DETAILED ACTION
Claims 1, 3, 5, 9, 11, 13, 17, and 19 are presented for examination.
This Office Action is in response to submission of documents on October 14, 2025.
Rejection of claims 1-20 under 35 U.S.C. 112(b) for being indefinite are withdrawn.
Other rejections of claims 2, 4, 6-8, 10, 12, 14-16, 8, and 20 are withdrawn in light of the cancellation of those claims in the Response.
Rejection of claims 1, 3, 5, 9, 11, 13, 17, and 19 under 35 U.S.C. 101 for being directed to unpatentable subject matter are maintained.
Rejection of claims 1-20 under 35 U.S.C. 102(a)(1) for being anticipated by Santoso are withdrawn.
Rejection of claims 1, 5-6, 8- 9, 13-14, and 16-17 under 35 U.S.C. 103 for being obvious over Maucec in view of Norbakht, and Chen are withdrawn.
Rejection of claims 2, 10, and 18 under 35 U.S.C. 103 for being obvious over Maucec, Norbakht, and Chen, and further in view of Bazargan are withdrawn.
Rejection of claims 3, 4, 11, 12, 19, and 20 are rejected for being obvious over Maucec in view of Norbakht, Chen, and Bargazan, and further in view of Agada are withdrawn.
Rejection of claims 7 and 15 under 35 U.S.C. 103 for being obvious over Maucec, Norbakht, and Chen in view of Agana are withdrawn.
New rejection of claims 1, 5, 9, 13, and 17 under 35 U.S.C. 103 as being obvious over Madasu in view of Chen, Bazargan, and Pivovar.
New rejection of claims 3, 11, and 19 under 35 U.S.C. 103 as being obvious over Madasu in view of Chen, Bazargan, Pivovar, and Agada.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
Regarding the rejection of the claims under 35 U.S.C. 112(b), the submitted amendments either clarify the language or have been cancelled. Accordingly, the rejections are withdrawn.
Regarding the rejection of the claims under 35 U.S.C. 101, Examiner is not persuaded by the arguments. The claims recite:
obtaining a plurality of measured production data from an oil reservoir;
selecting, by a computer processor, a reservoir simulation model of the oil reservoir…
performing, by the computer processor utilizing the reservoir simulation model, a first plurality of production predictions associated with the oil reservoir…
performing, by the computer processor, a Bayesian optimization of the reservoir simulation model using the first plurality of production predictions and the plurality of measured production data to produce an updated reservoir simulation model, wherein the Bayesian optimization automatically tunes a plurality of hyperparameters in the reservoir simulation model…
performing, by the computer processor, a history matching of the updated reservoir simulation model using the plurality of measured production data and a second plurality of production predictions that is generated by the updated reservoir simulation model.
These are the critical steps that are required in order to perform the history matching (i.e., any additional steps of the method do not contribute to the data that is utilized to generate the updated reservoir simulation model and/or perform the history matching). For example, as claimed, the “second model” is not utilized in generating any data that is used for the history matching because it is utilized to generate “posteriors,” which are not present in the claim elsewhere. Thus, the “second model,” its generation, and/or what is represents is not tied to the rest of the claim and therefore, when taken as an individual step, is not part of the process to perform history matching.
Regarding the hyperparameters, the claim recites what the hyperparameters are “used for,” but does not actually utilize the them to perform the recited actions (i.e., “control a covariance and a plurality of Markov Chain Monte Carlo (MCMC) properties in the reservoir simulation model”). Thus, the intended use of the hyperparameters does not contribute to the process and therefore their use is not recited as contributing to an improvement over existing technology. Further, because the claim does not indicate that the reservoir simulation model is comprised of an MCMC, this intended use does not affect the reservoir simulation.
As indicated in the Background of the Specification, “The other [method to perform history matching] is the modern approach, which uses computerized algorithm to directly tune parameters…The modern approach is highly affected by prior selection, likelihood calculation, and posterior samplings.” Published application at [0001]. It is an improvement in this “modern approach” that appears to be the technology that is improved by the claimed invention.
However, the critical steps of the claim are not directed to an improvement in selection of the priors, sampling posteriors, and/or a specific likelihood calculation. While “updated priors” are utilized to generate the “second model,” as indicated above, the “second model” is not claimed as contributing to performing the history matching.
