Petroleum production optimization utilizing adaptive network and genetic algorithm techniques

• US 6,236,894 B1
• Filed: 12/19/1997
• Issued: 05/22/2001
• Est. Priority Date: 12/19/1997
• Status: Expired due to Term

• Current Assignee: Atlantic Richfield Company (BP PLC)
• Original Assignee: Atlantic Richfield Company (BP PLC)
• Inventors: McCormack, Michael D., Stoisits, Richard F., MacAllister, Donald J., Crawford, Kelly D.
• Primary Examiner(s): Grant, William
• Assistant Examiner(s): Rapp, Chad
• Application Number:
  US08/994,975
• Time in Patent Office: 1,250 Days
• Field of Search: 340/853.3, 700/28, 700/29, 700/32, 702/9, 702/102, 702/45, 731/525.2, 731/521.9, 731/524.7, 738/647.2, 166/245, 166/306, 166/250.01, 166/250.15, 166/266, 166/364, 166/370, 166/373, 166/374, 166/52, 166/53, 166/263, 175/48, 175/50, 299/4, 299/2
• US Class Current: 700/28
• CPC Class Codes: G05B 13/0265 the criterion being a learning criterion
• CPC Class Codes: G05B 13/027 using neural networks only

Abstract

A computer system and method of operating the same to optimize the operating conditions of a petroleum production field, in which a plurality of wells are arranged according to drill sites, and connected to one or more central processing facilities, is disclosed. In this disclosed embodiment, gas compression capacity is a significant constraint on the operation of the complex production field, and surface line hydraulic effects of well production are to be considered in the optimization. A genetic algorithm is used to generate, and iteratively evaluate solution vectors, which are combinations of field operating parameters such as incremental gas-oil ratio cutoff and formation gas-oil ratio cutoff values. The evaluation includes the operation of an adaptive network to determine production header pressures, followed by modification of well output estimates to account for changes in the production header pressure. Convergence of the genetic algorithm identifies one of the solution vectors as containing an optimal combination of field operating parameters that may be used by production personnel to set the operating conditions of the field.

Claims

1. A method of operating a computer to determine optimized operating conditions for a petroleum production field having a plurality of wells, comprising the steps of:

producing a first generation of solution vectors, each solution vector including a plurality of field operating parameters associated with the production field;

evaluating product output from the production field for each of the first generation of solution vectors;

producing a second generation of solution vectors from selected solution vectors of the first generation, according to a genetic algorithm;

repeating the evaluating and producing steps until a convergence criterion is met; and

reporting the field operating parameters from a solution vector having an optimum product output, for use in operating the production field.

2. The method of claim 1, wherein the evaluating step comprises:

determining the status of each of the plurality of wells from the field operating parameters;

for each of the plurality of wells, estimating fluid output from the well based upon its status;

operating a surface line hydraulic model using the estimated fluid output from each well, to estimate production header pressures for each of the plurality of wells;

modifying the estimated fluid output from the well using the estimated production header pressures; and

determining the product output from the production field from the modified estimated fluid output.

3. The method of claim 2, wherein the surface line hydraulic model is an adaptive network model.

4. The method of claim 2, further comprising:

training the adaptive network model by way of backpropagation;

and wherein the step of operating the surface line hydraulic model is performed by operating the adaptive network model in recall mode.

5. The method of claim 2, wherein the plurality of wells are arranged according to a plurality of drill sites, each drill site associated with a group of wells;

wherein the evaluating step further comprises;

for each of the plurality of drill sites, after the step of estimating fluid output from the well, summing the fluid output for the group of wells associated with the drill site;

and wherein the step of operating the surface line hydraulic model estimates production header pressures at each of the plurality of drill sites using the summed fluid output for each of the plurality of drill sites.

6. The method of claim 5, wherein the surface line hydraulic model is an adaptive network model.

7. The method of claim 5, wherein the field operating parameters comprise incremental gas-oil ratio cutoff values and formation gas-oil ratio cutoff values associated with each of the plurality of drill sites.

8. The method of claim 1, wherein the field operating parameters comprise well producing status indications for each of the plurality of wells.

9. The method of claim 1, wherein the product output evaluated in the evaluating step comprises an oil output value and a gas output value from the production field;

and further comprising, for each solution vector;

comparing the gas output value from the evaluating step with a gas compression capacity value for at least one central processing facility in the production field;

responsive to the comparing step determining that the gas output value exceeds the gas compression capacity value, downgrading the oil output value for the solution vector;

and wherein the probability of selection in the producing step corresponds to the oil output value.

10. A computer system for determining optimized operating conditions for a petroleum production field having a plurality of wells coupled to a central processing facility by way of a network of surface lines, comprising:

computer-readable memory for storing well data corresponding to operating characteristics of each of the plurality of wells;

an output device;

a system computer, coupled to the computer-readable memory and to the output device, and programmed to perform the operations of;

randomly producing an initial generation of solution vectors, each solution vector including a plurality of field operating parameters associated with the production field;

for each of a plurality of generations of solution vectors beginning with the initial generation, performing the operations of;

evaluating product output from the production field for each of the solution vectors in a current generation using the operating characteristics of each of the plurality of wells;

producing a subsequent generation of solution vectors from selected solution vectors of the current generation, according to a genetic algorithm; and

reporting, on the output device, the field operating parameters from a solution vector having an optimum product output, for use in operating the production field.

