Method for dynamic calibration and simultaneous closed-loop inversion of simulation models of fractured reservoirs

1. A computer-implemented method for generating a fractured reservoir model executed by one or more processors, the method comprising:

receiving a seismic dataset of a surveyed subsurface, the seismic dataset comprising a set of parameters representing physical features of the surveyed subsurface;

identifying a dynamic response of each parameter of the set of parameters, wherein each parameter of the set is associated with a range of values;

selecting a subset of parameters from the set of parameters based on the dynamic response of each parameter;

sampling an outer boundary of a parameter uncertainty domain, the parameter uncertainty domain based on the range of values for each parameter of the subset of parameters;

adjusting the range of values associated with each parameter of the subset based on the sampling;

generating a geo-model based on the adjusted range of values associated with each parameter of the subset;

generating a discrete fracture network model based on the geo-model;

generating a scenario of a simulation model based on the discrete fracture network model and the geo-model;

performing a forward simulation based on the scenario of the simulation model;

determining that a misfit of the forward simulation is below a threshold by evaluating an objective function; and

producing a three-dimensional (3D) model based on the forward simulation;

in response to determining that the misfit of the forward simulation is not below the threshold:

redefining the objective function by an assisted history matching process in which a new parameter uncertainty domain is sampled for each of the geo-model, the discrete fracture network model, and the scenario of a simulation model; and

generating a history matched model based on the redefined objective function.

2. A computer-implemented method of claim 1 , wherein identifying a dynamic response of one or more parameters of the set of parameters comprises:

assigning one or more of a tolerance value range and a probability sampling distribution to each parameter of the set of parameters based on the uncertainty domain associated with each parameter;

determining a sensitivity metric for each of the parameters based on the one or more of the tolerance value range and the probability sampling distribution of each of the parameters;

ranking the parameters of the set of parameters based on the sensitivity metric associated with each of the parameters; and

selecting, from the set of the parameters, the subset of parameters including one or more higher ranked parameters, the one or more higher ranked parameters representing higher contribution to the dynamic response of the fractured reservoir model than one or more lower ranked parameters of the set of parameters.

3. A computer-implemented method of claim 2 , wherein assigning the one or more of the tolerance value range and the probability sampling distribution is based on selected types of basin geology and sedimentology associated with the seismic dataset.

4. A computer-implemented method of claim 1 , wherein sampling the outer boundary of the parameter uncertainty domain comprises a two level design of experiments (DoE) process.

5. A computer-implemented method of claim 4 , further comprising refining the sampling based on a three level design of experiments (DoE) process.

6. A computer-implemented method of claim 1 , wherein the set of parameters is selected from a group consisting of permeability, porosity, fracture density, fracture intensity, fracture geometry, fracture aperture, fracture orientation, stress, pore pressure, fluid contacts, fault transmissibility, a vertical to horizontal permeability (Kv/Kh) ratio, fluid properties, and aquifer strength.

7. A computer-implemented method of claim 1 , wherein generating the scenario of a simulation model based on the discrete fracture network model and the geo-model comprises:

defining uncertainty tolerance range of fluid contacts in the simulation model;

defining uncertainty tolerance range of fault transmissibility in the simulation model;

defining uncertainty tolerance range of a vertical to horizontal permeability (Kv/Kh) ratio in the simulation model;

defining uncertainty tolerance range of fluid properties (PVT) in the simulation model;

defining uncertainty tolerance range of aquifer strength in the simulation model; and

defining uncertainty tolerance range of external fluxes in the simulation model.

8. A computer-implemented method of claim 1 , further comprising generating a plurality of history matched models based on an iterative process of redefining the objective function; and

ranking the plurality of history matched models based on misfits determined from the objective functions associated with each of the history matched models.

9. A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:

receiving a seismic dataset of a surveyed subsurface, the seismic dataset comprising a set of parameters representing physical features of the surveyed subsurface;

identifying a dynamic response of each parameter of the set of parameters, wherein each parameter of the set is associated with a range of values;

selecting a subset of parameters from the set of parameters based on the dynamic response of each parameter;

sampling an outer boundary of a parameter uncertainty domain, the parameter uncertainty domain based on the range of values for each parameter of the subset of parameters;

adjusting the range of values associated with each parameter of the subset based on the sampling;

generating a geo-model based on the adjusted range of values associated with each parameter of the subset;

generating a discrete fracture network model based on the geo-model;

generating a scenario of a simulation model based on the discrete fracture network model and the geo-model;

performing a forward simulation based on the scenario of the simulation model;

determining that a misfit of the forward simulation is below a threshold by evaluating an objective function;

producing a three-dimensional (3D) model based on the forward simulation;

in response to determining that the misfit of the forward simulation is not below the threshold:

redefining the objective function by an assisted history matching process in which a new parameter uncertainty domain is sampled for each of the geo-model, the discrete fracture network model, and the scenario of a simulation model; and

generating a history matched model based on the redefined objective function.