While Examiner does not disagree that the disclosed invention, when analyzed as a whole, improves over the “modern approach” by using updated priors that are generated by a PCE model to construct a “second model,” the claims do not reflect the improvement. Accordingly, Examiner suggests amendments to the claims that reflect using the “second model” in the process, which would recite the asserted improvement. For example, referring to FIG. 7A, the output of the fine PCE mode (i.e., the “second model”) is utilized in the loop that includes the ”Bayesian Optimization” to tune the hyperparameters, which are provided to the “Adaptive Metropolis Algorithm” which, as it appears in [0055], is used in the Bayesian MCMC.
Accordingly, although the Examiner does is not persuaded that the claims do not recite an improvement in the field, the Specification clearly asserts such an improvement. If amended to better recite the improvement (i.e., using updated priors to tune the model), the claim would overcome the rejection under 35 U.S.C. 101 by either not reciting an abstract idea or including additional elements that recite such an improvement at Step 2A, Prong 2 of the analysis.
Regarding Step 2A, Prong 2, Examiner is not persuaded that the recited additional elements (i.e., “obtaining” data) integrate the judicial exceptions into a practical application and/or recite improvements in the field. Further, the additional elements do not recite “significantly more” than the recited abstract ideas because gathering data is an extra-solution activity, as determined by courts. See, e.g., In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989); In re Meyers, 688 F.2d 789, 794; 215 USPQ 193, 196-97 (CCPA 1982); OIP Technologies, 788 F.3d at 1363, 115 USPQ2d at 1092-93; CyberSource v. Retail Decisions, Inc., 654 F.3d 1366, 1375, 99 USPQ2d 1690, 1694 (Fed. Cir. 2011).
Thus, the rejection of the currently pending claims under 35 U.S.C. 101 is maintained.
Regarding the rejection of the claims under 35 U.S.C. 102(a)(1), Examiner acknowledges the change in inventorship which invalidates the cited prior art reference because the inventors and the authors of the reference are the same. Thus, under 35 U.S.C. 102(b)(1), the reference is not applicable prior art. Accordingly, the rejection under 35 U.S.C. 102(a)(1) is withdrawn.
Regarding the rejection of the currently pending claims under 35 U.S.C. 103, Examiner agrees that the references do not disclose all of the limitations of the claims, as amended. Accordingly, the rejections have been withdrawn. However, in light of the amendments, new rejections of the pending claims are presented herein as being obvious over Madasu in view of Chen, Bazargan, Pivovar, Agada.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to judicial exceptions without significantly more. The claims recite mathematical calculations and mental processes. This judicial exception is not integrated into a practical application because the additional elements that are recited in the claims are extra-solution activities that do not integrate the judicial exceptions into a practical application. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because courts have found that the steps of data gathering and reciting ideas of a solution are not significantly more than a judicial exception.
Claim 1
Step 1: The claim is directed to a process, falling under one of the four statutory categories of invention.
Step 2A, Prong 1: The claim 1 limitations include (bolded for abstract idea identification):
Claim 1
Mapping Under Step 2A Prong 1
1. A method comprising:
obtaining a plurality of measured production data from an oil reservoir;
selecting, by a computer processor, a reservoir simulation model of the oil reservoir;
identifying, by the computer processor, a mathematical model based on the reservoir simulation model;
identifying, by the computer processor, a plurality of history matching parameters based on the reservoir simulation model as initial priors;
constructing, by the computer processor utilizing the initial priors, a first model, wherein the first model is a coarse low-fidelity Polynomial Chaos Expansion (PCE) model;
obtaining, by the computer processor and utilizing the first model, updated priors;
constructing, by the computer processor and utilizing the updated priors and a Latin Hypercube Sampler (LHS), a second model, wherein the second model is a fine low-fidelity PCE model;
obtaining, by the computer processor and utilizing the second model, posteriors;
performing, by the computer processor utilizing the reservoir simulation model, a first plurality of production predictions associated with the oil reservoir, wherein the second model represents the reservoir simulation model;
performing, by the computer processor, a Bayesian optimization of the reservoir simulation model using the first plurality of production predictions and the plurality of measured production data to produce an updated reservoir simulation model,
wherein the Bayesian optimization automatically tunes a plurality of hyperparameters in the reservoir simulation model, and wherein the plurality of hyperparameters are used to control a covariance and a plurality of Markov Chain Monte Carlo (MCMC) properties in the reservoir simulation model;
performing, by the computer processor, a history matching of the updated reservoir simulation model using the plurality of measured production data and a second plurality of production predictions that is generated by the updated reservoir simulation model.
