System and method for automatic, full-field and multi-well permeability-thickness conditioning of geological models

Abstract

A method includes: accessing a full-field model for a reservoir containing wells without well-test events and wells with well-test events; importing records from all wells and well-test results from the test-available wells; creating a local grid refinement in a vicinity of each test-available well; launching a first iteration of simulation; using artificial intelligence to determine, for each test-available well, a first ratio of a simulated pressure derivative from results of the first iteration of simulation and an observed pressure derivative from the well-test results; launching a second iteration of full-field simulation using the model where the first ratio is used as a permeability multiplier in the vicinity of each test-available well; and responsive to meeting an objective of the simulation, generating a report delineating re-distributed permeability in the reservoir based on combining the records from all wells and the well-test results from the test-available wells corrected by respective permeability multipliers.

What is claimed is: 
     
         1 . A computer-implemented method comprising:
 accessing a full-field multi-well model for a reservoir containing a first plurality of wells each without no well-test event and a second plurality of wells each with a well-test event;   importing, into the full-field multi-well model, historical flowrate records from both the first and second plurality of wells, and well-test results from the second plurality of wells;   creating, in the full-field multi-well model, a local grid refinement in a vicinity of each well from the second plurality of wells;   launching a first iteration of full-field simulation using the full-field multi-well model covering both the first and the second plurality of wells of the reservoir;   determining, using artificial intelligence and for each well from the second plurality of wells, a first ratio of a simulated pressure derivative from results of the first iteration of full-field simulation and an observed pressure derivative from the well-test results;   launching a second iteration of full-field simulation using the full-field multi-well model for the reservoir where the first ratio is used as a permeability multiplier in the vicinity of each well from the second plurality of wells such that the observed pressure derivative and the simulated pressure derivative become more matched; and   responsive to meeting an objective of the full-field simulation, generating a report delineating re-distributed permeability in the reservoir based on combining the historical flowrate records—from the first and second plurality of wells—and the well-test results from the second plurality of wells corrected by respective permeability multipliers.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein the objective of the full-field simulation comprises one of: a number of iterations, or a difference between the simulated pressure derivative and the observed pressure derivative. 
     
     
         3 . The computer-implemented method of  claim 1 , wherein the vicinity is defined by a radius configurable by an operator through a user interface. 
     
     
         4 . The computer-implemented method of  claim 2 , wherein the vicinity of each well from the second plurality of wells defines a drainage area for a corresponding well. 
     
     
         5 . The computer-implemented method of  claim 1 , wherein the simulated pressure derivative and the observed pressure derivative are respective averages within a middle-time-region (MTR) of a corresponding pressure derivative curve. 
     
     
         6 . The computer-implemented method of  claim 5 , wherein the MTR is where the corresponding pressure derivative curve is characterized by an approximately constant level that varies within about 10%. 
     
     
         7 . The computer-implemented method of  claim 1 , further comprising:
 responsive to determining that the objective of the full-field simulation is not met, calculating, for each well from the second plurality of wells, a second ratio of a simulated pressure derivative from results of the second iteration of full-field simulation and the observed pressure derivative from the well-test results.   
     
     
         8 . The computer-implemented method of  claim 7 , further comprising:
 launching a third iteration of full-field simulation using the full-field multi-well model for the reservoir where the second ratio is multiplied with the first ratio to product a new permeability multiplier in the vicinity of each well from the second plurality of wells such that the observed pressure derivative and the simulated pressure derivative become more matched.   
     
     
         9 . The computer-implemented method of  claim 1 , further comprising:
 planning a location of a new well in the reservoir based on, at least in part, a regenerated static model incorporating the first ratio for kh calibration in the report.   
     
     
         10 . The computer-implemented method of  claim 9 , further comprising:
 estimating oil production in the reservoir based on, at least in part, the regenerated static model incorporating the first ratio for kh calibration in the report.   
     
     
         11 . A computer system comprising one or more computer processors configured to perform operations of:
 accessing a full-field multi-well model for a reservoir containing a first plurality of wells each with no well-test event and a second plurality of wells each with a well-test event;   importing, into the full-field multi-well model, historical flowrate records from both the first and second plurality of wells and well-test results from the second plurality of wells;   creating, in the full-field multi-well model, a local grid refinement in a vicinity of each well from the second plurality of wells;   launching a first iteration of full-field simulation using the full-field multi-well model covering both the first and the second plurality of wells of the reservoir;   determining, using artificial intelligence and for each well from the second plurality of wells, a first ratio of a simulated pressure derivative from results of the first iteration of full-field simulation and an observed pressure derivative from the well-test results;   launching a second iteration of full-field simulation using the full-field multi-well model for the reservoir where the first ratio is used as a permeability multiplier in the vicinity of each well from the second plurality of wells such that the observed pressure derivative and the simulated pressure derivative become more matched; and   responsive to meeting an objective of the full-field simulation, generating a report delineating re-distributed permeability in the reservoir based on combining the historical flowrate records—from the first and the second plurality of wells—and the well-test results from the second plurality of wells corrected by respective permeability multipliers.   
     
     
         12 . The computer system of  claim 11 , wherein the objective of the full-field simulation comprises one of: a number of iterations, or a difference between the simulated pressure derivative and the observed pressure derivative. 
     
     
         13 . The computer system of  claim 11 , wherein the vicinity is defined by a radius configurable by an operator through a user interface. 
     
     
         14 . The computer system of  claim 12 , wherein the vicinity of each well from the second plurality of wells defines a drainage area for a corresponding well. 
     
     
         15 . The computer system of  claim 11 , wherein the simulated pressure derivative and the observed pressure derivative are respective averages within a middle-time-region (MTR) of a corresponding pressure derivative curve. 
     
     
         16 . The computer system of  claim 15 , wherein the MTR is where the corresponding pressure derivative curve is characterized by an approximately constant level that varies within about 10%. 
     
     
         17 . The computer system of  claim 11 , wherein the operations further comprise:
 responsive to determining that the objective of the full-field simulation is not met, calculating, for each well from the second plurality of wells, a second ratio of a simulated pressure derivative from results of the second iteration of full-field simulation and the observed pressure derivative from the well-test results.   
     
     
         18 . The computer system of  claim 17 , wherein the operations further comprise:
 launching a third iteration of full-field simulation using the full-field multi-well model for the reservoir where the second ratio is multiplied with the first ratio to product a new permeability multiplier in the vicinity of each well from the second plurality of wells such that the observed pressure derivative and the simulated pressure derivative become more matched.   
     
     
         19 . The computer system of  claim 11 , wherein the operations further comprise:
 planning a location of a new well in the reservoir based on, at least in part, a regenerated static model incorporating the first ratio for kh calibration in report.   
     
     
         20 . The computer system of  claim 19 , wherein the operations further comprise:
 estimating oil production in the reservoir based on, at least in part, the regenerated static model incorporating the first ratio for kh calibration in the report.