10. The non-transitory, computer-readable medium of claim 9 , wherein identifying a dynamic response of one or more parameters of the set of parameters comprises:

assigning one or more of a tolerance value range and a probability sampling distribution to each parameter of the set of parameters based on the uncertainty domain associated with each parameter;

determining a sensitivity metric for each of the parameters based on the one or more of the tolerance value range and the probability sampling distribution of each of the parameters;

ranking the parameters of the set of parameters based on the sensitivity metric associated with each of the parameters; and

selecting, from the set of the parameters, the subset of parameters including one or more higher ranked parameters, the one or more higher ranked parameters representing higher contribution to the dynamic response of the fractured reservoir model than one or more lower ranked parameters of the set of parameters.

11. The non-transitory, computer-readable medium of claim 10 , wherein assigning the one or more of the tolerance value range and the probability sampling distribution is based on selected types of basin geology and sedimentology associated with the seismic dataset.

12. The non-transitory, computer-readable medium of claim 9 , wherein the set of parameters is selected from a group consisting of permeability, porosity, fracture density, fracture intensity, fracture geometry, fracture aperture, fracture orientation, stress, pore pressure, fluid contacts, fault transmissibility, a vertical to horizontal permeability Kv/Kh) ratio, fluid properties, and aquifer strength.

13. The non-transitory, computer-readable medium of claim 9 , wherein generating the scenario of a simulation model based on the discrete fracture network model and the geo-model comprises:

defining uncertainty tolerance range of fluid contacts in the simulation model;

defining uncertainty tolerance range of fault transmissibility in the simulation model;

defining uncertainty tolerance range of a vertical to horizontal permeability (Kv/Kh) ratio in the simulation model;

defining uncertainty tolerance range of fluid properties (PVT) in the simulation model;

defining uncertainty tolerance range of aquifer strength in the simulation model; and

defining uncertainty tolerance range of external fluxes in the simulation model.

14. The non-transitory, computer-readable medium of claim 9 , further comprising generating a plurality of history matched models based on an iterative process of redefining the objective function; and

ranking the plurality of history matched models based on misfits determined from the objective functions associated with each of the history matched models.

15. A computer-implemented system, comprising:

a computer memory; and

a hardware processor interoperably coupled with the computer memory and configured to perform operations comprising:

receiving a seismic dataset of a surveyed subsurface, the seismic dataset comprising a set of parameters representing physical features of the surveyed subsurface;

identifying a dynamic response of each parameter of the set of parameters, wherein each parameter of the set is associated with a range of values;

selecting a subset of parameters from the set of parameters based on the dynamic response of each parameter;

sampling an outer boundary of a parameter uncertainty domain, the parameter uncertainty domain based on the range of values for each parameter of the subset of parameters;

adjusting the range of values associated with each parameter of the subset based on the sampling;

generating a geo-model based on the adjusted range of values associated with each parameter of the subset;

generating a discrete fracture network model based on the geo-model;

generating a scenario of a simulation model based on the discrete fracture network model and the geo-model;

performing a forward simulation based on the scenario of the simulation model;

determining that a misfit of the forward simulation is below a threshold by evaluating an objective function;

producing a three-dimensional (3D) model based on the forward simulation;

in response to determining that the misfit of the forward simulation is not below the threshold:

redefining the objective function by an assisted history matching process in which a new parameter uncertainty domain is sampled for each of the geo-model, the discrete fracture network model, and the scenario of a simulation model; and

generating a history matched model based on the redefined objective function.

16. The computer-implemented system of claim 15 , wherein identifying a dynamic response of one or more parameters of the set of parameters comprises:

assigning one or more of a tolerance value range and a probability sampling distribution to each parameter of the set of parameters based on the uncertainty domain associated with each parameter;

determining a sensitivity metric for each of the parameters based on the one or more of the tolerance value range and the probability sampling distribution of each of the parameters;

ranking the parameters of the set of parameters based on the sensitivity metric associated with each of the parameters; and

selecting, from the set of the parameters, the subset of parameters including one or more higher ranked parameters, the one or more higher ranked parameters representing higher contribution to the dynamic response of the fractured reservoir model than one or more lower ranked parameters of the set of parameters.

17. The computer-implemented system of claim 16 , wherein assigning the one or more of the tolerance value range and the probability sampling distribution is based on selected types of basin geology and sedimentology associated with the seismic dataset.

18. The computer-implemented system of claim 15 , wherein the set of parameters is selected from a group consisting of permeability, porosity, fracture density, fracture intensity, fracture geometry, fracture aperture, fracture orientation, stress, pore pressure, fluid contacts, fault transmissibility, a vertical to horizontal permeability (Kv/Kh) ratio, fluid properties, and aquifer strength.

19. The computer-implemented system of claim 15 , wherein generating the scenario of a simulation model based on the discrete fracture network model and the geo-model comprises:

defining uncertainty tolerance range of fluid contacts in the simulation model;

defining uncertainty tolerance range of fault transmissibility in the simulation model;

defining uncertainty tolerance range of a vertical to horizontal permeability (Kv/Kh) ratio in the simulation model;

defining uncertainty tolerance range of fluid properties (PVT) in the simulation model;

defining uncertainty tolerance range of aquifer strength in the simulation model; and

defining uncertainty tolerance range of external fluxes in the simulation model.

20. The computer-implemented system of claim 15 , further comprising generating a plurality of history matched models based on an iterative process of redefining the objective function; and

ranking the plurality of history matched models based on misfits determined from the objective functions associated with each of the history matched models.