Abstract Idea: Mental Process
Selecting a model is a mental process that can be performed by a human (i.e., selecting a component “of interest”). The selection is an action that requires observation, evaluation, and judgment that can be performed by a human with knowledge in simulation models. See e.g., MPEP 2106.04(a)(2), Subsection III.
Abstract Idea: Mental Process
Identifying a model is a mental process that is relevant is an action that can be performed by a human. Determining “relevancy” is a process that requires evaluation and judgment to identify a model that meets the criteria of being “relevant.” MPEP 2106.04(a)(2), Subsection III.
Abstract Idea: Mental Process
Identifying parameters is a mental process that requires observation, evaluation, judgment, and opinion, which can be performed by a human. MPEP 2106.04(a)(2), Subsection III.
Abstract Idea: Mathematical Calculations
Constructing a model is a mathematical concept that includes performing one or more mathematical calculations. The resulting model is a mathematical concept that is comprised of mathematical functions that product an output. See MPEP § 2106.04(a)(2), Subsection I.
Abstract Idea: Mathematical Calculations
Constructing a model is a mathematical concept that includes performing one or more mathematical calculations. The resulting model is a mathematical concept that is comprised of mathematical functions that product an output. See MPEP § 2106.04(a)(2), Subsection I.
Abstract Idea: Mental Process
Performing a prediction is a mental process that requires evaluation and judgment to predict one or more aspects of the oil reservoir, which can be performed in the human mind by observing numbers and performing (e.g., using a pencil and paper). MPEP 2106.04(a)(2), Subsection III.
Abstract Idea: Mathematical Calculations
Utilizing a model is a mathematical concept that includes causing the model to perform mathematical functions to generate a result. MPEP § 2106.04(a)(2), Subsection I.
Abstract Idea: Mathematical Calculations
Tuning a parameter is a mathematical concept that includes utilizing an MCMC workflow to change one or more of the parameters to a more fitting value for the models to perform.
MPEP § 2106.04(a)(2), Subsection I.
Abstract Idea: Mathematical concept
“History matching” requires comparing a resulting value with a historical value and adjusting one or more parameters based on the difference between the predicted value and the actual value. Comparison between values is a mathematical concept that includes utilizing one or more calculations to perform the history matching. See MPEP 2106.04(a)(2), Subsection I.
Step 2A, Prong 2: The claim 1 limitations recite (bolded for additional element identification):
Claim 1
Mapping Under Step 2A Prong 2
1. A method comprising:
obtaining a plurality of measured production data from an oil reservoir;
selecting, by a computer processor, a reservoir simulation model of the oil reservoir;
identifying, by the computer processor, a mathematical model based on the reservoir simulation model;
identifying, by the computer processor, a plurality of history matching parameters based on the reservoir simulation model as initial priors;
constructing, by the computer processor utilizing the initial priors, a first model, wherein the first model is a coarse low-fidelity Polynomial Chaos Expansion (PCE) model;
obtaining, by the computer processor and utilizing the first model, updated priors;
constructing, by the computer processor and utilizing the updated priors and a Latin Hypercube Sampler (LHS), a second model, wherein the second model is a fine low-fidelity PCE model;
obtaining, by the computer processor and utilizing the second model, posteriors;
performing, by the computer processor utilizing the reservoir simulation model, a first plurality of production predictions associated with the oil reservoir, wherein the second model represents the reservoir simulation model;
performing, by the computer processor, a Bayesian optimization of the reservoir simulation model using the first plurality of production predictions and the plurality of measured production data to produce an updated reservoir simulation model,
wherein the Bayesian optimization automatically tunes a plurality of hyperparameters in the reservoir simulation model, and wherein the plurality of hyperparameters are used to control a covariance and a plurality of Markov Chain Monte Carlo (MCMC) properties in the reservoir simulation model;
performing, by the computer processor, a history matching of the updated reservoir simulation model using the plurality of measured production data and a second plurality of production predictions that is generated by the updated reservoir simulation model.
The limitation is an additional element that is an extra-solution activity of data gathering and transmission, which courts have found do not integrate the judicial exception into a practical application. See MPEP 2106.05(g).
Reciting a generic computer component is merely an application of a judicial exception. Further, courts have found that an abstract idea, performed by a generic computer, does not integrate the judicial exception into a practical application, but instead amount to reciting the exception and further reciting “apply it.” See, e.g., MPEP 2106.04(a)(2), Subsection III(C), MPEP 2106.05(f).