11. The computer system of claim 10, wherein the computer-readable memory is also for storing a surface line hydraulic model corresponding to operating characteristics of the network of surface lines;

and wherein the system computer is programmed to perform the evaluating operation by;

determining the status of each of the plurality of wells from the field operating parameters;

for each of the plurality of wells, estimating fluid output from the well based upon its status;

operating the surface line hydraulic model using the estimated fluid output from each well, to estimate production header pressures for each of the plurality of wells;

modifying the estimated fluid output from the well using the estimated production header pressures; and

determining the product output from the production field from the modified estimated fluid output.

12. The computer system of claim 11, wherein the surface line hydraulic model is stored in the computer-readable memory in the form of weighting factors for an adaptive network;

and wherein the system computer is programmed to perform the operation of operating the surface line hydraulic model by operating the adaptive network, in recall mode, using the estimated fluid output from each well.

13. The computer system of claim 11, wherein the plurality of wells are arranged according to a plurality of drill sites, each drill site associated with a group of wells;

wherein the evaluating operation further comprises;

for each of the plurality of drill sites, after the step of estimating fluid output from the well, summing the fluid output for the group of wells associated with the drill site;

and wherein the system computer is programmed to operate the surface line hydraulic model to estimate production header pressures at each of the plurality of drill sites using the summed fluid output for each of the plurality of drill sites.

14. The computer system of claim 13, wherein the field operating parameters comprise incremental gas-oil ratio cutoff values and formation gas-oil ratio cutoff values associated with each of the plurality of drill sites.

15. The computer system of claim 10, wherein the field operating parameters comprise well producing status indications for each of the plurality of wells.

16. The computer system of claim 10, wherein the product output evaluated in the evaluating operation comprises an oil output value and a gas output value from the production field;

wherein the system computer is further programmed to perform the operations of, for each solution vector;

comparing the gas output value from the evaluating operation with a gas compression capacity value for at least one central processing facility in the production field;

responsive to the comparing operation determining that the gas output value exceeds the gas compression capacity value, downgrading the oil output value for the solution vector;

and wherein the probability of selection in the operation of producing a subsequent generation of solution vectors corresponds to the oil output value.

17. A computer-readable memory configured so that, when read and used by a computer, the computer is directed to determine optimized operating conditions for a petroleum production field having a plurality of wells coupled to a central processing facility by way of a network of surface lines, said computer directed by a plurality of operations comprising:

producing an initial generation of solution vectors, each solution vector including a plurality of field operating parameters associated with the production field;

for each of a plurality of generations of solution vectors beginning with the initial generation, performing the operations of;

evaluating product output from the production field for each of the solution vectors in a current generation using operating characteristics of each of the plurality of wells;

producing a subsequent generation of solution vectors from selected solution vectors of the current generation, according to a genetic algorithm; and

reporting, on an output device, the field operating parameters from a solution vector having an optimum product output, for use in operating the production field.

18. The computer-readable memory of claim 17, the evaluating operation further comprises:

determining the status of each of the plurality of wells from the field operating parameters;

for each of the plurality of wells, estimating fluid output from the well based upon its status;

operating a surface line hydraulic model corresponding to operating characteristics of the network of surface lines, using the estimated fluid output from each well, to estimate production header pressures for each of the plurality of wells;

modifying the estimated fluid output from the well using the estimated production header pressures; and

determining the product output from the production field from the modified estimated fluid output.

19. The computer-readable memory of claim 18, wherein the computer-readable memory stores the surface line hydraulic model in the form of weighting factors for an adaptive network;

and wherein the operation of operating the surface line hydraulic model is performed by operating the adaptive network, in recall mode, using the estimated fluid output from each well.

20. The computer-readable memory of claim 17, wherein the plurality of wells are arranged according to a plurality of drill sites, each drill site associated with a group of wells;

wherein the evaluating operation further comprises;

for each of the plurality of drill sites, after the step of estimating fluid output from the well, summing the fluid output for the group of wells associated with the drill site;

and wherein the operation of operating the surface line hydraulic model to estimate production header pressures at each of the plurality of drill sites uses the summed fluid output for each of the plurality of drill sites.

21. The computer-readable memory of claim 20, wherein the field operating parameters comprise incremental gas-oil ratio cutoff values and formation gas-oil ratio cutoff values associated with each of the plurality of drill sites.

22. The computer-readable memory of claim 17, wherein the field operating parameters comprise well producing status indications for each of the plurality of wells.

23. The computer-readable memory of claim 17, wherein the product output evaluated in the evaluating operation comprises an oil output value and a gas output value from the production field;

wherein the plurality of operations further comprise, for each solution vector;

comparing the gas output value from the evaluating operation with a gas compression capacity value for at least one central processing facility in the production field;

responsive to the comparing operation determining that the gas output value exceeds the gas compression capacity value, downgrading the oil output value for the solution vector;

and wherein the probability of selection in the operation of producing a subsequent generation of solution vectors corresponds to the oil output value.