Flow simulator for generating reservoir management workflows and forecasts based on analysis of high-dimensional parameter data space
An apparatus used to generate forecasts from a high-dimensional parameter data space. The apparatus comprising a reservoir model and a flow simulator module. The reservoir model comprising a plurality input variables, output variables, and at least one algorithmic model. The input variables and output variables are generated by the flow simulator module and variables from a formation and reservoir properties database and a field production database. The flow simulator module generates the at least one algorithmic model and the output variables using at least one selected from a group comprising a full-physics flow simulator, proxy flow simulator for assisted history matching, and a proxy flow simulator for field development optimization. The full-physics flow simulator and the two proxy flow simulators generate the at least one algorithmic model using at least one selected from a group comprising the reservoir model, history matching input variables, and optimization input variables.
1. An apparatus for generating forecasts from an input high-dimensional parameter data space, the apparatus comprising:
a reservoir model that includes comprising a plurality of input variables, a plurality of output variables, and at least one algorithmic model, wherein the plurality of input variables and the plurality of output variables are generated by a flow simulator module based on variables from a formation and reservoir properties database and a field database, wherein the flow simulator module is configured to generate the at least one algorithmic model and the plurality of output variables using at least one simulator selected from a full-physics flow simulator, a proxy flow simulator for assisted history matching, and a proxy flow simulator for field development optimization; and
a model order reduction module configured to generate a subset of the plurality of input variables having a reduced parameter space relative to the input high-dimensional parameter data space by identifying input variables within the plurality of input variables useful to approximate at least one output variable of the plurality of output variables, wherein the model order reduction module uses a function decomposition algorithm to identify the input variables within the plurality of input variables useful to approximate the at least one output variable so as to generate the subset of the plurality of input variables with a reduced number of input variables such that the subset of input variables of the plurality of input variables most effectively approximates the at least one output variable using the selected at least one simulator, and wherein the at least one selected simulator is further configured to generate the at least one algorithmic model using at least one of the variables used by the reservoir model, associated history matching input variables, and optimization input variables depending on the selection of the at least one simulator.
2. The apparatus of claim 1 , wherein the flow simulator module is configured to generate the at least one algorithmic model and the plurality of output variables using at least one simulator selected from a proxy flow simulator for assisted history matching, wherein the proxy flow simulator for assisted history matching is further configured to generate trained algorithmic models trained using the history matching input variables, an algorithmic model generated by a full-physics flow simulator, the plurality of input variables and the plurality of output variables of the reservoir model, and test data generated by the full-physics flow simulator.
3. The apparatus of claim 1 , wherein the flow simulator module is configured to generate the at least one algorithmic model and the plurality of output variables using at least the proxy flow simulator for field development optimization, wherein the proxy flow simulator for field development optimization is further configured to generate trained algorithmic models using the optimization input variables, an algorithmic model generated by the full-physics flow simulator, the input variables and output variables of the reservoir model, and test data generated by the full-physics flow simulator.
4. The apparatus of claim 2 or 3 , wherein the trained algorithmic models are generated using design of experiments.
5. The apparatus of claim 1 , further comprising an assisted history matching module configured to generate trained algorithmic models trained using the history matching input variables.
6. The apparatus of claim 5 , wherein the reservoir model is updated at least once by the model order reduction module or the assisted history matching module.
7. The apparatus of claim 5 , further comprising an optimization module, wherein the reservoir model is optimized at least once by an objective function definition module and a field development optimization module and generates an optimized reservoir model to identify best field development scenario and optimization input variables to construct a proxy flow model for the optimization module.
8. The apparatus of claim 1 , wherein the formation and reservoir properties database and includes rock and fluid properties and the field database includes field production data.