Reciting a generic computer component is merely an application of a judicial exception. Further, courts have found that an abstract idea, performed by a generic computer, does not integrate the judicial exception into a practical application, but instead amount to reciting the exception and further reciting “apply it.” See, e.g., MPEP 2106.04(a)(2), Subsection III(C), MPEP 2106.05(f).
Reciting a generic computer component is merely an application of a judicial exception. Further, courts have found that an abstract idea, performed by a generic computer, does not integrate the judicial exception into a practical application, but instead amount to reciting the exception and further reciting “apply it.” See, e.g., MPEP 2106.04(a)(2), Subsection III(C), MPEP 2106.05(f).
Reciting a generic computer component is merely an application of a judicial exception. Further, courts have found that an abstract idea, performed by a generic computer, does not integrate the judicial exception into a practical application, but instead amount to reciting the exception and further reciting “apply it.” See, e.g., MPEP 2106.04(a)(2), Subsection III(C), MPEP 2106.05(f).
The limitation is an additional element that is an extra-solution activity of data gathering and transmission, which courts have found do not integrate the judicial exception into a practical application. See MPEP 2106.05(g).
Reciting a generic computer component is merely an application of a judicial exception. Further, courts have found that an abstract idea, performed by a generic computer, does not integrate the judicial exception into a practical application, but instead amount to reciting the exception and further reciting “apply it.” See, e.g., MPEP 2106.04(a)(2), Subsection III(C), MPEP 2106.05(f).
The limitation is an additional element that is an extra-solution activity of data gathering and transmission, which courts have found do not integrate the judicial exception into a practical application. See MPEP 2106.05(g).
Reciting a generic computer component is merely an application of a judicial exception. Further, courts have found that an abstract idea, performed by a generic computer, does not integrate the judicial exception into a practical application, but instead amount to reciting the exception and further reciting “apply it.” See, e.g., MPEP 2106.04(a)(2), Subsection III(C), MPEP 2106.05(f).
Reciting a generic computer component is merely an application of a judicial exception. Further, courts have found that an abstract idea, performed by a generic computer, does not integrate the judicial exception into a practical application, but instead amount to reciting the exception and further reciting “apply it.” See, e.g., MPEP 2106.04(a)(2), Subsection III(C), MPEP 2106.05(f).
Reciting a generic computer component is merely an application of a judicial exception. Further, courts have found that an abstract idea, performed by a generic computer, does not integrate the judicial exception into a practical application, but instead amount to reciting the exception and further reciting “apply it.” See, e.g., MPEP 2106.04(a)(2), Subsection III(C), MPEP 2106.05(f).
Step 2B: Regarding Step 2B, the inquiry is whether any of the additional elements (i.e., the elements that are not the judicial exception) amount to significantly more than the recited judicial exception. Obtaining data is the extra-solution activity of data gathering and transmission, which courts have found to be insignificantly more than the recited exception. See, e.g., In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989); In re Meyers, 688 F.2d 789, 794; 215 USPQ 193, 196-97 (CCPA 1982); OIP Technologies, 788 F.3d at 1363, 115 USPQ2d at 1092-93; CyberSource v. Retail Decisions, Inc., 654 F.3d 1366, 1375, 99 USPQ2d 1690, 1694 (Fed. Cir. 2011). Further, an idea of a solution and/or linking the judicial exception to a particular field or technology has been found by courts to be insignificantly more than the recited judicial exception. See McRO, 837 F.3d at 1314-15, 120 USPQ2d at 1102-03; DDR Holdings, 773 F.3d at 1259, 113 USPQ2d at 1107.
Accordingly, claim 1 is rejected for being directed to unpatentable subject matter.
Claim 3
Claim 3 recites wherein the first model is trained by a first training set, and the second model is trained by a second training set; and wherein, the first training set has a smaller size than the second training set. The claim recites mere data gathering to be utilized in training a model, which is a mathematical concept. Accordingly, claim 3 is rejected for being directed to unpatentable subject matter.
