over the selected pairs of parent strings relative to one another at a randomly selected value position to produce the succeeding generation.

4. The method of claim 1 wherein the fitness function further comprises:

assigning a first weighting factor to the sum of the pressures for each pattern; and

assigning a second weighting factor to the sum of the allocations for each well.

5. The method of claim 1 wherein the step of generating the first generation of allocation factor strings further comprises, for each string, performing a normalization procedure on the factors in the string wherein conservation of mass pertaining the injection and production of fluids is achieved in each well.

6. The method of claim 1 further comprising the step of performing material balance calculations using the allocation factors and reservoir data to produce the computed pressure values for use in evaluating the fitness function.

7. The method of claim 6 wherein the reservoir data used in performing the material balance calculations includes one or more of production and injection history data, initial pressure data, initial hydrocarbon pore volume data, initial fluid saturation data and fluid properties data.

8. The method of claim 1 wherein the specified criteria is one that produces pressures in a range of accuracy acceptable to field engineers.

9. The method of claim 1 wherein the first generation of strings comprises a number of strings which is more than half the number of allocation values in each string and less than twice the number of allocation values in each string.

10. The method of claim 1 wherein a well pattern represents a volume of the reservoir centered about a production well with fluid contributions from injection wells at the pattern boundaries.

11. The method of claim 10 wherein the step of constructing a first generation of allocation factor strings further comprises, for each string, performing a normalization procedure on the factors in the string wherein conservation of mass pertaining the injection and production of fluids is achieved in each well.

12. The method of claim 10 wherein the reservoir data used in performing the material balance calculations includes one or more of production and injection history data, initial pressure data, initial hydrocarbon pore volume data, initial fluid saturation data and fluid properties data.

13. The method of claim 1 wherein the predetermined criteria is one that produces pressures in a range of accuracy acceptable to field engineers.

14. The method of claim 1 wherein the fitness function comprises the summation of differences between computed and measured field pressures for each pattern, and the summation of the differences between a target allocation and the allocation specified within the string for each well.

15. A method of optimizing estimates of fluid allocation factors for well patterns in a hydrocarbon reservoir, the allocation factor for each well representing the fraction of fluid injected to or produced from a well into or out of a well pattern, and each well pattern representing a defined volume of the reservoir centered about a production well with fluid contributions from injection wells at the pattern boundaries, comprising the steps of:

constructing a first generation of allocation factor strings, where each string represents the allocation factors for each well in the reservoir;

for each string, performing a normalization procedure on the allocation factors in the string such that conservation of mass pertaining to the production and injection of fluids is achieved with respect to each well;

for each string, performing material balance calculations using the normalized allocation factors and reservoir data to produce pressure values for each well pattern;

determining a fitness function value for each of the strings by evaluating a fitness function;

determining if any of the fitness function values for any of the strings meet a predetermined criteria;

if the criteria is not met, then producing a succeeding generation of allocation factor strings according to a genetic algorithm and repeating the process of determining the fitness function values for each of the strings in the succeeding generation until the predetermined criteria is satisfied;

such that when the criteria is satisfied, the satisfying string represents the optimal set of allocation factors for the reservoir.

16. The method of claim 15 wherein the fitness function comprises the summation of differences between computed and measured field pressures for each pattern, and the summation of the differences between a target allocation and the allocation specified within the string for each well.

17. A computer program for producing a material balance solution for well patterns in a hydrocarbon reservoir that automatically optimizes the fluid allocation factors for each well used in determining the solution, the program stored on computer readable media, comprising:

instructions for randomly generating a first generation of allocation factor strings, each string in the generation assigning allocation factors to each of the wells in the reservoir representing the fraction of fluid injected into or produced from a well into or out of a well pattern;

instructions for determining a fitness function value for each of the strings by evaluating a fitness function, wherein the fitness function comprises the sum of the differences between computed and measured field pressures for each pattern, and the sum of the differences between target allocation factors and the allocation factors specified within the string for each well;

instructions for producing a succeeding generation of allocation factor strings according to a genetic algorithm;

instructions for repeating the of determining a fitness function value for each of the strings in the succeeding generation; and

instructions for identifying the string having fitness function value meeting a specified criteria, wherein the identified string includes optimized estimates of the allocation factors for the reservoir for use in determining the material balance solution.

18. The program of claim 17 wherein the identifying instructions further comprise determining whether the predetermined criteria has been satisfied, whereupon the string identification occurs when the criteria has been satisfied;

and further comprising;

instructions responsive to determining that the predetermined criteria is not satisfied, repeating the instructions of producing a succeeding generation of strings; and

determining a fitness function value for each of the strings in the succeeding generation.

19. The program of claim 17 wherein the producing a succeeding generation of allocation factor strings comprises:

instructions for creating a probability value for each of the allocation factor strings in first generation by determining the fractional contribution of each string in the generation to the sum of the fitness function values for the strings in the generation;

instructions for selecting parent sequences from the first generation according to the probability values;

instructions for randomly selecting pairs of parent strings; and

instructions for crossing over the selected pairs of parent strings relative to one another at selected value position to produce the succeeding generation.

20. A system for producing a material balance solution for well patterns in a hydrocarbon reservoir that automatically optimizes the fluid allocation factors for each well used in determining the solution, comprising:

means for randomly generating a first generation of allocation factor strings, each string in the generation assigning allocation factors to each of the wells in the reservoir representing the fraction of fluid injected into or produced from a well into or out of a well pattern;

means for determining a fitness function value for each of the strings by evaluating a fitness function, wherein the fitness function comprises the sum of the differences between computed and measured field pressures for each pattern, and the sum of the differences between target allocation factors and the allocation factors specified within the string for each well;

means for producing a succeeding generation of allocation factor strings according to a genetic algorithm;

means for repeating the of determining a fitness function value for each of the strings in the succeeding generation; and

means for identifying the string having fitness function value meeting a specified criteria, wherein the identified string includes optimized estimates of the allocation factors for the reservoir for use in determining the material balance solution.

21. A method of producing a material balance solution for well patterns in a hydrocarbon reservoir that automatically optimizes the fluid allocation factors for each well used in determining the solution, the allocation factors representing the fraction of fluid injected to or produced from a well into or out of a well pattern, comprising the steps of:

constructing a first generation of allocation factor strings, where each string represents a random estimate of the allocation factors for each well in the reservoir;

for each string, performing material balance calculations using the allocation factors and reservoir data to produce pressure values for each well pattern;

for each string, determining a fitness function value using a fitness function that has as inputs the produced pressure values for each well pattern and the allocation factors for each well;

determining if the fitness function value for any of the strings meet a predetermined criteria indicative of accuracy of the estimated allocation factors;

if the criteria is not met, then producing a succeeding generation of allocation factor strings according to a genetic procedure and repeating the determining of the fitness function value for each of the strings in the succeeding generation until the criteria is satisfied;

such that when the criteria is satisfied, the satisfying string represents a nearly optimal estimate of allocation factors for the reservoir and a material balance solution is produced using the estimate of allocation factors.