9. A system for generating forecasts from an input high-dimensional parameter data space, the apparatus comprising:
a reservoir model that includes a plurality of input variables, a plurality of output variables, and at least one algorithmic model, wherein the plurality of input variables and the plurality of output variables are generated by a flow simulator module based on variables from a formation and reservoir properties database and a field database, wherein the flow simulator module is configured to generate the at least one algorithmic model and the plurality of output variables using at least one simulator selected from a full-physics flow simulator, proxy flow simulator for assisted history matching, and a proxy flow simulator for field development optimization;
a model order reduction module configured to generate a subset of the plurality of input variables having a reduced parameter space relative to the input high-dimensional parameter data space by identifying input variables within the plurality of input variables useful that can be used to approximate at least one output variable of the plurality of output variables, wherein the model order reduction module uses a function decomposition algorithm to identify the input variables within the plurality of input variables useful to approximate the at least one output variable so as to generate the subset of the plurality of input variables with a reduced number of input variables such that the subset of input variables of the plurality of input variables most effectively approximates the at least one output variable using the selected at least one simulator, and
wherein the at least one simulator is further configured to generate the at least one algorithmic model using at least one of the variables used by the reservoir model, associated history matching input variables, and optimization input variables depending on the selection of the at least one simulator.
10. The system of claim 9 , wherein the flow simulator module is configured to generate the at least one algorithmic model and the plurality of output variables using at least one simulator selected from a proxy flow simulator for assisted history matching, wherein the proxy flow simulator for assisted history matching is further configured to generate trained algorithmic models trained using the history matching input variables, an algorithmic model generated by a full-physics flow simulator, the plurality of input variables and the plurality of output variables of the reservoir model, and test data generated by the full-physics flow simulator.
11. The system of claim 9 , wherein the flow simulator module is configured to generate the at least one algorithmic model and the plurality of output variables using at least the proxy flow simulator for field development optimization, wherein the proxy flow simulator for field development optimization is further configured to generate trained algorithmic models using the optimization input variables, an algorithmic model generated by the full-physics flow simulator, the input variables and output variables of the reservoir model, and test data generated by the full-physics flow simulator.
12. The system of claim 10 or 11 , wherein the trained algorithmic models are generated using design of experiments.
13. The system of claim 9 , further comprising an assisted history matching module, wherein the model order reduction module and the assisted history matching module calibrate the reservoir model using variables from the field database, uncertainty input variables, sensitivity analysis, and an optimization algorithm,
wherein a field development optimization module and an objective function definition module optimize the reservoir model to define best case scenarios for well site operations.
14. The system of claim 13 , wherein the reservoir model is updated at least once by the model order reduction module or the assisted history matching module.
15. The system of claim 13 , further comprising an optimization module, wherein the reservoir model is optimized at least once by an objective function definition module and a field development optimization module and generates an optimized reservoir model to identify best field development scenario and optimization input variables to construct a proxy flow model for the optimization module.
16. The system of claim 10 , wherein the formation and reservoir properties database includes rock and fluid properties and the field database includes field production data.
17. A method for generating forecasts from an input high-dimensional parameter data space, the method comprising:
selecting at least one simulator from a full-physics flow simulator, proxy flow simulator for assisted history matching, and a proxy flow simulator for field development optimization;
generating a reservoir model that includes a plurality of output variables and at least one algorithmic model using the at least one selected simulator, and a plurality of input variables from a reservoir properties database and a field database;
generating at least one updated reservoir model using at least one the variables used by the reservoir model, history matching input variables, and optimization input variables; and
identifying input variables within the plurality of input variables for use in approximating at least one output variable by-applying a function decomposition algorithm so as to generate a subset of the plurality of input variables with a reduced number of input variables such that the subset of input variables of the plurality of input variables most effectively approximates the at least one output variable using the selected at least one simulator,
wherein the at least one simulator is further configured to generate the at least one algorithmic model using at least one of the variables used by the reservoir model, associated history matching input variables, and optimization input variables depending on the selection of the at least one simulator.
18. The method of claim 17 , further comprising using a proxy flow simulator for assisted history matching to generate trained algorithmic models trained using the history matching input variables, an algorithmic model generated by a full-physics flow simulator, the input variables and output variables of the reservoir model, and test data generated by the full-physics flow simulator.