Claim 5
Claim 5 recites determining, by the computer processor, if a plurality of medians of the posteriors matches the plurality of measured production data (a mental process that includes evaluation and judgment), upon determining that the plurality of medians of the posteriors matches the plurality of measured production data, identifying, by the computer processor, the plurality of medians of the posteriors as plurality of final posteriors (mental process that requires observation); and upon determining that the plurality of medians of the posteriors does not match the plurality of measured production data, performing, by the computer processor, a plurality of inspections (additional element of an idea of a solution, which is not integrated into a practical application and does not amount to significantly more than the judicial exception). Accordingly, claim 5 is rejected for being directed to unpatentable subject matter.
Claim 9
Claim 9 recites “A system comprising: a computing device with a computer processor, the computing device executing a history matching manager configured to:” perform the steps recited in claim 1. Recitation of generic computer components is an additional element of an application of a judicial exception, which amounts to reciting “apply it.” Courts have found that such a limitation does not integrate the judicial exception into a practical application and further does not amount to significantly more than the abstract idea. For at least the same reasons as asserted regarding claim 1, claim 9 is directed to unpatentable subject matter.
Claims 11 and 13
Claim 11 and 13 recite substantially the same limitations as recited in claims 3 and 5. Accordingly, for at least the same reasoning as claims 3 and 5, claims 11 and 13 are directed to unpatentable subject matter.
Claim 17
Claim 17 recites “A non-transitory computer readable medium storing instructions executable by a computer processor, the instructions comprising functionality for:” performing the steps recited in claim 1. Recitation of generic computer components is an additional element of an application of a judicial exception, which amounts to reciting “apply it.” Courts have found that such a limitation does not integrate the judicial exception into a practical application and further does not amount to significantly more than the abstract idea. For at least the same reasons as asserted regarding claim 1, claim 17 is directed to unpatentable subject matter.
Claim 19
Claim 19 recites substantially the same limitations as recited in claim 3. Accordingly, for at least the same reasoning as claim 3, claim 19 is directed to unpatentable subject matter.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 5, 9, 13, and 17 are rejected under 35 U.S.C. 103 for being obvious over Madasu, et al., (U.S. Patent Pub. No. 2021/0270998, hereinafter “Madasu”) in view of Chen, et al. (U.S. Patent Pub. No. 2020/0217978, hereinafter “Chen”), Bazargan, et al., ("An Efficient Polynomial Chaos-based Proxy Model for History Matching and Uncertainty Quantification of Complex Geological Structures," hereinafter "Bazargan"), and Pivovar, et al., (U.S. Patent No. 12,001,766, hereinafter “Pivovar”).
Claim 1
Madasu discloses:
A method comprising: obtaining a plurality of measured production data from an oil reservoir;
At block 206, for each of multiple historical times, a measurement value from the oilfield may be obtained. The historical times may include times that span an entire history of a well system (e.g., the entire history of one or more production wells in the well system). For example, the measurement values may include surface flow rate measurements obtained by a flow sensor such as flow sensor 145 of FIG. 1 and/or surface pressure measurements obtained by a surface pressure sensor such as pressure sensor 147 of FIG. 1. Madasu at [0039].
selecting, by a computer processor, a reservoir simulation model of the oil reservoir;
In an implementation, the oilfield model may be based in part on known or measured geophysical/geologic and seismic properties of the oilfield and/or on well system data including various measurements collected downhole from one or more wells drilled within a reservoir in the oilfield (e.g., in the form of a production well for an oil and gas reservoir). Further, multiple production wells may be drilled for providing access to the reservoir fluids underground. Measured values such as surface flow rate values and/or surface pressures may be collected regularly from each production well, as will be described in further detail below with respect to a production well example as illustrated in FIG. 1. Madasu at [0020].
identifying, by the computer processor, a mathematical model based on the reservoir simulation model;
Petroleum reservoirs are typically geologically complex and large in size. In order to facilitate oil and gas recovery, oilfield models including reservoir features and/or well system features are generated. In an example, oilfield models may be developed and parameterized based on, for example, geophysical data and production data. Geophysical data, such as seismic and wireline logs, may provide ranges for model parameters that describe physical properties (e.g., porosity or permeability) of one or more portions of the reservoir. Production data (e.g., measured water saturation and pressure information such as downhole pressures) may provide ranges for model parameters that describe the fluid flow dynamics of the reservoir and/or well system components for the reservoir. Madasu at [0021].
identifying, by the computer processor, a plurality of history matching parameters based on the reservoir simulation model as initial priors;
At block 204, a prior probability distribution may be provided for the at least one adjustable parameter. The prior probability distribution for each adjustable parameter may be a simple range, a weighted range, or a collection of weighted ranges (as examples). The prior probability distributions for adjustable parameters such as NL, PB, PY, S, P, NF, H, L A, C, and/or PYP can be determined based on known geophysical features of the oilfield, reservoir and/or well system components and/or measurements obtained during drilling and/or wireline measurement during the production stage of the wellbore. Madasu at [0038].
obtaining, by the computer processor
At block 214, the processor may apply a modification to the at least one adjustable parameter based on the prior probability distribution and the model error. As indicated by arrow 221, the processor may repeat the computing of block 208, the comparing of block 210, the determining of block 212, and the applying of block 214, until convergence of the model error (e.g., until the model error is below a threshold error and/or until the changes in the model error for each repetition fail to decrease by more than a convergence threshold), to generate a history-matched oilfield model that facilitates well system operations for the oilfield. Madasu at [0043].
performing, by the computer processor utilizing the reservoir simulation model, a first plurality of production predictions associated with the oil reservoir,
At block 208, for each of the multiple historical times, one or more processors may execute code or instructions stored in a non-transitory machine-readable medium to generate an output value of the model using the at least one adjustable parameter. Madasu at [0040].
performing, by the computer processor, a Bayesian optimization of the reservoir simulation model using the first plurality of production predictions and the plurality of measured production data to produce an updated reservoir simulation model, wherein the Bayesian optimization automatically tunes a plurality of hyperparameters in the reservoir simulation model, and
In this way, one or more of the operations described above in connection with blocks 208-216 can be performed to generate a history-matched oilfield model that facilitates well system operations for the oilfield, by performing a Bayesian optimization of at least one adjustable parameter using modifications to the at least one adjustable parameter based on a prior probability distribution, using measurement values and corresponding model prediction values, each generated using a corresponding modification of the at least one adjustable parameter. Madasu at [0045].
performing, by the computer processor, a history matching of the updated reservoir simulation model using the plurality of measured production data and a second plurality of production predictions that is generated by the updated reservoir simulation model.
As indicated by arrow 221, the processor may repeat the computing of block 208, the comparing of block 210, the determining of block 212, and the applying of block 214, until convergence of the model error (e.g., until the model error is below a threshold error and/or until the changes in the model error for each repetition fail to decrease by more than a convergence threshold), to generate a history-matched oilfield model that facilitates well system operations for the oilfield. Madasu at [0043].
Madasu does not appear to disclose:
constructing, by the computer processor utilizing the initial priors, a first model,
wherein the first model is a coarse low-fidelity Polynomial Chaos Expansion (PCE) model;
obtaining, by the computer processor and utilizing the first model, updated priors;
constructing, by the computer processor and utilizing the updated priors and a Latin Hypercube Sampler (LHS), a second model,
wherein the second model is a fine low-fidelity PCE model;
obtaining, by the computer processor and utilizing the second model, posteriors;
wherein the second model represents the reservoir simulation model;
wherein the plurality of hyperparameters are used to control a covariance and a plurality of Markov Chain Monte Carlo (MCMC) properties in the reservoir simulation model;
Chen, which is analogous art, discloses:
constructing, by the computer processor utilizing the initial priors, a first model,
A seismic AVA dataset A is input to the 2D or 3D stochastic inversion 10 which generates a Bayesian model using MCMC. This inversion results in an ensemble of estimated coarse-scale (e.g., spatial resolution of 20-100 m) seismic parameters B. Chen at [0038].
The MCMC is a “first model.”
wherein the first model is a coarse low-fidelity
obtaining, by the computer processor and utilizing the first model, updated priors;
As illustrated in FIG. 2, the “Parameters B” (analogous to “updated priors”) are generated by the “Bayesian model with MCMC.”
constructing, by the computer processor and utilizing the updated priors
Referring again to FIG. 2, the deep learning operation 12 generates an ensemble of fine-scale reservoir parameters C which may include seismic parameters. Chen at [0040].
The “second model” is the c-GAN model, which takes as input the “updated priors.”
wherein the second model is a fine low-fidelity
Chen is analogous art to the claimed invention because both are related to optimizing Markov Chain Monte Carlo methods for predicting variables related to oil reservoirs. It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the application, to combine Chen with Madasu to construct a system that updates priors before generating posteriors. Motivation to combine includes reducing computing resources required to execute a reservoir simulation by providing for an initial coarse model and subsequently utilizing the generated updated priors to generate posteriors.
Bazargan, which is analogous art to the claimed invention, discloses:
Polynomial Chaos Expansion (PCE) model
Polynomial chaos expansion, which was discussed in Chapter 2, constitutes a promising means to construct an efficient pseudo-simulator (proxy). Polynomial chaos expansion has a significant advantage over other response surfaces and proxy models as it guarantees the convergence in probability and also in distribution to the output random variable of interest, i.e. cumulative oil production. Bazargan at pg. 44, Paragraph 1.
Bazargan is analogous art to the claimed invention because both are directed to generating proxy models for history matching. It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the application, to combine the references with Bazargan to result in a system that utilizes polynomial chaos expansion models to update priors and generate posteriors. Motivation to combine includes reducing computing resources to generate simulation results because, when using PCE, fewer simulation runs are required.
Pivovar, which is analogous art to the claimed invention, discloses:
constructing, by the computer processor and utilizing the updated priors and a Latin Hypercube Sampler (LHS), a second model,
The method obtains N samples of values for the design parameters within the initial trust region. In some implementations, the method includes constructing the N samples using Latin Hypercube Sampling (LHS). Pivovar at col. 3, lines 53-56.
obtaining, by the computer processor and utilizing the second model, posteriors;
The method then trains an optimization model according to the N samples and the corresponding computed aggregate residuals. Various machine learning techniques can be used. In some implementations, the optimization model is a machine learning model. Pivovar at col. 4, lines 10-14.
wherein the second model represents the
The method is used to design nuclear reactor cores. The design process uses a plurality of design variables for a nuclear reactor. The design variables are essentially the independently controlled variables. In addition, the design process uses a plurality of metric variables related to the nuclear reactor. The metric variables include, for example, variables to measure thermal-hydraulic properties, variables to measure neutronics properties, and variables to measure stress properties. The metric variables are essentially the dependent variables, which measure the viability of a design. Other metric variables can also be used. Pivovar at col. 3, lines 34-42.
wherein the plurality of hyperparameters are used to control a covariance and a plurality of Markov Chain Monte Carlo (MCMC) properties in the reservoir simulation model;
In some implementations, the Markov Chain Monte Carlo algorithm uses the Metropolis-Hastings algorithm, which adapts a covariance matrix. Pivovar at col. 7, lines 4-6.
Pivovar is analogous art to the claimed invention because both are related to utilizing a model of a physical system to generate a model, using LHS, that can determine posteriors of the modeled system. It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to adapt the process described in Pivovar, which is directed to nuclear reactors, to an oil reservoir because, in both instances, the models are comprised of mathematical relationships that describe the physical phenomena being modeled. Motivation to combine includes reduction in calculation time to generate the model. As disclosed, “The disclosed methods provide the ability to understand (within the time frame of hours) interactions too complex for a human to explicitly process. The disclosed approach offers more than a single design that meets design criteria. Rather, the disclosed approach operates during the running of the computer analysis without interrupting the run and analyzes trade-offs between multiple designs that naturally arise with evolving design discovery and a spectrum of design inputs and changing mission profiles.” Pivovar at col. 3, lines 22-30.
Claim 5
Pivovar discloses:
further comprising: determining, by the computer processor, if a plurality of medians of the posteriors matches the plurality of measured production data,
The markov proposal distribution is a multivariate Gaussian distribution initially centered at either the current vest vector or one of the ranked vectors determined in Step 10. b. The posterior distribution used to sample from is the internal forest of the random forest. Hence, there is always a 50% probability at the median of the best design, but there may not be a >0% probability for surrounding designs. Pivovar at col. 12, lines 56-64.
upon determining that the plurality of medians of the posteriors matches the plurality of measured production data, identifying, by the computer processor, the plurality of medians of the posteriors as a plurality of final posteriors; and
Every time the MCMC adapts the covariance matrix of the proposal distribution in Step 11, a support vector machine (SVM) is trained on the accepted proposals of the markov chain. A different SVM is trained for each categorical variable.
c. Once the SVM are trained, each time the multinomial distribution is called, the probabilities of the SVM are used for the prior probabilities thus creating a hierarchical model that learns from the random walk. Pivovar at col. 13, lines 17-21.
upon determining that the plurality of medians of the posteriors does not match the plurality of measured production data, performing, by the computer processor, a plurality of inspections.
The new candidate designs identified in step 11 are then used as the next generation stochastic sample. Use these points as the next function call. Pivovar at col. 13, lines 5-7.
It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to combine Pivovar with the other references to result in a system that, once a second model is constructed, according to Pivovar, the posterior distributions are investigated further if they do not match the measured data that is identified as in Madasu. Motivation to combine includes improved accuracy and reduced computing times, as disclosed in Pivovar. See Pivovar at col. 3, lines 22-30.
Claim 9
Madasu discloses:
A system comprising: a computing device with a computer processor, the computing device is configured to perform a method comprising:
In one or more implementations, such a processing system may include a computing device (e.g., a server) and a data storage device (e.g., a database). Such a computing device may be implemented using any type of computing device having at least one processor, a memory and a networking interface capable of sending and receiving data to and from control unit 132 via a communication network, such as a processor 338 described in FIG. 3, the computing device 500 described hereinafter in connection with FIG. 5, and/or the server 606 described hereinafter in connection with FIG. 6. Madasu at [0031].
Perform the method recited in claim 1.
Because claim 1 is rejected under 35 U.S.C. 103, claim 9 is rejected for at least the same reasons using the same reference. Accordingly, claim 9 is rejected for being obvious over Madasu, Chen, Bazargan, and Pivovar.
Claim 13
Claims 13 recites substantially the same limitations as claim 5. Accordingly, claim 13 is rejected under 35 U.S.C. 103 for at least the same reasons.
Claim 17
Madasu discloses:
A non-transitory computer readable medium storing instructions executable by a computer processor, the instructions being configured to perform a method comprising:
At block 208, for each of the multiple historical times, one or more processors may execute code or instructions stored in a non-transitory machine-readable medium to generate an output value of the model using the at least one adjustable parameter Madasu at [0040]..
Performing the steps of the method recited in claim 1.
Because claim 1 is rejected under 35 U.S.C. 103, claim 17 is rejected for at least the same reasons using the same references. Accordingly, claim 9 is rejected for being obvious over Madasu, Chen, Bazargan, and Pivovar.
Claims 3, 11, and 19 are rejected under 35 U.S.C. 103 as being obvious over Madasu in view of Chen, Bazargan, Pivovar, and further in view of Agada, et al., (“Reduced Order Models for Rapid EOR Simulation in Fractured Carbonate Reservoirs,” hereinafter Agada).
Claim 3
Madasu, Chen, Bazargan, and Pivovar do not appear to disclose:
wherein the first model is trained by a first training set, and the second model is trained by a second training set; and wherein, the first training set has a smaller size than the second training set.
Agada, which is analogous art, discloses:
wherein the first model is trained by a first training set, and the second model is trained by a second training set; and wherein, the first training set has a smaller size than the second training set.
One way of reducing the computational cost is by using data-driven model reduction techniques that construct an approximation (or proxy) of the simulation response based on a limited number of simulation runs….The modelling process typically involves generating an initial reduced order model (ROM) with training simulations. Subsequently, an approximate solution to the objective function is obtained by evaluating the ROM. For validation purposes, approximate solutions from the ROM are compared to model predictions using high-fidelity simulation (e.g. Black-Oil or compositional simulation). If the comparison shows a mismatch, the ROM is iteratively updated with more training runs and testing points added until the mismatch is eliminated (Koziel and Yang, 2011). Agada at pg. 2, paragraph 1.
“More training runs” results in a model with higher fidelity, such as the second model.
Agada is analogous art to the claimed invention because both are related to simulations of oil reservoirs. It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the application, to combine the references with Agada to train models that use PCE. Motivation to combine includes training models with greater accuracy and using fewer computing resources compared to traditional usage of Monte Carlo methods.
Claims 11 and 19
Claims 11 and 19 recite substantially the same limitations as claim 3. Accordingly, for at least the same reasons and based on the same references, claims 11 and 19 are rejected under 35 U.S.C. 103 for being obvious over Madasu, Chen, Bazargan, Pivovar, and Agada.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Chen et al., U.S. Patent Pub. No. 2021/0110089.
Da Veiga, et al., U.S. Patent Pub. No. 2014/0019108.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Communication
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSEPH MORRIS whose telephone number is (703)756-5735. The examiner can normally be reached M-F 8:30-5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Ryan Pitaro can be reached at (571) 272-4071. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
JOSEPH MORRIS
Examiner
Art Unit 2188
/JOSEPH P MORRIS/Examiner, Art Unit 2188
/RYAN F PITARO/Supervisory Patent Examiner, Art Unit 2188
Prosecution Insights